On 2016 Aug 28, Steven Finney commented:

Schultz and van Vugt (2015) (henceforth SVV) compare their Tap Arduino device against a PC-based system using the FTAP software package (Finney SA, 2001a), and claim that FTAP has high latency and variability (M = 14.6 ms, SD = 2.8). This claim is incorrect: as SVV show later in their paper, the single specific percussion pad that they used in their FTAP condition is responsible for the bulk of their measured latency, not FTAP. SVV also contains a number of additional false and misleading claims.

FTAP is a software package that runs on a Linux PC; it must be attached to a MIDI input device (a keyboard or percussion pad) for input, and to a MIDI output device (a tone generator) for output. The end-to-end latency of an experimental system using FTAP (as measured by SVV) will be the (linear) sum of the input device latency, the output device latency, and the latency of the FTAP-MIDI system itself. As reported in Finney SA, 2001a and Finney SA, 2001b, an FTAP-MIDI system running on a standard 200 MHz Linux PC processed MIDI data with millisecond accuracy and precision, and this is easily replicated on current hardware (SVV themselves report such a result on p 5 of their article, and further data is provided on the FTAP web page at http://www.sfinney.com/ftap). Since FTAP adds no latency that is relevant at the millisecond level, the "FTAP" latency reported by SVV must be due to their input and/or output device.

SVV thus err in confounding the MIDI I/O devices with the FTAP software. To compound the error, in their FTAP condition SVV used a single input device (a Roland HPD-15 percussion pad) that they knew to be defective (they report that it both missed responses and generated superfluous responses). They measure the audio latency of the percussion pad to be 9 milliseconds, and then repeatedly present the "FTAP" and "percussion pad" measurements as if they were independent, when the percussion pad latency is actually the major component of their "FTAP" latency. (This lack of independence in their conditions arguably invalidates their statistical measures). Their latency data thus demonstrate that their specific percussion pad was faulty but say nothing about FTAP.

SVV also repeatedly and mistakenly assert that USB and MIDI processing will add significant latency in a PC system (e.g., p 2, "delays can be introduced by the [USB] polling speed"; p 7, "temporal noise...probably due to the MIDI-USB and USB-MIDI conversions"). This claim is also disproved by their own data. The FTAP distribution provides a loop test that rigorously tests MIDI and USB input and output (both software and hardware), along with FTAP itself. SVV report (p 5) that their FTAP configuration had a loop test result of 1.01 ms; this shows that neither USB nor MIDI processing added latency that was significant at the millisecond level. (See Finney SA, 2001a, Finney SA, 2001b, and the FTAP web page for further discussion of the configuration and interpretation of the loop test)

Finally, SVV suggest that Tap Arduino is an inexpensive replacement for FTAP; this is also incorrect. Tap Arduino is simply a button box that can produce auditory feedback; it cannot run any experiments without being attached to a PC. SVV provide no data that such a complete Tap Arduino system has millisecond accuracy. In fact, they explicitly note that using Tap Arduino in a synchronization experiment has issues with the "asynchrony between the Arduino response and the onset of computer-generated audio" (p 11) that may require "expensive options" to solve.

SVV are to be commended for highlighting the importance of end-to-end measurement of systems such as FTAP and MAX, and for making a valiant, if flawed, first attempt at such measurements. However, the only useful thing that their article demonstrates is that the Roland HPD-15 percussion pad (with one particular set of configuration parameters) is not a suitable input device for millisecond data collection.

A validated Arduino-based MIDI input device that could be used with FTAP or MAX would be a valuable contribution; unfortunately, Tap Arduino is not that device.

On 2017 May 10, C Baum commented:

NB to readers: The article itself was not retracted; the journal merely removed its electronic reproduction that had been created in error. See publisher's message below.

This article has been removed: please see Elsevier Policy on Article Withdrawal (http://www.elsevier.com/locate/withdrawalpolicy). Please note that a special edition of Canadian Journal of Diabetes (Volume 39 Supplement 4) was published electronically in error and has since been removed. This edition was planned as a print-only reproduction of a selection of earlier published articles from Canadian Journal of Diabetes: electronic publication of the edition created duplicate items, and were therefore removed. The original articles remain and are unaffected. Andrew Miller, Executive Publisher.

On 2017 Feb 19, David Keller commented:

Higher physician spending on medical tests is associated with lower risk of malpractice lawsuit

The observed association of higher physician spending on medical procedures and tests with lower risk of malpractice lawsuits does not seem surprising. In general, the need for a diagnostic test is assessed based on the pretest likelihood of disease, which cannot be determined with perfect accuracy. Playing the percentages in this way works well most of the time, except when it doesn't. All it takes is one missed diagnosis to ruin your whole decade. Even a completely useless test will force the physician to think about the patient, if only briefly, when the result is reported. That additional attention may be enough to trigger thoughts that lead to the correct diagnosis. In addition, there is the possibility that "shotgun" testing may return an informative result due to serendipity. Many effective pharmaceuticals are discovered by indiscriminate screening, and while this method cannot be justified on a cost-benefit-harm basis in clinical medicine, it is sometimes helpful when the approved diagnostic algorithm is stuck in the mud. You can't win the lottery if you don't buy a ticket.

The above comment was published online by British Medical Journal as a "rapid response"; it is included here for the record, for convenience, and for the opportunity to engage readers in the U.S.

Reference:

BMJ 2015;351:h5516 URL: http://www.bmj.com/content/351/bmj.h5516/rr-5

On 2016 Jan 25, Michael Peabody commented:

We have received questions from a few researchers about our paper. Below are some answers in case they are of broader interest.

Q: Did you evaluate how well methods performed without clade exclusion (i.e. no species were removed from the database being compared to)?

A: Yes we did; the results are shown in Figure S1 in Additional File 2.

Q: Why are there so few species in your test datasets for the evaluations?

A: A smaller number of species in the test sets makes the system easier to comprehend, which enables us to better understand the performance of the different methods. The MetaSimHC dataset was used since it was previously published and proposed as an evaluation dataset, plus it contains diverse taxa. For the in vitro dataset, making a mock community containing many species is non-trivial. Note we chose species relevant to a larger study we were performing and encourage researchers to customize any test dataset to suit their analysis needs. We also purposefully included some more closely related organisms, to evaluate how well methods differentiated more closely related taxa.

Q: If I'm just starting out, which method should I try?

A: MEGAN5 has some great features for performing all sorts of analyses and visualizations - good for exploring your data and potentially a good first method for starting out. We did find in this analysis (and subsequent analyses) that MEGAN4 and MEGAN5 with default settings tend to overpredict reads (e.g. assign a read to a similar species in the same genus if the actual species was removed from the database), however they otherwise perform quite well with BLASTX-like programs (BLASTX, RAPSearch2, Diamond). If the goal is trying to analyze community composition rather than classify all of the reads, a marker based method (which we would expect to have high precision and run relatively quick) such as MetaPhyler or MetaPhlAn should work well. MetaPhlAn2 was recently released and should be worth checking out (http://www.nature.com/nmeth/journal/v12/n10/full/nmeth.3589.html). Kraken is an incredibly quick method to run and should identify species if they are in the database used. CARMA3 and DiScRIBinATE were methods we found to be more conservative/less prone to overpredict reads relative to other methods with similar sensitivity, so are worth checking out if you are concerned about reads being overpredicted (and thus predicting species that aren't actually in the sample). See the paper discussion for more information.

Q: Why do certain methods give false species predictions?

A: This is method dependent. For example, MEGAN4 and MG-RAST rely on relatively simple LCA approaches. If a read makes a single hit to one species and it reaches the bit-score threshold, the read will be assigned directly to the species level, regardless of how good the hit/alignment is in terms of other metrics such as % identity (although depending on parameter choices such as the minimum support parameter the read may be reassigned by MEGAN4).

Q: Why is MG-RAST not in the clade exclusion analysis? I see it only in the analysis without clade exclusion.

A: Many methods, including MG-RAST, could not be evaluated with clade exclusion because we didn't have access to manipulate their underlying database. Others were not evaluated simply due to time constraints, as there are a very large number of methods to perform metagenomics taxonomic sequence classification (and more keep coming out). Hopefully more methods will be evaluated in this way and we have evaluated more methods already since this publication (see below).

We also have a few additional comments:

Strain level classification can be quite difficult, and complicated by notable within-strain variability that can occur, so we only looked at classifications down to the species level.

Pseudomonas fluorescens Pf-5 is now known as Pseudomonas protegens Pf-5.

Due to the way we evaluated sensitivity and precision, more conservative methods which tend to assign reads to higher taxonomic levels if the species is not in the database showed higher levels of sensitivity and precision. However, if you are only interested in more specific taxonomic ranks like species and genus level classifications, these methods may not be as useful if they end up classifying few reads to these taxonomic levels. Less conservative methods make the trade-off of assigning reads to more specific taxonomic levels, with higher rates of overprediction.

We have evaluated MEGAN5 on BLASTX, RAPSearch2, RAPSearch2 fast mode, and Diamond for the MetaSimHC dataset with 250bp simulated reads and overpredictions considered incorrect. As we move from BLASTX to Diamond in this list of heuristic search methods, we generally see a tradeoff of slightly decreased sensitivity, for a substantially improved (i.e. shortened) running time - see http://www.slideshare.net/Mpeabody/comparison-of-megan5-with-different-similarity-search-methods. However, precision stays about the same. We have compared MEGAN4 vs MEGAN5 using default parameters for each, and find MEGAN5 has slightly increased sensitivity and similar precision relative to MEGAN4.

Hope this is helpful.

-Michael Peabody

On 2015 Nov 10, Peter Karp commented:

Although I like the notion that access statistics could be used to direct curation efforts for a given database, note there is a converse view of this problem: if I'm going to write programs that compute across an entire database, I want the entire database to be as accurate as possible, otherwise my large-scale compute will not return accurate results. For example, if I'm going to build a metabolic model from a database, I want the whole metabolic network within the database to be well curated, regardless of how often researchers happen to query a particular enzyme. If I'm going to search the whole human genome for a gene that has certain properties, I want the whole genome to be annotated/curated as accurately as possible, regardless of how often each gene is visited by the research community. Thus, taking shortcuts in curation could backfire when we want to query across an entire biological system.

On 2016 Jan 12, Yuri Lazebnik commented:

A role for parasites and cell fusion in the emergence of transmissible cancers? Yuri Lazebnik & George Parris. The finding of cancerous tapeworm cells in the human body calls for a comprehensive mechanistic explanation to understand the significance of this discovery. One clue may be the prominent feature of the cells – aberrant cell fusion. Normally, cell fusion shapes some of the tapeworm organs. The observed promiscuous fusion, however, would be expected to generate abnormal cells Duelli D, 2003 and promote their clonal evolution by causing chromosomal instability Duelli DM, 2007, Duelli D, 2007, Lazebnik Y, 2014, by changing gene expression or causing dedifferentiation Bulla GA, 2010, Koulakov AA, 2012, and by serving as an equivalent of sexual reproduction, a powerful force in the evolution of species Parris G, 2006. Fusion with human cells, followed by unilateral genome reduction common to interspecies hybrids, or acquiring human DNA through other forms of gene transfer (both possibilities can be tested by searching for the stretches of DNA containing both human and worm sequences) would “humanize” the cells, thus helping them bypass host defenses. Such a breeding ground could produce evolving and transmissible cancer cells that are a species of its own. If so, the study by Muehlenbachs et al. may be a glimpse into how transmittable cancers, including those in humans Lazebnik Y, 2015, can originate.

On 2016 Apr 24, Clive Bates commented:

This article has received an extensive critique: Smears or science? The BMJ attack on Public Health England and its e-cigarettes evidence review November 2015. [Disclosure: I am the author].

The investigative article was published in the BMJ following PHE's publication of its E-cigarette evidence review update. The highly publicised finding of PHE's expert reviewers was that e-cigarettes are likely to be at least 95% lower risk than smoking - including an allowance for uncertainty.

The BMJ's article attempted to undermine that assessment and link it, erroneously, to the tobacco industry. The critique linked above makes ten distinct criticisms of the BMJ's investigative journalism and its unwarranted hostility to PHE's work.

1. Playing the man: the descent into personal attacks at the expense of substance
2. Exploiting the ambiguity of graphics: creating misleading connections between people
3. Failure to examine the underlying science: is the PHE 95% relative risk estimate actually reasonable?
4. Failure to acknowledge the problem PHE is tackling: widespread misperception of e-cigarette risks compared to smoking
5. Inappropriate dismissal of quantified estimates: these are useful to help people anchor risk perceptions
6. Hypocritical and abusive use of conflict of interest disclosure: it is for transparency, not disparagement
7. Bias and imbalance: selective quoting and inadequate scrutiny of PHE's critics
8. Unaccountable sources: reliance on anonymous hostile briefing by public officials
9. Activism rather than objectivity: are BMJ and Lancet becoming protagonists and losing their neutrality?
10. A new 'scream test': why has PHE's claim created such consternation?

On 2016 Aug 01, GT Hayman commented:

The abstract above is not the correct one for the title. It is actually the abstract for the first paper under 'Similar articles'.

On 2015 Dec 24, Andrea Messori commented:

Meta-analysis on thrombectomy in acute ischemic stroke: is the degree of heterogeneity influenced by the type of device?

In the overall analysis of 8 randomized controlled trials (RCTs), the meta-analytic results reported by Badhiwala et al (1) for the end-point of functional independence at 90 days contained a significant degree of heterogeneity (I² = 75.4%, p-value for heterogeneity <0.01 according to the results shown in Figure 1 published on page 1839 by these authors; data from 8 RCTs). Since the presence of heterogeneity is considered a negative factor in terms of reliability of the meta-analytic results (2-5), this finding represents an important limitation of the study by Badhiwala et al.; in their article, these authors have correctly pointed out this limitation. Likewise, the meta-analysis recently published by Elgendy et al. (6) on the same topic shared the same limitation and in fact contained a significant degree of heterogeneity (I² = 54.0%, p-value for heterogeneity= 0.021 according to the results shown in Figure 1 published on page 2501 by these authors; data from 9 RCT).

In their papers, Badhiwala et al. (1) and Elgendy et al.(6) have discussed which factors could be responsible for this significant heterogeneity. For example, the time window for the thrombectomy procedure, the type of device, and the use of functional perfusion imaging before randomization were thought to be implicated. In this context, the type of device is likely to play an important role as shown in previous studies (7).

For this reason, given that the Solitaire device was the one most commonly employed in this clinical material, we identified which trials among those examined Badhiwala et al. and by Elgendy et al. were based the comparison between a group receiving the Solitaire device (treatment group) and the control. group. The following four trials were found to be based on this comparison: ESCAPE, EXTEND-A, SWIFT-PRIME, and REVASCAT. In these four trials, the crude rates for the end-point of functional independence at 90 days were the following (comparison: Solitaire group vs control group): ESCAPE, 43 /147 vs. 87 /164; EXTEND-A, 14 /35 vs. 25 /35; SWIFT-PRIME, 33 /83 vs. 59 /98; REVASCAT, 29 /103 vs. 45 /103 (data shown in Figure 1 published on page 1839 by Badhiwala et al).

We have analyzed these data by application of the random-effect model (DerSimonian and Laird method as implemented in the Open Meta-analysis software) and we have obtained the following meta-analytic results: odds-ratio = 2.47 (95% confidence interval[CI], 1.84 to 3.33; I² = 0%, p-value for heterogeneity = 0.689); relative risk = 1.66 (95%CI, 1.40 to 1.97; I² = 0%, p-value for heterogeneity = 0.821). This re-analysis has some interest because, after the selection of a single device, the degree of heterogeneity was markedly reduced and changed from a statistically significant level in the overall analysis to 0%. In conclusion, our re-analysis indicates that the type of device can have an important role influencing the results of these two meta-analyses.

References

1) Badhiwala JH, Nassiri F, Alhazzani W, Selim MH, Farrokhyar F, Spears J, Kulkarni AV, Singh S, Alqahtani A, Rochwerg B, Alshahrani M, Murty NK, Alhazzani A, Yarascavitch B, Reddy K, Zaidat OO, Almenawer SA. Endovascular Thrombectomy for Acute Ischemic Stroke: A Meta-analysis. JAMA. 2015 Nov 3;314(17):1832-43.

2) Gagnier JJ, Morgenstern H, Altman DG, Berlin J, Chang S, McCulloch P, Sun X, Moher D; Ann Arbor Clinical Heterogeneity Consensus Group. Consensus-based recommendations for investigating clinical heterogeneity in systematic reviews. BMC Med Res Methodol. 2013 Aug 30;13:106.

3) Pigott T, Shepperd S. Identifying, documenting, and examining heterogeneity in systematic reviews of complex interventions. J Clin Epidemiol. 2013 Nov;66(11):1244-50.

4) Laurin D, Carmichael PH. Combining or not combining published results in the presence of heterogeneity? Am J Clin Nutr. 2010 Sep;92(3):669-70,

5) Bollen CW, Uiterwaal CS, van Vught AJ. Pooling of trials is not appropriate in the case of heterogeneity. Arch Dis Child Fetal Neonatal Ed. 2006 May;91(3):F233-4.

6) Elgendy IY, Kumbhani DJ, Mahmoud A, Bhatt DL, Bavry AA. Mechanical Thrombectomy for Acute Ischemic Stroke: A Meta-Analysis of Randomized Trials. J Am Coll Cardiol. 2015 Dec 8;66(22):2498-505.

7) Messori A, Fadda V, Maratea D, Trippoli S. New endovascular devices for acute ischemic stroke: summarizing evidence by multiple treatment comparison meta-analysis. Ann Vasc Surg. 2013 Apr;27(3):395-6.

On 2015 Nov 19, Kenneth Witwer commented:

Shu et al have conducted a fascinating analysis of miRNA sequences. It is unclear, however, how this analysis relates to dietary, circulating, and "transportable" categories of RNA that are hypothesized to be absorbed from the diet in functional form. Although the stated intention was "to heavily rely on experimental data to identify features that can differentiate secreted miRNAs from the rest," the data in question are either unreliable (the circulating miRNA data is questionable, based on only one biological and technical measurement), incompletely described (assignment of "exosomal" status based on the two cited databases is unclear) or missing (the pivotal milk uptake experiment--unless I'm missing something). Thus, the practical validity of the sequence analysis cannot be assessed.

Endogenous miRNAs are classified by Shu et al as circulating or not based on a list from Weber JA, 2010. This preliminary publication reported only one qPCR threshold cycle measurement for each of several hundred miRNAs using only one sample of pooled plasma. Other issues, such as a lack of correlation with results of other studies and a failure to detect abundant plasma miRNAs, such as miR-16 and miR-223, were previously noted in Watson AK, 2012. Thus, the "circulating" classification made by the authors is not supported by reliable data. Perhaps the authors might wish to revisit their study with a more comprehensive ranking of plasma miRNAs supported by reliable public sequencing and microarray data.

Which miRNAs are packaged into extracellular vesicles is an even more complicated question than simple presence in circulation. Since the majority of miRNAs in circulation appear to be in free protein complexes, not EVs, contaminants of EV preparations have strong potential to skew experimental results. It would be helpful if the authors could clarify how they used the EVpedia and ExoCarta databases to identify EV-packaged miRNAs and how this information (presumably including abundance ranks?) was used in the study.

Also unclear was where to obtain the sequencing data from the described milk feeding experiments. Although all data were said to be found on a university website, I could not find the sequencing data there or elsewhere. A public link to these data and further clarification of how they were used to validate the findings would be very helpful, as well as consistent with journal guidelines. Perhaps I missed this link?

I would note that the evidence in support of the dietary miRNA transfer hypothesis described as "unambiguous" consists of a study by the authors. The results of this study have not been confirmed. Alternative hypotheses (Witwer KW, 2014) were omitted, as well as published evidence that contradicts the hypothesis, most strikingly a recent study (Title AC, 2015) in which no miRNA uptake was observed from milk in miR-200c and miR-375 knockout mouse pups.

In conclusion, the sequencing analysis looks quite interesting, but the underlying assumptions are debatable at best.

On 2016 Aug 08, UFRJ Neurobiology and Reproducibility Journal Club commented:

The article by Schreiber et al. touches on an interesting subject: namely, the relation between actin remodeling and synaptic plasticity. However, some of the experiments presented raise concerns related to the analysis of nested data, which violates the assumption of independency between experimental units required by conventional statistical methods – a rather common problem in neuroscience papers (Aarts E, 2014).

On figure 3 (K-O) the legend mentions a sample size is of clusters (i.e. GFP-positive PURA-containing mRNP particles), observed in hipocampal cell cultures derived from transgenic and wild-type animals. However, there is no mention of how many different animals were used. If more than one cluster measured came from the same animal, they do not constitute independent samples; thus, statistics based on clusters should not be used to evaluate hypotheses relating to the effect of genotype. However, it is unclear on the text whether this was the case. On figure 7 (F) it is stated that 15/16 slices used were obtained from 9/10 animals per genotype; therefore, some of the animals contributed with more than one slice, leading to the same problem of non-independence between units. Still on figure 7 (A – B) and on figure 8 (E) the sample size is given in number of cells, and again it is unclear whether they came from different animals or not. The presence of nested data in an experiment tends to lead experimental values obtained from the same animal/cell to be more similar than if they were obtained from different animals. This causes differences between units within experimental groups to be smaller than those between groups, and thus increases the type 1 error rate of statistical tests that assume independent observations (such as t tests and ANOVA), leading to more false-positive results. A simple correction for this problem would be to calculate a mean value for each animal, and then use animal-level statistics for each comparison. Alternatively, multi-level analysis can be used to try to separate variances at the different levels (e.g.slice-level vs. animal-level variation). (Aarts E, 2014).

Another concern in this article is the analysis of subgroups. On figure 3 (F-O), the clusters were divided into 2 groups based on their mobility, and each of these groups was evaluated in 3 parameters: total distance moved, maximal velocity and maximal distance to origin. This amounts to 6 analyses in total, 2 of which had a statistically significant result. It is not clear from the methods whether dividing the clusters was an a priori decision, or whether the authors perceived the whole group as heterogeneous and therefore decided to divide it. Such a posteriori analysis of subgroups will inevitably increase the number of comparisons performed, and thus the chance of obtaining false-positive results by chance (Lagakos SW, 2006). In this case, statistical significance thresholds could have been corrected by the authors to account for the number of comparisons generated by subgroup analysis.

On 2016 Feb 04, Martine Crasnier-Mednansky commented:

The authors did not do their homework.

In E. coli, sorbitol is transported by the PTS (see PMC comment attached to Aidelberg G, 2014).

Jacques Monod (Ph.D. Thesis, 1942) classified xylose and arabinose as two class B sugars, and further reported that growth on a mixture of two class B sugars does not lead to diauxie (of interest here, Monod also reported similar growth rates for his E. coli xylose- and arabinose-growing strains). Why are the authors discussing the results of a paper Groff D, 2012 reporting strategies to eliminate a diauxie with xylose and arabinose?

On 2018 Feb 01, Andrew Willetts commented:

The poor quality of the modelling studies of 3,6-diketocamphane monooxygenase (3,6-DKCMO) presented by Isupov et alresults from a combination of a number of well characterised deficiencies (1) in relevant structural and biochemical features that were used to develop their proposals for this Type II Baeyer-Villiger monooxygenase (2) from camphor-grown Pseudomonas putida NCIMB 10007.

An important contributory factor that stymies their studies is Isupov et al’s mistaken understanding of the nature and significance of the facial diastereoselectivity of the hydride transfer that occurs between the donated FNR cofactor and the enzyme in the active site of this flavin-dependent two-component monooxygenase (fd-TCMO [3]). Both Isupov et al’s structural and modelling studies place a significant dependence on comparative data for the luciferase from Vibrio harveyi, the first fd-TCMO to be recognised and well-characterised at both the structural and functional levels (4). However, while it has been conclusively shown that 3,6-DKCMO and the two enantiocomplementary 2,5-diketocamphane monooxygenases (2,5-DKCMOs) from camphor-grown P.putida NCIMB 10007 deploy requisite FNR exclusively in (si)-facial hydride transfers (5-7), the luciferase from V.harveyi (8,9) and all other bacterial luciferases thus characterised to date (10), deploy FNR exclusively in (re)-facial hydride transfers. This fundamental dichotomy between 3,6-DKCMO and the luciferase is not taken into account by Isupov et al who incorrectly state that ‘all enzymes of the bacterial luciferase superfamily catalyse their reaction on the si side of the ring’.

Consequential errors that result from this significant misunderstanding occur principally, but not exclusively, in Sections 3.7 (Comparison with other bacterial luciferase-family proteins) and 3.8 (The reaction mechanism) of Isupov et al’s paper. Typically, if this important biochemical difference had been appreciated, then Figure 5 of their paper might have been interpreted differently. Also, because 3,6- and the isoenzymic 2,5-DKCMOs have been shown to exhibit the same extremely high (si)-facial diastereoselectivity (5-7) with respect to the hydride transfers that characterise key biochemical events in their active sites, the outcome of which is the successive formation and stabilisation of their respective Criegee intermediates (11), Isupov et al’s prediction that these enantiocomplementary enzymes will exhibit ‘a different mode of cofactor binding’ seems highly unlikely.

The established differences in facial diastereoselectivity between these particular fd-TCMOs may help to explain why the commercially available (Sigma Aldrich) flavin reductase component of the luciferase from Vibrio harveyi failed to promote any significant electrophilic biooxidation of a small number of tested organosulfides by purified preparations of 3,6-DKCMO (12). Similar low levels of product(s) were detected in both experimental and equivalent control reactions (eg, <0.01% sulfoxide and <0.002% sulfone formed after 30h incubation with methyl phenyl sulphide +/- 3,6-DKCMO). It was concluded that the negligible levels of oxidation observed were principally, if not exclusively, the result of abiotic autooxidation, and consequently this particular research initiative was abandoned in mid-1996. Also, because they were outside the remit of the PhD programme of Jean Beecher (supervisor Dr Andrew Willetts, degree awarded 1997, University of Exeter), no equivalent studies of potential nucleophilic biooxidation of ketone substrates were considered or undertaken (13,14). Consequently, Dr Beecher’s thesis is notable for the total absence of any relevant content relating to either electrophilic or nucleophilic biooxidations with a hybrid P.putida-V.harveyi multienzyme complex. The claims in Isupov et al that ‘the commercially available Vibrio harveyi flavin reductase (Sigma) was able to demonstrate activity with purified 36DKCMO oxygenating enzyme in biotransformation reactions (McGhie, 1998)’, and that ‘commercially available NADH-FMN oxidoreductase from Vibrio harveyi has successfully been used for reduction of cofactor in activity measurements (McGhie, 1998)’ are incompatible with the above pre-1998 outcomes. McGhie is an accredited co-author of Isupov et al’s paper, and McGhie (1998) is in reference to her PhD awarded by the University of Exeter (supervisor Dr Littlechild). Most significantly, there is a total absence of either supporting data, or corresponding experimental protocols, or discussion relevant to both electrophilic and nucleophilic biooxidations by a hybrid P.putida-V.harveyimultienzyme complex in McGhie’s 1998 thesis, the sole relevant entry being a single sentence on p74 which claims that the hybrid complex can support ‘lactonising activity’ (= nucleophilic biooxidation), citing the source as ‘Beecher, personal communication’, which is clearly inconsistent with Jean Beecher’s pre-1998 studies (vide infra). The incorrect claim included in McGhie’s PhD thesis and the equivalent two incorrect claims included in Isupov et al are clearly interrelated. References. 1. Willetts, A. & Kelly, D.P. (2016). Microorganisms, 4, 38: 2. Willetts, A. (1997). Trends Biotechnol., 15, 55-62: 3. Ellis, H.R. (2010). Arch. Biochem. Biophys. , 497, 1-12: 4. Campbell, Z.T., Weichsel, A., Montfort, W.R. & Baldwin, T.O. (2008). Biochem. 48, 6085-6094: 5. Grogan, G. (1995). PhD Thesis, University of Exeter: 6. Beecher, J.E., Grogan, G., Roberts, S. & Willetts, A. (1996). Biotechnol. Lett., 18, 571-576: 7. Kelly, D.R., Knowles, C.J., Mahdi, J.G., Wright, M.A., Taylor, I.N., Roberts, S., Wan, P., Grogan, G., Pedragosa-Moreau, S. & Willetts, A.(1996). Chem. Commun., 20, 2333-2334: 8. Yamazaki, S., Tsai, L. & Stadyman, T.C. (1980). J. Biol. Chem., 255, 9025-9027: 9. Yamazaki, S., Tsai, L. & Stadyman, T.C., Teshima, T., Nakaji, A. & Shiba, T. (1985). Proc. Nat. Acad. Sci. USA, 82, 1364-1366: 10. Villa, R. & Willetts, A. (1997). J. Mol. Catal. B: Enzym., 2, 193-197:11. Yachnin, B.J., Sprules, T., McEvoy, M.B., Lau, P.C.K. & Berghuis, A.M. (2012). J. Amer. Chem. Soc., 134, 7788-7795: 12. Willetts, A. & Beecher, J.E. (1995; 1996). Laboratory records, unpublished data: 13. Willetts, A. (1996). Laboratory records, unpublished data: 14. Beecher, J.E. (1997). PhD Thesis, University of Exeter.

On 2017 Jul 12, Andrew Willetts commented:

I thank Drs Littlechild and Isupov for their recent Comment, which raises some interesting new issues.

As previously (1-3),the terms ‘model/structural model’ refer exclusively to material presented in Sections 3.5 – 3.7 in Isupovet al(4), and not the experimentally determined crystalline structures of 3,6-diketocamphane monooxygenase (3,6-DKCMO) presented in Sections 3.2 and 3.3. As in (5), misconceptions will arise if that important distinction is not appreciated and respected.

Isupov et al’s study (4) of 3,6-DKCMO, a flavin-diffusible two-component monooxygenase (TCMO) from camphor-grown Pseudomonas putida NCIMB 10007, determines to high resolution (1.9 A and low R-values) the three-dimensional structures of both the native enzyme and the corresponding oxidised flavin (FMN)-bound complex, which have been deposited with the PDB (code: 4UWM). In strict contrast, however, their modelling proposals detailed in sections 3.5 - 3.7 offer no sound biochemical perspective on which amino acid residues promote the formation and stabilization of the C4a-hydroperoxyflavin intermediate essential for catalytic activity, and consequently offer no mechanistic insight on 3,6-DKCMO. The binding of FMN proposed by Isupov et al, which they report as retained by the reduced form of the flavin FNR, is based on the electron density shown in Figure 3. Significantly, it is not possible to accurately assign the orientation of the isoalloxazine ring of FMN from these crystallographic data, and consequently the locations of N5 and the heterocyclic moiety of this tricyclic ring, which both serve as key determinants of functional activity, are ill-defined. Figures 4a and 5 therefore have little experimental basis.

It is likely that a combination of structural and biochemical issues may be contributing to this striking dichotomy between high resolution and low functional definition.

Structural issues - how valid are Isupov et al’s resolved crystal structures?

As a flavin-diffusible-TCMO, FMN has no direct involvement with the sequence of biochemical events catalysed by the oxygenating moiety of 3,6-DKCMO. However, at the time the structural studies were undertaken (6), 3,6-DKCMO was mistakenly characterised as a flavoprotein using bound FMN as a coenzyme (7), which would explain why Isupov et alchose this biochemically inert form of the cofactor in developing their structural studies. The structure of the FMN complex was solved by MR with the refined structure of native apo 3,6-DKCMO, which in turn was solved by dependency on MR of a synthetic α2 dimer of the luciferase from Vibrio harveyi (6) derived from Fisheret al’s (8) 2.4A resolution of the crystal structure of the native apo α/β heterodimeric form of this bacterial enzyme (PDB code: 1luc). This dependency on outcomes from Fisher et al’s 1995 study is important in this context because it predates the recognition of the relevance of structural allostery in influencing the conformation, and hence functional activity, of ligand-regulated heteromultimeric proteins (review: 9). Thus whereas 3,6-DKCMO is homodimeric, V.harveyi luciferase is an α/β heterodimer in which the β subunit, although not catalytically active, is known to have a profound allosteric effect on the 3-D shape and hence catalytic activity (>5-fold) of the α subunit resulting from a structural allosteric pathway triggered by the binding of the fully reduced flavin cofactor (FNR) as the effector ligand. A more recent study of this luciferase by Campbell et al(10) recognised and partially characterised this cofactor-triggered allosteric effect. Specifically it highlighted the importance of a mobile loop adjacent to the active site in promoting the relevant conformational changes, a structural feature that was crystallographically disordered in Fisheret al’s study (8). Thus their 1995 crystallographic studies (8) will have been compromised because these multiple major structural effects were not taken into account, and as a consequence, any structural study reliant on the synthetic α2 dimer (6) derived therefrom will be similarly compromised.

Biochemical issues - factors relevant to the active site that could result in low functional definition

1. The respective Kd values for FMN & FNR differ by >500% (1) which calls into question Isupov et al’s active site modelling proposals (1 – 3). It is likely that any FMN binding is either random or at best very weak, as evidenced by other flavin-diffusible TCMOs (11).

2. 3,6-DKCMO and V.harveyi luciferase have experimentally confirmed opposite facial diastereoselectivity with respect to the alignment of FNR and divestment of its reducing power within the active sites of the two enzymes (12). This factor was not taken into account by Isupov et al’s modelling studies which were developed in the mistaken belief that 3,6-DKCMO like ‘all other enzymes of the bacterial luciferase superfamily catalyse their reaction on the (si)-side of the ring’.

3. The tricyclic ring structure of F420 is an unwise precedent for modelling FNR into the active site because it lacks the catalytically crucial N5 atom.

Littlechild and Isupov’s Comment leaves unresolved two other interrelated issues (3), providing only evasion and obfuscation. The claim (4) that ‘Partial sequencing of the large CAM plasmid has now identified a flavin reductase adjacent to the 3,6-DKCMO gene on the CAM plasmid’ was made on the basis of partial sequencing of a corrupted sample of plasmid plus chromosomal DNA (13). The claim is incompatible with directly relevant seminal research by Iwaki et al (14) which reported no evidence for an FR-coding open-reading frame on the CAM plasmid. The three principal authors of (14) are also cited as co-authors of (4), which was published 30 months later, yet incomprehensibly the latter includes no details of or reference to their own directly relevant prior seminal research.

A number of these considerations will also be highly relevant to any equivalent structural studies being conducted (5) on either of the two the isoenzymic forms of 2,5-diketocamphane monooxygenase from Pseudomonas putida NCIMB 10007 (14).

References: 1. Willetts, A.; Kelly, D.R. (2016). Microorganisms,4, 38, doi: 10.3390/microorganisms4040038: 2. Willetts, A. (2016). PubMed Commons, Oct 27: 3. Willetts, A. (2017). PubMed Commons, Jun 05: 4. Isupov, M.N.; Schröder, E.; Gibson, R.P.; Beecher, J.; Donadio, G.; Saneei, V.; Dcunha, S.A.; McGhie, E.J.; Sayer, C.; Davenport, C.F.; Lau, P.C.; Hasegawa, Y.; Iwaki, H.; Kadow, M.; Balke, K.; Bornscheuer, U.T.; Bourenkov, G.; Littlechild, J.A. (2015). Acta Crystallogr. D, 71, 2344; 5. Littlechild, J.A.; Isupov, M.N. (2017). PubMed Commons, Jun 08: 6. Isupov, M.N.; Lebedev, A.A. (2008). Acta Crystallogr. D, 64, 90: 7. Taylor, D.G.; Trudgill, P.W. (1986). J. Bacteriol., 165, 489: 8. Fisher, A.J.; Raushel, F.M.; Baldwin, T.O.; Rayment, I. (1995). Biochem., 34, 6581: 9. Raman, S.; Taylor, N.; Genuth, N.; Fields, S.; Church, G.M. (2014). Trends Genet., 30, 521: 10. Campbell, Z.T.; Weichsel, S.; Montfort, W.R.; Baldwin, T.D. (2009). Biochem., 48, 6085: 11. Ellis, H.R. (2010). Arch. Biochem. Biophys., 497, 1; 12. Villa, R.; Willetts, A. (1997). J. Mol. Catal.: B Enzymatic, 2, 193: 13: Littlechild, J.A.; Isupov, M.N (2017). Submission to the Editors, Acta Crystallogr. D, Mar 18: 14. Iwaki, H.; Grosse, S.; Bergeron, H.; Leisch, H.; Morley, K.; Hasegawa, Y.; Lau, P.C.K. (2013). Appl. Environ. Microbiol., 79, 3282.

On 2017 Jun 13, Andrew Willetts commented:

None

On 2017 Jun 13, Andrew Willetts commented:

None

On 2017 Jun 08, Michail N Isupov commented:

We dispute the comments made on PubMed by Willetts regarding the paper ‘The oxygenating constituent of 3,6-diketocamphane monooxygenase from the CAM plasmid of Pseudomonas putida: the first crystal structure of a type II Baeyer-Villiger monooxygenase’ Michail N Isupov, Ewald Schröder, Robert P Gibson, Jean Beecher, Giuliana Donadio, Vahid Saneei, Stephlina A Dcunha, Emma J McGhie, Christopher Sayer, Colin F Davenport, Peter C Lau, Yoshie Hasegawa, Hiroaki Iwaki, Maria Kadow, Kathleen Balke, Uwe T Bornscheuer, Gleb Bourenkov, Jennifer A Littlechild, Acta Cryst. D Biol. Cryst, 71, 2015. In both pubmed comments Willetts fails to understand that it is the fit of the model to experimental data (crystallographic R-factor and FreeR) that serves as a measure of the quality of crystallographic structures and not the sequence similarity of the molecular replacement model. The use of an artificial dimer of bacterial luciferase, which has only 16% sequence similarity to 3,6-diketocamphane monooxygenase , to obtain the lost phase information and solve the well refined structure of the 3,6-DKCMO as reported in the paper is a very impressive use of molecular replacement which has already been reported in an earlier Acta Cryst paper (Isupov and Lebedev, 2008 Acta Cryst.D 64, 90-98) and was selected for a presentation at the CCP4 annual workshop in 2007 on Molecular Replacement. The X-ray data that was collected for both structures reported in our paper extends to beyond 2 Å, and the R-factors of the refined models, which were deposited in the Protein Data Bank, were compliant with their deposition criteria (pdb ref. Rose, P. W., Prlic´, A., Bi, C., Bluhm, W. F., Christie, C. H., Dutta, S., Green, R. K., Goodsell, D. S., Westbrook, J. D., Woo, J., Young, J., Zardecki, C., Berman, H. M., Bourne, P. E. & Burley, S. K. (2015) Nucleic Acids Res. 43, D345–D356). The 3,6-DKCMO structural paper was refereed by experienced crystallographers and found acceptable for publication in this well respected structural biology journal. The X-ray data shows the FMN molecule to make a number of hydrogen bonds in the monooxygenase enzyme active site. Willetts believes that the FNR (the reduced form of FMN) binding to the enzyme will result in a completely different hydrogen bonding pattern to that observed in the reported structure of the FMN monooxygenase complex. However he does not suggest any experimental evidence for this except for the increase of the enzyme affinity for the reduced form of the cofactor. We suggest that binding of the reduced flavin molecule FNR with the isoalloxazine ring in the ‘butterfly’ conformation as shown results in retention of the hydrogen bonding observed in the structure of the FMN monooxygenase complex and in an increase of hydrophobic interactions. The latter would result in the observed large increase in affinity of enzyme for the reduced cofactor. The H-bonding between N5-H and a Ser/Thr residue has not been observed in any structure determined to date of the bacterial luciferase enzyme family. In our 3,6-DKCMO structure there appears to be no Ser/Thr amino acid residue that could take on this role within the enzyme active site. The other comments of Willetts do not directly concern the main topic of the paper which describes the structural studies on the oxygenating subunits of 3,6-DKCMO, since they address the reductase enzyme. The question which we and others had investigated regarding whether this enzyme was found on the CAM plasmid or whether it was an enzyme on the Pseudomonas genome was unanswered for a number of years. We unfortunately omitted to include a reference from Iwaki et al, 2013 on this topic. We have agreed to include this reference in our next structural paper on the related 2,5-DKCMO oxygenating subunits which will shortly be submitted.

Jennifer Littlechild and Michail Isupov

On 2017 Jun 05, Andrew Willetts commented:

The modelling studies of 3,6-diketocamphane monooxygenase(3,6-DKCMO) from camphor-grown Pseudomonas putida NCIMB 10007 presented by Isupov et al in this paper are characterised by serious weaknesses (1) which result from a combination of reasons (2) including deploying molecular replacement resolution based on another protein with only 16% sequence homology, a value significantly below the lower limit accepted by structural biologists as highly likely to generate randomness (3). However, in addition to these major flaws, this paper is most notable for the inclusion of an unsupported claim to have identified a flavin reductase (FR) gene adjacent to the 3,6-DKCMO gene on the CAM plasmid, a claim which is contra to well-founded pre-existing knowledge, and which only serves to undermine the validity of the science data base.

For over 50 years, 3,6-DKCMO was mistakenly believed to be a loosely linked multienzyme complex consisting of a flavoprotein monooxygenase supported by another subunit that generated the reduced flavin coenzyme (reduced FMN = FNR) necessary for catalytic activity, and which has been variously described as ‘electron transport oxidase’ (4), NADH;(acceptor) oxidoreductase (5), or NADH dehydrogenase (6). However, recent seminal studies reported in May 2013 by Iwaki et al (7) confirmed that 3,6-DKCMO is not a flavoprotein, but rather an FMN-dependent member of the two-component monooxygenase (TCMO) group (8). Like other similar TCMOs, 3,6-DKCMO requires a separate FR to supply readily diffusible FNR as the reduced flavin cofactor necessary to undertake successful oxygenating activities. Extensive studies of the PCR amplified isolated pure CAM plasmid DNA by Iwaki et al confirmed unequivocably that there was no orf corresponding to a FR anywhere on the plasmid. Rather, Iwaki et al’s comprehensive study identified Fred, a chromosome-coded 36 kDa homodimeric FR as the principal source of FNR for 3,6-DKCMO during late log phase growth of P.putida on camphor-based minimal medium, a conclusion confirmed by a second more recent study (1).

The claim included by Isupov et al in their November 2015 publication that ‘Partial sequencing of the large CAM plasmid has now identified a flavin reductase adjacent to the 3,6-DKCMO gene on the CAM plasmid (Littlechild and Isupov, unpublished data)’ is incompatible with Iwaki et al’s extensively researched conclusions published 30 months previously (7). Their unsupported claim defies any obvious logical explanation. This is all the more so as firstly, Isupov et al were aware as long ago as 2008-9 that their relevant CAM plasmid DNA samples were contaminated with chromosomal DNA from P.putida (9), and secondly as Profs Iwaki, Hasegawa and Lau (the senior authors of [7]) are listed as co-authors of Isupov et al’s November 2015 publication. In this latter respect it is highly significant that Isupov et al’s paper fails to include a single reference to or citation of Iwaki et al’s seminal research and its highly relevant outcomes reported 30 months previously.

Perhaps one or more of the senior authors of either or both relevant publications can provide some insight, enlightenment, and/or explanation of these bizarre contradictions and inconsistencies?

References: 1. Willetts, A. & Kelly, D.R. (2016). Microorganisms,4, 38, doi: 10.3390/microorganisms4040038: 2. Willetts, A. PMC (2016). Oct 27: 3. Rost, B. (1999). Protein Eng 12 85-94: 4. Conrad, H.E., Lieb, K. and Gunsalus, I.C. (1965). J. Biol. Chem., 240, 4029-4037: 5. Trudgill, P.W., DuBus, R. and Gunsalus, I.C. (1966). J. Biol. Chem.,241, 1194-1205; 6. Taylor, D.G and Trudgill, P.W. (1986). J. Bact., 165, 489-497: 7. Iwaki, H., Grosse, S., Bergeron, H., Leisch, H., Morley, K., Hasegawa, Y. and Lau, P.C.K. (2013). Appl. Envron. Microbiol., 79, 3282-3293; 8. Ellis, H.R. (2010). Arch. Biochem. Biophys., 497, 1-12: 9. Littlechild, J.A & Isupov, M.N (2017). Submission to the Editors, Acta Cryst. D.

On 2016 Oct 27, Andrew Willetts commented:

I submitted a Letter for consideration for publication in Acta Cryst. Section D addressing a number of significant deficiencies (both errors and omissions) that characterise Isupov et al’s open access publication, but this was rejected by the Section Editor.

To attempt to redress the balance, here I provide an outline of just some of the points raised in that Letter to allow a wider audience the opportunity to reflect on the merits or otherwise of Isupov et al’s model. Science progresses by debate and discussion, not by censorship.

Comments on modelling the native apo structure of 36DKCMO

Type II Baeyer-Villiger monooxygenases (1) such as the 3,6-diketocamphane monooxygenase (36DKCMO) induced in camphor-grown Pseudomonas putida NCIMB 10007 are not true flavoproteins, as the flavin serves as a cosubstrate rather than a coenzyme. Along with the various bacterial luciferases, they are members of the FMN-dependent two-component monooxygenases (2).

Because of the relatively low-grade nature of the collected data, it was necessary to use Molecular Replacement (MR) to achieve resolution of the native apo structure. However, the choice of a synthetic α2 dimer of the luciferase from Vibrio harveyi for structure determination is misconceived, given that the two enzymes differ significantly in key structural and functional characteristics. Thus whereas 36DKCMO is homodimeric (3), the luciferase from V. harveyi is an α/β heterodimer (4). Of particular relevance, research (5) that combined extensive crystallographic studies undertaken on the α/β luciferase complex allied to mutational changes to the key amino acid residue βTyr151, has confirmed earlier predictions (6) that the β subunit plays an important allosteric role in establishing the catalytically active form of the α subunit, a feature that the synthetic α2 dimer does not take into account. The importance of this dynamic structural relationship is emphasised by a comparison which demonstrated that the quantum efficiency of the bioluminescent reaction undertaken by the α subunit was 5+ orders of magnitude lower than that undertaken by the α/β heterodimer (7).

Comments on modelling the flavin-bound 36DKCMO by Isupov et al

Isupov et al‘s use of FMN as the ligand of choice for the development of modelling studies of the active site of 36DKCMO is doubly problematical: i) Firstly, 36DKCMO is not a flavoprotein and does not generate requisite reduced FMN (FNR) in situfrom pre-bound FMN coenzyme. Rather, studies have confirmed that it accepts FNR generated from FMN by one or more competent flavin reductases (FR) present in camphor-grownP. putida NCIMB 10007 (8,9), rapid free diffusion of FNR between the two participating enzymes being a characteristic of the FMN-dependent two-component monooxygenases (2). There is a >500-fold difference in the dissociation constant (Kd) of FNR (0.24+/- 0.02 μM), and FMN (125.3 +/- 5.1 μM) recorded for 36DKCMO which suggests that any interaction of the oxidised flavin cosubstrate with the monooxygenase will be comparatively random. ii) Secondly, FMN and FNR differ in certain key structural characteristics which are likely to significantly influence the orientation of the flavin within the 3-D structure of 36DKCMO. The tricyclic isoalloxazine ring that is the basic functional feature of all flavins is planar in the fully oxidised state, whereas it is bowed through the N5-N10 axis into the so-called ‘butterfly’ conformation in FNR. The observed extent of bowing of the reduced flavin can be as high as 35^o (10), and is an idiosynchratic characteristic of any given FNR-dependent enzyme, Although the significance of the difference in conformation between the oxidised and reduced forms of the isoalloxazine ring of flavin cofactors is acknowledged by Isupov et al, they chose an arbitrary deviation from polarity (20^o) for the reduced isoalloxazine ring of FNR within the proposed active site of the enzyme, based on the past precedent of Adf (11). However, Adf is an F420-dependent enzyme, and the 5-deazoisoalloxazine ring of F420 differs from the tricyclic ring of FNR in a number of important structural and functional features, most significantly the absence of the key heteroatom N5 (vide supra). Because no illustration of their proposed FNR model is presented by Isupov et al, nor is it stated about which axis of the tricyclic ring the 20^o conformational modification was modelled in, it is not possible to critically examine their claim that FNR ‘appears to retain the hydrogen bonding observed for the oxidised FMN’. However, very large difference in the Kd values for the two forms of the flavin cosubstrate recorded with 36DKCMO (vide infra) is not compatible with this claim.

That Isupov et al’s model is significantly flawed is confirmed by a comparative study of related enzymes. Many crystal structures of other Class C (FMN-dependent) and Class D (FAD-dependent) two-component flavin-dependent oxygenases (12) have been solved recently, and although the similarity between them is generally low, the overall pattern of structural folding is well preserved (13). Above all, the feature of H-bonding interactions between the N5-H of the relevant flavin cofactor and a hydroxyl group of Ser or Thr in the active site of the enzyme, which is crucial for subsequent C4a-hydroperoxyflavin formation and stabilisation, is conserved in every other TCMO characterised at this level (14, 15). It is significant that none of the schematic drawings of the active site of flavin-bound 36DKCMO complex presented by Isupov et al are consistent with this key functional feature of TCMOs.

In summary, Isupov et al’s model is likely to be a poor basis for gaining insight into the molecular mode of action of 36DKCMO and other Type II Baeyer-Villiger monooxygenases.

References: 1. Willetts, A. (1997). Trends Biotechnol., 15, 55-62: Ellis, H.R. (2010). Arch. Biochem. Biophys., 497, 1-12: 3. Jones, K.H., Smith, R.T, & Trudgill, P.W. (1993). J. Gen. Microbiol., 139, 797-805: 4. Hastings, J.W., Potrikus, C.J., Gupta,S.C., Kurfurst, M. & Makemson, J.C. (1985). Adv. Microb. Physiol., 26, 235-291: 5. Campbell, Z.T., Weichsel, A., Montfort, W.R. & Baldwin, T.O. (2009). Biochem., 48, 6085-6094: 6. Meighen, E.A. & Bartlet, I. (1980). J. Biol. Chem., 255, 11181-11187: 7. Waddle, J. & Baldwin, T.O. (1991). Biochem. Biophys. Res. Commun., 178, 1188-1193: 8. Willetts, A. & Kelly, D.R. (2014). Microbiology , 160 , 1784-1794: 9. Willetts, A. & Kelly, D.R. (2016). Microorganisms,4, 38, doi: 10.3390/microorganisms4040038: 10. Lennon, B.W., Williams, C.H. & Ludwig, M.L. (1999). Prot. Sci., 8, 2366-2379: 11. Aufhammer, S.W., Warkentin, E., Berk, H., Shima,S., Thauer, R.K. & Ermler, U. (2004). Structure, 12, 361-370: 12. van Berkel, W.J., Kamerbeek, N.M. & Fraaije, M. (2006). J. Biotechnol., 124, 670-689: 13. Chaiyen, P., Fraaije, M.W. & Mattevi, A. (2012). Trends Biochem. Sci., 37, 373-380: 14. Thotsaporn, K., Chenprakhon, P., Sucharitakul, J., Mattevi, A., Chaiyen, P. (2011). J. Biol. Chem., 286, 28170-28180: 15. Visitsatthawong, S., Chenorakhon, P., Chaiyen, P., Surawatanawong, P. (2015). J. Amer. Chem. Soc., 137, 9363-9374.

On 2016 Jan 14, Lydia Maniatis commented:

Conceptual problems The authors argue that when the visual system gets things wrong, the problem is that the scene is “impoverished” and “unnatural.” Such terms are not specific enough to allow them to be evaluated scientifically. If we use the term “natural” to mean created by natural processes, then the stimuli used in this study were not natural. If the authors mean more than that, they need to both specify what they mean and to control experimentally for the relevant factors. Otherwise, the definition of natural becomes tautological – any scene that produces more or less veridical percepts is defined as “natural.” A non-tautological step forward require examining cases where vision fails, discerning a potential distinction between the features of the cases were it succeeds and the cases where it fails, and testing the assumption that the distinguishing factors matter. Simply defining conditions where vision fails as “impoverished” and “unnatural” doesn't allow such a test (and/or the claims are easily falsified); it leads to confusion, since, for example, it would be difficult to argue that a photograph is an “impoverished” stimulus, or, again, that man-made objects are “natural.” If the authors are suggesting that all types of asymmetry compromise the accuracy of aspects of the 3D percept, then they need to rationalize the claim theoretically and/or frame it precisely enough that it can be tested. (I'm quite sure any clearly-framed symmetry claim can be easily falsified.)

It is surprising, finally, that the authors claim not to understand the value of illusions in studying perception. The value is in testing hypotheses vis a vis the principles underlying visual performance. For example, the pictorial perception of 3D shapes, the Ames room, contradict Marr's 2.5D sketch concept, i.e. the primacy of depth maps in perception of a third dimension.

The confusion as to the role of illusions is correlated with an epistemological confusion as to the use and logic of falsification. “It follows,” write the authors, “that with the types of models we are using, falsification is not your “best friend” as it often is elsewhere. In 3D vision, there is usually no model you can turn to, so there is nothing to falsify. Simply put, scientific discovery in 3D vision is not accomplished by using an ANOVA or Bayesian tests to reject some hypotheses. Discovery is accomplished by correctly guessing which cost function is actually being used by the visual system. objects within this space veridically.” But to corroborate a guess (the asumptions incorporated in a model) we must test it, and the test may prove it wrong, i.e. may falsify it. This does not mean testing merely by using ANOVA's to show significance, or lack thereof, in cases where the outcome is foggy, or “Bayesian” tests (to show...whatever), but actually showing that our assumptions – e.g. that binocular vision is not necessary or sufficient to the perception of 3D shape - do, or do not, hold up to rigorous testing.

Finally, it's not made clear which “Gestalt-like” constraints are being referred to. (Organisational principles are necessary to all perception; is there meant to be more specificity than this?)

On 2016 Jan 14, Lydia Maniatis commented:

This study does not seem to extend the reasonable (a rare thing) approach of the Pizlo group to 3D shape perception (a fundamental and much-neglected topic). The study addresses a. the ability of humans to accurately perceive the proportions of “3D indoor scenes” and b. the ability of a robot to reproduce such scenes using a computational model constructed (in part) by the investigators. However, while in the past the group was interested in discerning principles underlying human performance and using them as the basis for building computational models (which could then serve both the theoretical goal of testing the hypothesized principles and of constructing a computer that can “see like us”), here there appears to be no theoretical connection between human and robot performance.

The way the robot goes about the task is not the way humans can or do. It involves constructing a depth map (using a camera's inbuilt program) on the basis of binocular disparity and triangulation. Pizlo (2008) is clear on the fact human 3D perception is not built on the basis of depth maps, or '2.5D sketches' exploiting binocular disparities: “There seems little reason to afford binocularity, disparity...a critical role in natural shape perception” p. 176).

Conversely, the robot does not employ monocular principles – it cannot see with one eye. Yet in humans, monocular performance was not too much worse than binocular (which is necessary but not sufficient for the best accuracy), and the former can override the latter (e.g. the zero disparity in a view of a photograph of a 3D scene/objects doesn't prevent 3D percepts from arising).

Thus, none of the (monocular) principles that the investigators have effectively applied in previous work on 3D shape perception were in play in this computational model (which, if I understand correctly, does little more than group points associated with objects, whose relative locations are given by the camera software). If the goal is to model human vision, then the present study has not taken any additional steps toward that goal. If the goal is to make robots that can navigate the environment, then non-human mechanisms – e.g. some kind of radar – might be more straightforward.

At any rate, the human and the robot sections of this paper are unconnected, theoretically and practically.

Problems with methods Focussing strictly on the experiments with humans, we can note that the only goal the method allowed, in principle, was to confirm that under some conditions (mirror-symmetrical objects in a rectangular room with a floor perpendicular to the direction of gravity), human perception of relative sizes and relative locations of objects is reasonably good. Performance was not great, with high within and between subject variability, and scale biases.

It is not clear that the problem was not with perception, but with the task, which involved a pictorial representation of the ground plan. For non-artists, copying from life to paper/tablet can be a difficult task, and flaws in the picture don't necessarily correspond to flaws in perception (meaning that the experiment might better have used artists or draftsmen). Memory also comes into play, as the relative distances in the picture are repeatedly, and serially, checked against those of the scene. It is also known that between object/figure distances are less salient perceptually than within-object distances (see e.g. Arnheim, Art and Visual Perception (1974) p. 236-8. This book should be read by all vision scientists!)

Given these difficulties, it's doubtful that the high degree of variability in performance found by the researchers corresponds to real variability in perception. All this doesn't matter, though, because the investigators' criteria for veridicality were permissive: “The intra-trial Bias was removed by applying a uniform size scaling of the recovered scene in each trial.” Moreover, their definition of veridical corresponds narrowly to the absence of systematic, non-Euclidean geometrical distortion: “We can, therefore, conclude that our subjects’ visual space was not distorted, which means that they saw the 3D spatial arrangement of objects within this space veridically.” Given the apparent difficulty of the task and unreliability of the data, it's not actually clear that such distortions would have been detectable. Still, the conclusion that in certain conditions we see the world more or less veridically should not come as a surprise to scientists, given our ability to navigate it and act on it effectively. (In the case of those metaphysicians, e.g. D. D. Hoffman, who do not concede realism to any degree, no evidence can undermine their position (see e.g. Maniatis, Perception 2015, Vol. 44(10) 1149–1152).

The fact that the objects in the room were mirror-symmetrical and the floor perpendicular to the direction of gravity only shows that such characteristics of a scene are not inconsistent with the degree of performance rated by the investigators' as “veridical.” It does not show that performance would not be equivalent if asymmetrical objects were used, or if the floor were tilted or slanted, because there were no such control conditions. Indeed, one could argue that performance would be equivalent under ANY conditions. The claim would be incorrect, but the results of this experiment could not be used to contradict such a claim. That is, the study does not actually exclude any possible outcomes/theoretical assumptions with respect to the features of a stimulus that enable veridical perception.

On 2016 Jan 19, BSH Cancer Screening, Help-Seeking and Prevention Journal Club commented:

The HBRC discussed this paper during the journal club held on January 12th 2016. We read the paper with great interest as the topic is particularly relevant for our group. We also considered it as a very timely and useful contribution to the discussions taking place in the wider screening community on issues of overdiagnosis/overtreatment and informed decision making.

The study examined how using different terms for describing ductal carcinoma in situ (DCIS) might affect women’s concern and treatment preferences. This is important because, as stated by the authors, about 20% of screening detected breast cancers are DCIS. Overdiagnosis and overtreatment can be an important problem for DCIS and it has been suggested that a possible approach to reduce this problem may be to change the name and use alternative terms avoiding the word cancer and carcinoma.

The group agreed that the authors should be acknowledged for their study which used a randomised design within a broader community survey. This allowed to include women with different socio-demographic characteristics and different prior screening experiences. However, as already pointed out by the authors, the group felt that the women’s responses to hypothetical scenarios might not necessarily reflect real life situations. For example, in a real life clinical context, after receiving a DCIS diagnosis, treatment options would be discussed with a clinician and the information and advice provided by the doctor would be important for helping the patient to make the best treatment decision. This was obviously not possible within the study context. Moreover, as also pointed out by the authors, the group felt that patients’ answers might have been strongly influenced by the statement ‘if research shows that watchful waiting is a safe and effective option’. This does not reflect the information that patients would get in a real clinical setting, considering that evidence is still lacking. Nevertheless, we agreed that it is extremely interesting to examine the effects of different terms on understanding, psychological outcomes and treatment preferences. The group would encourage more work in this area.

We also discussed advantages and disadvantages of the cross-over design. It was highlighted that once a person has been presented with a specific scenario which includes the word cancer, it is unlikely that responses to scenarios/terms presented subsequently might not be influenced by the initial exposure. Therefore it was felt that the between group comparison was more relevant than the before-after comparison within participants.

The group would have liked to have more information on how the terms used in the two study arms were selected and whether other alternative terms were excluded based on pilot testing. Future studies could explore different terminology. Furthermore, the relatively small sample size limited the possibility of examining whether results might be different for people with previous screening experience or belonging to specific age-groups. Larger studies and also research exploring the effect of different terminology in the context of other diseases (e.g. pulmonary lesions following imaging, diabetes/pre-diabetes) on understanding, patient concerns and treatment decisions would be welcome.

In conclusion, the HBRC group read the article with great interest and would encourage further studies in this area.

Conflicts of interest: We report no conflict of interests and note that the comments produced by the group are collective and not the opinion of any one individual.

On 2015 Nov 18, David Keller commented:

An interesting & important discovery, using a powerful new technique, but further study is required

This retrospective observational study demonstrated powerful and convincing associations between the use of levodopa (L-dopa) and reduced incidence, and delayed onset, of age-related macular degeneration (AMD). Bravo to these investigators, who correctly note: "Our data indicates prospective clinical trials to determine whether L-DOPA can prevent AMD are warranted". Prospective randomized trials of levodopa are indeed now required, in order to prove a cause and effect relationship between taking levodopa and the diminished and/or delayed risk of AMD. The reason is that retrospective, non-randomized studies can only demonstrate associations, not causality. Thus, the statement "exogenous L-dopa was protective against AMD" must be understood to mean that taking L-dopa was associated with lower risk of AMD, not necessarily that taking L-dopa caused a lower risk of AMD. This distinction is frequently lost when results from studies such as this are presented in the popular press, and even in the medical literature.

On 2016 Aug 30, Robert Goulden commented:

Thank you David - I think that's a fair summary of our discussion, and you usefully highlight areas where the evidence can reasonably be interpreted in different ways. I hope our exchange will be read by those interested in the article and that will enrich their understanding of this complex question.

On 2016 Aug 30, David Keller commented:

A reader disagrees with the conclusion of this study

The author's conclusion that "moderate alcohol consumption is not associated with reduced all-cause mortality" conflicts with prior studies, which have reported reduced all-cause mortality with moderate alcohol consumption, compared with non-consumption, or heavy consumption. I noted that the data in this study actually show minimum mortality for subjects who drank from 1 to 6 drinks per week, compared with non-drinkers or for heavier drinkers.

The author replied with this explanation: "You’re absolutely correct that mortality is lower in regular moderate alcohol consumers than in long-term abstainers. My paper does not contradict the existing literature on this point. The important question, however, is whether that difference is due to alcohol consumption, or due to other (confounding) factors. If it’s due to confounding factors (which I argue it is) then the comparison is not especially helpful – in fact, it’s misleading."

So, despite the title of this study, it actually confirms prior reports of significant reduction in mortality associated with moderate alcohol drinking, compared with non-drinking or heavy drinking. The author argues that this reduced mortality is not due to alcohol consumption itself, but to unidentified "confounding factors".

Occam's razor is the principle that favors the simplest explanation for an observed phenomenon, when more complicated theories exist in equipoise. For example, Occam's razor favored Newton's laws of motion until they could no longer explain all of the experimental data, at which time the more complicated Theory of Relativity was called for.

The simplest explanation for the results observed in this and prior studies is that moderate alcohol consumption is the cause of the associated reduced mortality in the populations studied. A randomized, placebo-controlled experiment is required to test the more complicated theory that some unknown confounding factor is responsible for the mortality reduction associated with moderate drinking. Meanwhile, Occam's Razor favors consideration of reduced overall mortality when discussing the potential beneficial and adverse effects of moderate alcohol consumption.

On 2016 Aug 29, Robert Goulden commented:

Thanks David. As suggested; here's those two emails:

You’re absolutely correct that mortality is lower in regular moderate alcohol consumers than in long-term abstainers. My paper does not contradict the existing literature on this point. The important question, however, is whether that difference is due to alcohol consumption, or due to other (confounding) factors. If it’s due to confounding factors (which I argue it is) then the comparison is not especially helpful – in fact, it’s misleading. In contrast, comparing moderate drinkers to occasional drinkers is much more informative in helping us figure out if alcohol reduces mortality. These groups (occasionals vs. moderates) are much more similar across a range of variables than abstainers vs. moderates, and thus any differences in mortality can more plausibly attributed to the effects of alcohol. The lack of a mortality difference between those who drink less than once per week, versus those who drink around 1 drink per day, surely calls into question a beneficial effect of alcohol? Of course, people may not be reporting their consumption levels accurately, but in that sense my study suffers the same weakness as just about every other study on this question. Hence the need for an RCT to settle this question once and for all; I sincerely hope somebody gets round to organising one sooner rather than later!

The reason I think the benefit [of moderate consumption vs. long-term abstention] survives [adjustment] is that there is significant residual confounding. This is a judgement – by definition you cannot measure unmeasured confounding – but which I think is supported by the lack of any difference between the occasional and moderate drinkers. As the Naimi et al. paper from 2005 suggests, the differences between non-drinkers and drinkers is so marked across a huge numbers of variables that it’s likely that no amount of adjustment can truly balance out the groups. Despite my extensive adjustments, in reality I’m only scratching the surface of the thousands of potential variables which affect people’s health outcomes. This gets to the heart of why observational studies, even based on pretty good quality data like the Health and Retirement Study, can only take us so far.

On 2016 Aug 29, David Keller commented:

Thank you for your outstanding replies to my questions, including your last two private emails to me, which I found very helpful and would urge you to consider posting. PubMed Commons does not permit the recipient of an email to post it, but the author can do so. I do not think anyone will object, however, if I quote your final sentence:

Hence the need for an RCT [randomized controlled trial] to settle this question once and for all; I sincerely hope somebody gets round to organising one sooner rather than later!

I concur completely, until which time we will have to agree to disagree regarding the benefits of moderate alcohol consumption.

A votre sante!

On 2016 Aug 27, Robert Goulden commented:

Hi Dr Keller,

One of the key methodological feature of my paper - in addition to extensive confounder adjustment and measurement of consumption at several different time points - is the use of occasional drinkers as the reference group. For the reasons I've outlined in the paper and in my previous comment, I believe they are a more appropriate comparator group than long-term abstainers, because the comparison is less likely to be subject to confounding. In contrast, the HR for long-term abstainers is likely subject to significant unmeasured confounding and I therefore felt should be left out of the abstract, where there is insufficient space to explain its limitations. To an extent that's a value judgement, but I think a reasonable one.

If my abstract had not mentioned that occasional drinkers were the reference group, I think that would have been grounds to publish a correction to the abstract (though not to retract the paper!). But given that I've clearly stated that occasional drinkers are the reference group, I'm not sure any change is warranted.

In a separate comment you raise a very valid point about who the 'occasional drinkers' are. You correctly suggest that someone who binges on a bottle of whisky once every 8 days would fall into that category. This reflects the structure of the questioning in the Health and Retirement Study (see question C129 onwards in the script used by interviewers), whereby those who drink 'less than once per week' are not probed for further details on how much they drink. My guess is that such drinkers are sufficiently rare to not significantly affect the results, but even if they weren't, they would presumably make the occasional drinkers group less healthy (if we can all agree a bottle of whisky every 8 days is probably bad for you!). Therefore, in so far as there are bingers among the occasional drinker group, the direction of bias would be to favour the regular light drinker group. Instead, they are almost indistinguishable (HR 1.02) from the occasional drinkers.

I hope this addresses your questions, but am happy to continue the conversation.

Regards

Rob

On 2016 Aug 27, David Keller commented:

Dear Professor Goulden,

Thanks for your helpful reply to my comments. PubMed Commons is a unique forum for authors to fully explain and defend their work, and readers to make their best effort to question results or methods.

My main point has still not been answered. Line 5 of your Table 2 indicates that the Hazard Ratio for all-cause mortality for non-drinkers was 1.19 (1.11-1.27), and for light drinkers, it was 1.02 (0.94-1.11). The associated confidence intervals touch but do not overlap. Does this not indicate a significant decrease in the risk of all-cause mortality, associated with the light consumption of alcohol? If so, then how can you conclude that "Moderate alcohol consumption is not associated with reduced all-cause mortality" in the population you studied? Your own fully adjusted data tells us that a significant decrease in the Hazard Ratio for all-cause mortality, from 1.19 to 1.02, is associated with drinking 1-6 alcoholic beverages per week. Is this not an example of alcohol consumption which is indeed associated with reduced all-cause mortality? If so, shouldn't a retraction be issued?

David L. Keller, MD, FACP

On 2016 Aug 27, Robert Goulden commented:

Hi David

Many thanks for your comments. Your key objection, in the comment here, your letter to Am J Med, and your PubMed commons comment on my response to that letter, is around the use of occasional drinkers as the reference category. You are completely right that this use of a reference category is the principal (but not sole) reason that I don’t find evidence of a benefit to moderate alcohol consumption. I had hoped that the use of this reference category was adequately explained in the paper, but it’s an important point so I’m happy to discuss it further.

Studies of the association between alcohol consumption and health are plagued with the problem of confounding. Non-drinkers and moderate drinkers differ in myriad important ways which conventional regression analyses cannot adequately adjust for (1). This leaves us with the difficult question of how to isolate the effects of alcohol on health, as opposed to the effect of all the other health-related variables which differ between non-drinkers and drinkers. One proposed solution is to use occasional drinkers as the reference category – not a novel “statistical maneuver” developed by me – but an approach used in the largest ever study of this question (2) (they were called light drinkers in that study, but the volume of alcohol consumed [0-2 g/day] meant they were occasional drinkers, as the accompanying editorial noted (3)). As I say in my paper, occasional drinkers “drink at levels for which a physiologic effect of alcohol is not plausible, but are likely to be more similar in other characteristics to moderate drinkers than long-term abstainers, thus reducing confounding”.

When this approach to addressing confounding is taken, my results are actually consistent with the wider literature, as Stockwell and Naimi note in their commentary on my paper (4). Their systematic review (5) reported “similar findings” and had my paper been included in their meta-analysis (it was published after their search window) it would have been coded as “high quality”. Other approaches which try to isolate the causal effect of alcohol and minimize confounding, such as mendelian randomization, also find no evidence of a benefit of moderate alcohol consumption (6).

Making sense of all the evidence is tricky, but my gut instinct (for what it's worth!), based on my paper and the wider literature, is that moderate alcohol consumption (up to 21 drinks per week) likely has very little effect (positive or negative) on health. My paper only finds unambiguous evidence of harm for those drinking over 21 drinks per week; of course, whether that association is driven by residual confounding is hard to say, but it’s certainly consistent with well-established links between heavy alcohol use and adverse outcomes such as liver disease, certain cancers, and trauma.

I think the claim that my abstract is ‘misleading’ isn’t warranted; I explicitly state in the abstract that occasional drinkers are the reference category, but by necessity this choice can only be fully explained in the main text.

Finally, you make some explicit causal claims about alcohol’s effects on health which I think go beyond what the observational literature can tell us. You state “the non-drinker can lower his Hazard Ratio for all-cause mortality…by starting the light consumption of alcohol” and “an average non-drinker can significantly lower their risk of call-cause mortality by adding one standard 14 gram serving of ethanol per day”. Until we have an RCT to demonstrate this, I think claims about what alcohol can and cannot do should be made more cautiously. If such a study were performed, and indeed showed benefit, I’d be the first to raise a glass to the results!

Regards,

Rob

References

1: Naimi TS, Brown DW, Brewer RD, Giles WH, Mensah G, Serdula MK, et al. Cardiovascular risk factors and confounders among nondrinking and moderate-drinking U.S. adults. Am J Prev Med. 2005 May;28(4):369–73.

2: Bergmann MM, Rehm J, Klipstein-Grobusch K, Boeing H, Schütze M, Drogan D, et al. The association of pattern of lifetime alcohol use and cause of death in the European Prospective Investigation into Cancer and Nutrition (EPIC) study. Int J Epidemiol. 2013 Dec 1;42(6):1772–90.

3: Stockwell T, Chikritzhs T. Commentary: Another serious challenge to the hypothesis that moderate drinking is good for health? Int J Epidemiol. 2013 Dec 1;42(6):1792–4.

4: Stockwell T, Naimi T. Study raises new doubts regarding the hypothesised health benefits of “moderate” alcohol use. Evid Based Med. 2016 Jul 7;ebmed-2016-110407.

5: Stockwell T, Zhao J, Panwar S, Roemer A, Naimi T, Chikritzhs T. Do “Moderate” Drinkers Have Reduced Mortality Risk? A Systematic Review and Meta-Analysis of Alcohol Consumption and All-Cause Mortality. J Stud Alcohol Drugs. 2016 Mar;77(2):185–98.

6: Holmes MV, Dale CE, Zuccolo L, Silverwood RJ, Guo Y, Ye Z, et al. Association between alcohol and cardiovascular disease: Mendelian randomisation analysis based on individual participant data. BMJ. 2014;349:g4164.

On 2015 Dec 15, David Keller commented:

Trials of Parkinson disease treatments should report effects on dyskinesia along with improvements in "off" time

Patients with Parkinson disease (PD) can be described as being in one of the following 3 symptomatic states at any given time:

1) "Off" - the under-medicated state, when the patient feels the oppressive grip of bradykinesia, and tremors or dystonia may also worsen. Without treatment, the "off" state is the default pathological disease condition in PD.

2) "On, without dyskinesia" - the desired well-medicated near-normal state, when the patient is able to move with a minimum of bradykinesia, tremor or dyskinesia;

3) "On, with Dyskinesia" - the relatively over-medicated state, when the patient experiences disabling and tiring writhing movements.

Levodopa, dopamine agonists, and other PD medications which relieve "off"-state bradykinesia also tend to cause dyskinesias when their effects are excessive.

The current nomenclature in neurology lumps together the desired state of "on, without dyskinesia", with the uncomfortable and disabling state of "on, with dyskinesia". This imprecision causes confusing miscommunication, for the following reasons:

PD patients take medications with the goal of achieving a near-normal "on" state for the maximum amount of time per day. Dyskinesias and the "off" state can be disabling to PD patients to an equivalent degree, and both are detested and unpleasant. PD patients are not benefited by a medication which reduces "off" time if it increases "on" time with dyskinesia by an equal or greater amount.

The net increase in "on" time without dyskinesia is what benefits patients, and is what clinicians want to know concerning a new PD treatment.

Addendum: I have been posed the following question: if dyskinesia occurs in the over-medicated state, why can't the patient simply reduce the amount of medication, taking greater care not to exceed the amount required? The answer is that the "amount required" to emerge from the paralyzing grip of the off state can vary markedly from day to day. In advanced Parkinson disease, the therapeutic window can close completely - the patient is plunged into dyskinesia the moment he breaks free of the off state.

On 2016 Jun 07, David Keller commented:

Reply to the Attack on Off-label Prescribing

The study by Eguale and colleagues ("the authors") demonstrated that off-label prescriptions backed by strong scientific evidence are just as safe as on-label prescriptions. [1] An accompanying editorial, citing their results, called for a "crackdown" against off-label prescriptions, with no exemption for those with strong scientific evidence. [2] My letter to the editor defended scientific off-label prescribing from this blanket prohibition against off-label prescribing. [3] In their reply to my letter, [4] the authors misunderstood or misrepresented my main points, which is best seen by directly comparing quotations from their reply with quotations from my letter. Below, false or misleading statements by the authors are numbered in sequence, and my rebuttals are numbered to match.

1) The authors wrote: "Unscientific prescribing constitutes 4 of 5 off-label uses, and this unscientific prescribing has resulted in a 54% increased risk of adverse drug events compared with on-label uses."

Rebuttal to #1: I never advocated unscientific prescribing. Quite the opposite, I wrote: "We need to crack down on unscientific prescribing" and we should take steps to "discourage unscientific prescribing." The authors provide no evidence or arguments here to counter my advocacy of scientific off-label prescribing.

2) The authors wrote: "To crack down on unscientific prescribing, drug regulatory bodies need to demand strong scientific evidence from pharmaceutical companies to safeguard the public."

Rebuttal to #2: We have agreed repeatedly on the need "to crack down on unscientific prescribing", which in itself implies the need for strong scientific evidence. I dispute the implication that the evidence needs to come from pharmaceutical companies, with their powerful financial biases. I am more impressed by the results of studies which are not funded by the company with so much future profit to gain or lose according to the outcome.

3) The authors wrote: " When physicians prescribe medications for unscientific, off-label uses, their patients suffer adverse drug events."

Rebuttal to #3: Actually, patients always suffer adverse drug events. They meant to say that patients suffer MORE adverse drug events from unscientific off-label prescriptions than from on-label prescriptions. The authors should have added that patients do NOT experience any increase in adverse events from off-label prescriptions backed by strong science than they do from on-label prescribing. Don't these authors recall proving this in their study? Does this constant conflation of "off-label" with "unscientific" indicate confusion on the part of the authors, or are they trying to confuse the readers?

4) The authors wrote: "The expensive and long approval process required by the US Food and Drug Administration (FDA) is exactly what produces the scientific evidence that physicians need to prescribe appropriately."

Rebuttal to #4: False. The authors repeatedly state that "4 out of 5 drugs that are prescribed off-label are prescribed for unscientific use". This means that 20% of off-label prescriptions are for scientific use, and these very authors have proved that such use is just as safe as on-label prescriptions. They are ignoring the results of their own important research.

Consider an inexpensive, generic drug, which is approved for one indication, but is also used widely for another, unapproved indication, with the support of strong scientific evidence. What is to be gained by "cracking down" on doctors who prescribe this drug for the unapproved indication? Not an increase in safety, as these authors have demonstrated in their recent study. And "strong scientific evidence" must, by definition, include evidence of efficacy.

The major effect of forcing generic drugs to be approved by the FDA for unapproved indications that have strong scientific evidence already will be a major increase in the cost of these drugs. Patients who took cheap, generic colchicine safely and effectively for gout for years got a rude shock when it was forced through the FDA approval process. The only difference noticed by patients was the tenfold increase in price they were forced to pay as cheap generic colchicine morphed overnight into an expensive branded-only version, which left many of patients unable to afford to take colchicine any longer.

References

1: Eguale T, Buckeridge DL, Verma A, Winslade NE, Benedetti A, Hanley JA, Tamblyn R. Association of Off-label Drug Use and Adverse Drug Events in an Adult Population. JAMA Intern Med. 2016 Jan;176(1):55-63. doi: 10.1001/jamainternmed.2015.6058. PubMed PMID: 26523731.

2: Good CB, Gellad WF. Off-label Drug Use and Adverse Drug Events: Turning up the Heat on Off-label Prescribing. JAMA Intern Med. 2016 Jan;176(1):63-4. doi: 10.1001/jamainternmed.2015.6068. PubMed PMID: 26524668.

3: Keller DL. In Defense of Off-label Prescribing. JAMA Intern Med.2016;176(6):861. doi:10.1001/jamainternmed.2016.1403.

4: Eguale T, Verma A, Tamblyn R. In Defense of Off-label Prescribing—Reply. JAMA Intern Med. 2016;176(6):861-862. doi:10.1001/jamainternmed.2016.1406.

On 2015 Nov 04, Friedrich Thinnes commented:

For a model trying to explain the inverse relationship of Alzheimer´s Demenz and cancer see:

1) Thinnes FP. Why cancer survivors have a lower risk of Alzheimer disease. Mol Genet Metab. 2012 Nov;107(3):630-1. doi: 10.1016/j.ymgme.2012.06.016. Epub 2012 Jul 4.

2) Thinnes FP. Alzheimer disease controls cancer - concerning the apoptogenic interaction of cell membrane-standing type-1 VDAC and amyloid peptides via GxxxG motifs. Mol Genet Metab. 2012 Aug;106(4):502-3. doi: 10.1016/j.ymgme.2012.06.004. Epub 2012 Jun 15. No abstract available.

On 2016 Apr 02, Daniel Tsin commented:

The first report of human transvaginal cholecystectomy, carried out during a vaginal hysterectomy, was done on August 20, 1999 and published in 2003 by Tsin et al. ( Culdolaparoscopic cholecystectomy during vaginal hysterectomy.Tsin DA, Sequeria RJ, Giannikas GJSLS. 2003 Apr-Jun; 7(2):171-2.) PMID 12856851

On 2016 Apr 08, Tom Kindlon commented:

Main finding: no difference between all 4 groups. Data shows improvement for APT or SMC-only groups not due to extra CBT or GET

Despite how the results have been presented, the main result from this study is that there was no difference at long-term follow-up between the 4 cohorts: cognitive behaviour therapy + specialist medical care (CBT); graded exercise therapy + specialist medical care (GET); adaptive pacing therapy + specialist medical care (APT) or specialist medical care alone (SMC).

I keep seeing it being claimed that the reason the SMC and APT groups caught up with the CBT and GET groups was possibly due to them getting extra CBT or GET after the initial 12 months. What keeps being missed is that we have data to explore that possibility and it is simply not the case. This data may be missed by many as it is in the appendix. The relevant table can be found at this link*: https://twitter.com/TomKindlon/status/718976687384502272

One person has done the calculations which can be seen in this image**: https://twitter.com/TomKindlon/status/718977817044774912

The lack of improvement for those who got extra CBT or GET over those who did not can't be assumed to be due to the fact that some of them had less than 10 sessions of extra CBT or GET as those who had less than 10 sessions improved on more measures than those who had 10+ sessions. See***: https://twitter.com/TomKindlon/status/718978753225953283

*or if you're a member of Phoenix Rising, here: http://forums.phoenixrising.me/index.php?threads/pace-trial-lancet-psychiatry-paper-on-long-term-follow-correspondence.42491/page-2#post-687716

** or if you're a member of Phoenix Rising, here: http://forums.phoenixrising.me/index.php?threads/pace-trial-lancet-psychiatry-paper-on-long-term-follow-correspondence.42491/page-2#post-687705

*** or if you're a member of Phoenix Rising, here: http://forums.phoenixrising.me/index.php?threads/pace-trial-follow-up-heres-the-table-looking-at-the-effects-of-having-cbt-or-get-after-52-weeks.40884/page-2#post-657952

On 2015 Dec 11, Anders von Heijne commented:

Interesting results, but my guess would be that the referrals could be an even better source than the reports for machinebased analysis, if the referrals for radiological exams are of a sufficient informational quality.

On 2017 Apr 12, Siegfried Hekimi commented:

In our recent paper we provide evidence that CLK-1 functions exclusively in ubiquinone biosynthesis and has no obvious role in the nucleus. In a paper entitled: A single biochemical activity underlies the pleiotropy of the aging-related protein CLK-1. By Liu et al. http://www.nature.com/articles/s41598-017-00754-z?WT.feed_name=subjects_energy-metabolism

On 2016 Mar 27, Stephen Strum commented:

I have not completed the entire article but I will say that I find the results to be counter-intuitive and not at all in keeping with what I have seen in over 25 years using IAD (intermittent androgen deprivation) plus another 8 years in using CAD (continuous androgen deprivation). Absolutely, unequivocally, there are many signs & symptoms relating to androgen deprivation (AD) that resolve once testosterone recovery occurs in those using an IAD approach. So no apparent QoL benefits seen in this study has me wondering about how careful the internal medical issues were documented in the medical record.

Moreover, we found major differences in mortality that correlated with the ability to achieve an undetectable PSA defined as ≤ 0.05 during ADT. This was published in Urology in 2007: Prostate Cancer-Specific Survival and Clinical Progression-Free Survival in Men with Prostate Cancer Treated Intermittently with Testosterone Inactivating Pharmaceuticals by Scholz, Lam, Strum et al, & the mean follow-up of patients was ten years.

Men in the current study should have been analyzed to assess who achieved an undetectable PSA on either ADT or IAD & how long was that undetectable status maintained while on ADT. We found that finding to be of major value in ascertaining the prognosis of men receiving ADT. If such subset analysis is not done, then those having an excellent response to ADT can be camouflaged by those men not doing well. Issues of testosterone nadir on ADT are also important to control.

On 2016 Mar 18, ZHONGMING ZHAO commented:

My lab recently moved to the University of Texas Health Science Center at Houston. The database is now available at https://bioinfo.uth.edu/ccmGDB/.

On 2015 Nov 03, Richard Kellermayer commented:

I greatly appreciate the responses by Alessandro Borghi and colleagues to our comments in http://www-ncbi-nlm-nih-gov.ezproxyhost.library.tmc.edu/pubmed/26517976. Unfortunately, I still don’t find the answers to our questions within their responses:

1. Why didn’t they identify the ATP2C1 mutation in the patient of their original report, http://www-ncbi-nlm-nih-gov.ezproxyhost.library.tmc.edu/pubmed/25430969? The genetic analysis would have aided the clinical diagnosis, and may have provided pharmacogenetic insights.

2. Why they propose that HeLa cells are a relevant model for Hailey-Hailey disease (HHD) that exclusively affects the epidermis?

3. Why they failed to reference our pioneering work (http://www-ncbi-nlm-nih-gov.ezproxyhost.library.tmc.edu/pubmed/16143464) on examining the role of magnesium in a yeast model of HHD, which has been shown to be relevant for testing therapeutic agents against the disease? Our publication was the only pertinent reference in PubMed in respect to magnesium and HHD prior to 2015.

On the contrary to Alessandro Borghi and colleagues, a layperson who has family members suffering from the disease kindly called my attention to the findings of Borghi, et al. in October of 2015, by stating the following: “Just wanted you to know that your pioneering research on PMR1 has inadvertently led to a clinical breakthrough in the treatment of Hailey-Hailey disease.”

Irrespective to the questions above, I wish the utmost success to Borghi and colleagues in their future endeavors towards helping the patients suffering from HHD.

Richard Kellermayer, M.D., Ph.D.

On 2017 Mar 02, Prajak Barde commented:

Lucio Gnessi et al.1 concluded that Xyloglucan is a fast, efficacious and safe option for the treatment of acute diarrhea. The conclusion was based on statistically significant difference in diarrheal symptoms with Xyloglucan in comparison with diosmectite and Saccharomyces in 150 adult patients. However in absence of clear prior hypothesis, it is difficult to understand whether the study sample size was adequate to conclude efficacy of Xyloglucan2. Author neither commented on the prevalence of diarrheal symptoms in control group, nor pre-defined expected difference in prevalence [effect size (d)] between Xyloglucan and control.

In addition, author did not discuss the magnitude of effect of Xyloglucan and other two treatment modalities. There is understandable variability in baseline data with mean number of dehydrating stool higher in S. Bouliardii, therefore concluding on the basis of direct comparison post treatment effects would be inappropriate. Though, the result are encouraging with clear efficacy of Xyloglucan, presenting mean changes from baseline would have been appropriate to demonstrate the effect of individual drugs, to substantiate the results of the study and to build a future hypothesis. In absence of this information, it looks like a pilot study.

On 2016 Oct 24, Sean Ekins commented:

Here are the slides from an ACS talk in 2015 http://www.slideshare.net/ekinssean/combining-metabolitebased-pharmacophores-with-bayesian-machine-learning-models-for-mycobacterium-tuberculosis-drug-discovery

On 2016 Sep 12, Massimiliano Tognolini commented:

2,5-dimethylpyrrolyl benzoic acid (compound 1) is a promiscuous and reactive chemical substance able to interfere with a multitude of biological targets. MOreover: - “… newly synthesized compound 1 did not show detectable inhibition of ephrin—A5 binding to EphA4 in ELISAs or EphA2 phosphorylation in cells stimulated with ephrin-A1…” when tested immediately after its chemical synthesis as white and crystalline powder. Moreover, the same Authors further explained these findings stating that “…when left exposed to air at room temperature in dry form, compound 1 acquired a progressively darker brown color. Concomitantly, the compound became progressively more active in ELISAs measuring the inhibition of ephrin-A5-EphA4 binding” (Noberini R et al Chem. Biol. Drug Des.2011; 78, 667–678). - Compound 1, spontaneously forms a mixture of reactive polymers, with an average molecular weight of 40kDa, likely responsible for the aspecific effect of compound (Zhu W et al. Chem. - Eur. J. 2013; 19, 8397–8400).

On 2017 Oct 05, ROBERT HURST commented:

Unfortunately, the KU-7 line is really HeLa cells instead of bladder cancer PMC3805942

On 2015 Nov 12, RUBEN ABAGYAN commented:

Additional disclosure. One of the co-authors, R.A., has an equity interest in Molsoft, LLC. The terms of this arrangement have been reviewed and approved by the University of California, San Diego in accordance with its conflict of interest policies.

On 2016 Aug 10, Ahmed Al-Shammari commented:

This work spot the light on the one health concept where we can see the effect of pollution on Human, animlas and plant health

On 2015 Oct 30, KEVIN BLACK commented:

We have begun an online draft of a 2015 update of this article, and would be grateful for any constructive feedback. The platform allows readers to leave comments, using the little balloon icon at the upper right corner of each section of the article. For further explanation, see the comments below the 2014 highlights article.

On 2016 Aug 02, Sudheendra Rao commented:

Hi, there has been a lot of discussion on this paper on nature comments sections but I am surprised to see that authors have not responded to any of them. I really hope that the authors take some time out and respond either here or at Nature Comments section.

On 2017 Aug 05, Seán Turner commented:

Family name 'Egibaceraceae' (sic) in the title should be 'Egibacteraceae'.

On 2016 Jan 29, Daniel Corcos commented:

Welch et al. draw unwarranted conclusions from the incidence curves they show. In their paper, there is no information about the incidence of prostate cancer before initiation of widespread screening. However, as PSA screening is more likely to detect metastatic cancer, since high levels of PSA are associated with metastasis (1), it is very likely that screening leads to early detection of metastatic cancer, with the expected consequence of a peak of detected incidence at initiation of screening, followed by a return to basal level. In contrast mammography screening detects only primary cancers, with no bias for metastatic cancer. The incidence of metastatic cancer should be related to the incidence of primary cancer, which has risen during the same period (2). If breast cancers were metastatic at the time of detection as they claim, there would be no benefit from early surgery (3).

I conclude that there is no difference in cancer dynamics that could be inferred from these incidence curves, and that early detection of breast cancer is still the best option in order to shorten time to surgery. However, a 30% increase of primary breast cancer after implementation of x-rays screening is worrying and suggests that x-ray screening may be carcinogenic, at least in women with hereditary DNA repair defect (4). A retrospective study comparing the percentage of DNA repair defects in metastatic breast cancer before and after initiation of widespread mammography would be useful.

1) http://www.ncbi.nlm.nih.gov/pubmed/26659430 2) http://seer.cancer.gov/ 3) http://www.ncbi.nlm.nih.gov/pubmed/15126770 4) http://www.ncbi.nlm.nih.gov/pubmed/22788243

On 2016 May 17, Lydia Maniatis commented:

(My two PubPeer comments)

Here's the funny thing about vision science these days.

We are told about a prolific area of research - 'visual averaging': “New research tools and imaging techniques coupled with solid psychophysical work have added substantially to the large base of work done in the 20th century.”

Then we learn (caps mine) that: “Some lively debates regarding the scope of the putative sampler/averager and EVEN ITS EXISTENCE have sharpened the questions being asked and reminded researchers to consider scaling issues, experimenter and observer bias, and the multidimensional nature of stimuli and ensembles.”

It is the nature of a dogma and the school that collects around it to act on the basis of assumptions that are never challenged. Research projects are designed so that they cannot challenge the dogma, but only answer questions in its terms. (That something like this is going on here is evidenced by the author's concession that claims of possible non-existence of “averagers” remain viable despite decades of research activity.) It's the nature of science to test its fundamental assumptions before proceeding to elaborate on them. If a claim is falsifiable, in a healthy research environment it will be quickly falsified. (If it cannot be tested even in principle, then it's outside the game of science and cannot constitute a legitimate field of empirical investigation). In the type of environment we have now, it will remain in good standing for the foreseeable future. (This is the case, for example, with the bizarre notion of “spatial frequency filters.”)

On some of the absurdities coming out of “averaging” proponents, see comments here: https://pubpeer.com/publications/90941136CC181AFE4896477BF5BB44 and here:

https://pubpeer.com/publications/26067519

The faith-based tendency to accumulate evidence (supposedly) in favour of the dogma is related to this field's (invalid) adoption of inductive procedures rather than hypothesis-testing. If we just keep measuring things and fitting algorithms to the data, the truth will naturally emerge:

"The psychophysical function and the environmental information on which it is based should be allowed to emerge without predilection, lest the true mental algorithm be obscured by expectation (Anderson, 1968; Levin, 1975).... It is well known that psychological measurement is not impervious to the effects of context, scaling, measurement error, and other issues that can hinder the revelation of the true psychophysical relationship." Expectations, aka hypotheses, should stay well away lest they interfere with revelation.

Despite decades of patient effort and mountains of "evidence," we're still waiting for the true psychophysical function to be revealed. What if it doesn't exist? What if it can never be spotted amongst the confounds, known and unknown?

As we wait, we might ask what it means that "perceptual averages" have never been perceived? In what sense is an unperceived percept perceptible? (If averages were perceptible, there would be no question of their existence; we would be trying to explain perceptual facts rather than waiting for these facts to be revealed.)

On 2015 Dec 15, Martine Crasnier-Mednansky commented:

This article is remarkable for demonstrating an interaction between unphosphorylated Enzyme IIA^Glc and the EAL domain of CsrD. Such interaction is physiologically connected to the role of cAMP in E. coli because there is an established correlation between the phosphorylation state of Enzyme IIA^Glc and the level of cAMP, phosphorylation of Enzyme IIA^Glc typically causing an increase in the cAMP level (and concomitantly eliminating the effect of Enzyme IIA^Glc on CsrD).

The authors discussed the present regulatory interaction with no mention to the role of cAMP. However, considering (1) the inhibitory function of (CsrD-controlled) CsrA on glycogen biosynthesis, (2) the ability of stationary-phase E. coli to accumulate glycogen as a carbon reserve, and (3) the increase in cAMP occurring upon entry into the stationary phase (because of Enzyme IIA^Glc phosphorylation) (MAKMAN RS, 1965), it seems CsrD and CRP-cAMP both work 'in concert' for the timeliness of physiological processes during entry into stationary phase, particularly when cells are growing on glucose.

On 2016 Oct 24, Gwinyai Masukume commented:

It is unclear if indeed there has been a rapid rise in the sex ratio at birth because the data used to derive the estimates was more than 50% incomplete. In Nigeria civil registration of births is about 30% World Health Organisation - Global Health Observatory, UNICEF.

On 2015 Oct 28, Riccardo Percudani commented:

The authors suggest that the increased HGPRT activity in bottlenose dolphins after diving could maintain ATP concentrations through "ATP synthesis from IMP" (abstract). Could they provide information about the biochemical pathway involved? No pathways for ATP synthesis from IMP are reported in KEGG, and the authors do not cite a specific reference.

On 2015 Nov 20, Lydia Maniatis commented:

This paper has all the hallmarks of the “Bayesian program” - a predictable prediction confirmed, draped over a sketchy, ad hoc (I will dare to say unintelligible) mathematical model structured around a normal distribution, which model is then said to have “influenced” the results (as opposed to simply being (roughly and post hoc) correlated with them) – spiced, here, by the improper discarding of inconvenient datasets.

To begin with the last point, in Experiment 1, two out of fifteen subjects' datasets were “excluded from analysis.” The brief rationale “because they performed below 55% correct” might mislead readers into assuming it means that, for the purposes of the experiment, subjects were performing at chance level. However, although the task did involve a forced choice between presence or absence of a contour, the question at issue was whether certain features of contours make them more or less detectable. This question could have been answered using this data set – out of the 55% correct, did more of the detected contours contain the features that the authors predicted would lead to more reliable detection? Given the weak and noisy effects, it's possible that these data would have undermined the story; their removal is, at any rate, highly improper.

Based on the data that were used we are told, as part of the “basic analysis,” that “subjects were significantly better at detecting leaf shapes than animal shapes.” So what? There was no prediction, no experimental question, regarding the difference in the detectability of leaf vs animal shapes. And while this is the first result reported, the authors make no attempt to explain it.

Let me take a stab at it. Looking at their sample forms, I notice that the leaf shapes are fronto-parallel and symmetrical (or very nearly so), while the animal forms are not – some are side views, some three-quarters views, but none except a fish shape are bilaterally symmetrical. The symmetry of the fish is around the horizontal (and thus tends not to be salient for human viewers). This very clear and obvious shape factor is very pertinent to the authors' discussion about local vs global shape properties; but it is not at any point discussed. Despite these suggestive results, the authors do not test or refer to symmetry in their artificial shape condition. (Symmetry is a confound in their Experiment 2, because their least “complex” shape is also the one that is bilaterally symmetrical.)

Fortunately, this oversight doesn't prevent the authors from detecting the well-known fact that detection of shapes (i.e. perceptual organization of the visual field) depends on global properties, and that more “complex” (i.e. more discontinuous) shapes are less salient than less complex ones. (Because the shapes are presented in a highly unnatural setting (a field of noise) it is not clear what these particular results imply for more typical conditions. The authors are aware of this, and describe their results as showing that “subjects' detection of closed contours in noise is impaired, slightly but reliably...”)

Despite the prominence of the term, it is not clear that the authors have tested the role of “closed contours,” because they have not included an “open-contour” condition. And, in fact, they don't actually claim to have specifically tested the role of complexity in closed contours: “Our results show that when observers seek a closed contour amid noise, they are influenced by the complexity of the bounded shape.” Given their experimental parameters, there is no difference between this statement and the statement: “Our results show that when observers seek a closed contour amid noise, they are influenced by the complexity of the CONTOUR.” The role of contour complexity was the subject of another recent paper by these authors (http://jov.arvojournals.org/article.aspx?articleid=2291654). As far as the analysis goes, the data offered in the present paper would not seem to allow a differentiation between closed and open contours. The paper is thus effectively a repeat of the previous one.

The conclusions are trivial in the sense that they corroborate one of the most well-corroborated claims in visual perception: “Our results corroborate the Gestalt view that a complete understanding of contours is not, in and of itself, sufficient to understand whole shapes [I would note that a line (a long skinny object) also has a shape].” So what does the study offer? “The quantification of the complexity of whole shapes...” Given the unnatural task difficulty, noisy data, small effect sizes, unexplained results, it's not likely that the present effort has any value in this direction. The authors don't dare say that it does - concluding (very) modestly that: “...we hope that more attention to the problem of representing whole shapes, and specifically to skeletal representations, will spur the development of a more comprehensive account.” These hopes are not supported by this study.

Finally, although the authors pay lip service to “global shape” their models are fundamentally local “and-sum” approaches; their “features” are not truly global features, such as symmetry or compactness.

References that should have been included:

On shape skeletons:

1. Arnheim, Rudolf. Art and visual perception: A psychology of the creative eye. Univ of California Press, 1954.

2. 2.Firestone, Chaz, and Brian J. Scholl. "“Please Tap the Shape, Anywhere You Like” Shape Skeletons in Human Vision Revealed by an Exceedingly Simple Measure." Psychological science (2014): 0956797613507584.

On quantification of global shape features:

Articles by Pizlo and colleagues presenting computational models of shape.

On 2015 Dec 01, Lydia Maniatis commented:

Very brief summary: The central claim about the discovery of a perceptual difference in the effect on a target of a narrow gradient versus a wide gradient when both are invisible has not been corroborated because:

a. the wide gradient is visible;

b. the effect of the narrow gradient is the very same as that which would occur in the absence of this gradient. The attempt to spin it differently is based on past experiments in different conditions using narrow but visible, not invisible, surrounds.

On 2015 Nov 30, Lydia Maniatis commented:

his article claims to introduce a new illusion whose novelty and theoretical interest hinge on the presence of invisible luminance gradients - one shallow, one steep. The problem, as any reader can confirm for themselves, is that the shallow gradient is quite visible.

The discussion section contains an admission of the inaccuracy of the invisibility claim and an attempt to downplay the relevance of this failure to control for visibility: “A problem...is how to interpret the contrast effect obtained with the display pair C-D, in which the luminance gradients were barely visible...We ...conclude that the validity of a high-level interpretation of our results based on illumination cues does not hinge critically on whether the luminance gradients in our study were actually invisible or “almost” invisible.”

Contrast the above statements with others such as that “The experimental results clearly indicate that luminance gradients do not need to be visible in order to influence the lightness of embedded surfaces....the effect produced by pair C-D is not surprising. This effect is just an invisible gradient version of the effect demonstrated earlier by Agostini and Galmonte (1997, 2002);” “the luminance range of gradual transitions has been reduced to make them “invisible” and “unnoticeable;” “we manipulated the width of invisible luminance gradients.” The difference between invisible and barely visible is that the latter is visible, and the former invisible.

Given that the main empirical claim (from the summary) is that “the width of an invisible luminance gradient determines the lightness of a target that it surrounds...wide invisible gradients generate contrast effects, while narrow invisible gradients generate assimilation effects,” the visibility of the shallow gradient means that visibility/invisibility was, in fact, an uncontrolled confound in these experiments, and that the central claim has not been corroborated.

The meaningfulness of the “assimilation” description of the invisible narrow gradient is also called into question when the authors acknowledge that b/c the gradient was so narrow, “one might argue that the outer background could have had a larger effect on the disk appearance than the local surround.” They attempt to deflect this possibility by citing previous experiments in which a visible, narrow local surround produced assimilation. In this comparison between experiments, just as in the comparisons within this experiment, the visibility/invisibility factor constitutes a confound that undermines any conclusions as to a distinction between wide and narrow gradients.

But wasn't subjects' ability to perceive the shallow gradient an empirical question? The procedure used actually did not allow this key question to be settled, something which would not have been particularly difficult. Observers were simply asked to describe the displays. When they did not spontaneously report on the differences between the targets, they were directed by the experimenter to do so. Because they did not spontaneously report on the smoky or cloudy effect of the shallow gradients, the authors concluded (or at least state) that “none of the observers noticed the gradual luminance gradient.” But in this case, why do the authors themselves later describe the gradients as “barely visible”? They were obviously aware that a failure to spontaneously report on some aspect of the stimuli did not necessarily indicate a failure to perceive, and they took measures to fill in some relevant gaps. Since the shallow gradients are, in fact, visible, treating the failure to report as failure to perceive is transparently poor practice.

The discussion is conceptually confused, ad hoc and fully-hedged, but it does not seem worthwhile for the present purposes to analyze the various flawed attempts at interpretation of a confounded and biased dataset.

Finally, it should be noted that the authors refer to other “notable approaches” to lightness “including Gilchrist's Anchoring theory...critiqued by Rudd (2010, 2013, 2014)...” It would have been relevant here to mention that a paradigmatic claim of “anchoring theory” has been straightforwardly and definitively falsified (Maniatis, Lydia M. "A simple test of the “anchoring account” of simultaneous contrast." Journal of Vision 15.5 (2015): 13-13).

On 2015 Oct 28, Ziguo Zhang commented:

Human anaphase-promoting complex (APC/C) is a multimeric (14 different proteins, 19 subunits, functioning together with one of its co-activators, Cdh1 or Cdc20) RING E3 ubiquitin ligase that controls chromosome segregation and mitotic exit.

On 2017 Sep 06, Tao Liu commented:

The webserver has moved to http://ccb1.bmi.ac.cn:81/srnatarbase/

On 2015 Oct 31, Christopher Southan commented:

Useful paper - this post picks up the PubChem structure mappings http://cdsouthan.blogspot.se/2015/10/a-kinase-gift-horse.html

On 2015 Nov 14, Arnaud Chiolero MD PhD commented:

What is missing is the expected absolute reduction in breast cancer mortality from mammography screening. It would help women make an informed decision.

On 2017 Jan 02, Martin Mayer commented:

The new online format of JAMA Internal Medicine no longer shows the comments that were submitted and posted for this article (one from Peter Doshi and Tom Jefferson, and one from me). The comments to which I refer are not the same as the above-noted comments that were published as Letters to the Editor (though Doshi and Jefferson did also publish a Letter to the Editor). The comments to which I refer are available here for interested readers.

On 2017 May 16, Fang-Cheng Yeh commented:

Erratum at Fig.7a's legend:

"Low" FDR leads to more specific results but may miss true findings, while "high" FDR is more sensitive but may include false positive findings.

On 2016 Feb 07, Fang-Cheng Yeh commented:

The tool and source code for connectometry are available at http://dsi-studio.labsolver.org

On 2016 Apr 14, RODERIC GUIGO commented:

The literature cite by Eduard is relevant to the topic of the relationship between histone modifications and splicing. These works, as well as others, show that there is a weak (and sometimes controversial) effect of histone modifications and exon inclusion. This weak effect could reflect a general weak effect on most exons, or a stronger effect on a smaller set of exons. The main contribution of Curado et al. (2015) is the identification of a small set of exons in which histone modifications seem to be more strongly associated with inclusion levels, and that this association seems to be mediated by proximity (linear or on tri-dimensional space) to promoter regions. The findings in Gonzalez-Vallines et al. (2015) suggest the possibility that our promoter-like exons have enhancer-like characteristicis.

On 2016 Mar 23, Eduardo Eyraa commented:

The authors failed to cite prior relevant literature related to the associations between chromatin signals and alternative splicing: Zhou Y, 2012, Shindo Y, 2013, Ye Z, 2014, Agirre E, 2015, González-Vallinas J, 2015, among others. Some of these articles already proposed specific predictive models of splicing regulation based on chromatin signals that can explain a considerable number of events. For instance, Agirre E, 2015 already showed that nearly 70% of differentially spliced events between two cell lines can be explained by a model based on the differential enrichment of chromatin signals. The proposed model included the binding activity of CTCF, which, incidentally, is also used in the model by Curado et al. On the other hand, Curado et al. describe their promoter-like exons in the following way: "...while no promoters themselves, they are, on average, closer to TSS and enriched by ChIA-PET tags associated with RNA Polymerase II". In one of the prior publications mentioned above, González-Vallinas J, 2015, the authors describe a model in which potential intragenic transcriptional enhancers, upon activation or deactivation, would change the local properties of the chromatin, thereby affecting the RNA processing of the host gene, and in particular, its alternative splicing. These potential intragenic enhancers are also shown to have enhancer-like properties, including the presence of ChIA-PET links with nearby promoters. Additionally, they produce eRNAs, they can occur on exons as well as on introns, and associate with differential splicing of nearby exons. It is unfortunate that the authors did not mention these prior results in their article, given the relevance they have for what they are discussing. If you are considering citing this publication, please consider whether you should also cite Agirre E, 2015 or González-Vallinas J, 2015.

On 2016 Jun 25, Donald Forsdyke commented:

HISTORY OF LECTIN PATHWAY RESEARCH

The history of this field has recently been reviewed (1,2).

1 Forsdyke DR (2016) Microbes & Infection 18, 450-459. Almroth Wright, opsonins, innate immunity and the lectin pathway of complement activation: a historical perspective. Forsdyke DR, 2016

2 Sims RB, Schwaeble W, Fujita T (2016) Complement research in the 18th-21st centuries: progress comes with new technology. Immunobiology 221:1037-1045 Sim RB, 2016

On 2015 Nov 16, Donald Forsdyke commented:

COMPLEMENT ACTIVATION BY PLANT AND ANIMAL LECTINS - THE LECTIN PATHWAY

The “proof-of-principle” that “lectins can be made more selective through molecular engineering” was demonstrated with the plant mannose-binding lectin Concanavalin-A in the 1970s (1). The functions then studied were mitogenicity, complement activation, and serum-binding activity. The dimeric form was found to retain mitogenicity without complement activation. The tetrameric form retained both activities (2).

The discovery that plant lectins activate the mammalian complement system (1) was initially attributed to a single antibody-independent “alternative” pathway of complement activation (3). However, subsequently the observation was extended to animal mannose-binding lectins and a second “alternative” pathways was characterized (4). This was named the “lectin pathway” and the other retained the “alternative pathway” name. Thus, there are three complement activation pathways – the classical, the lectin and the alternative.

In this light there are caveats regarding the proposed topical application of lectins to prevent HIV infection.

(i) There is suggestive evidence that lectins can cross mucous membranes, thus accessing body fluids and exerting toxic, perhaps complement-mediated, effects (5). While the elegant studies of Swanson et al. (6) show how mitogenic effects might be avoided, the possibility of toxic effects has not been excluded.

(ii) Reaction of soluble lectins with targets can be extensively buffered by competing lectin-binding activities, both associated with cell surfaces and in body fluids. When HIV buds from infected cells its envelope includes normal cell surface components that bind lectin. Furthermore, when plant lectins activate lymphocytes cultured in serum-containing medium, doubling the serum concentration doubles the lectin requirement (7). Thus, very high lectin concentrations may be needed for in vivo microbicidal effects. Is in vitro BanLec mitogenicity dependent on the BanLec/serum ratio? If so, is molecular engineering able to decrease competitive binding by serum and natural secretions?

(iii) While avoidance of initial HIV infection is important, use of lectins has not been thought promising in this respect. Hopefully the lectin work is not diverting resources from studies of the “shock-and-kill” approach that promises complete HIV eradication (8)?

(1) Milthorp P, Forsdyke DR (1970) Inhibition of lymphocyte activation at high ratios of concanavalin A to serum depends on complement. Nature 227:1351-1352. Milthorp P, 1970

(2) Forsdyke DR (1977) Role of receptor aggregation in complement-dependent inhibition of lymphocytes by high concentrations of concanavalin-A. Nature 267:358-360. Forsdyke DR, 1977

(3) Eidinger D, Gery I, Elleman C (1977) The inhibition of murine lymphocyte mitotic responses by human and mouse sera. 1. Evidence for a role of antibody-independent activation of the alternative complement pathway. Cellular Immunology 30:82-91. Eidinger D, 1977

(4) Degn SE, et al. (2014) Complement activation by ligand-driven juxtaposition of discrete pattern recognition complexes. Proc. Natl. Acad. Sci. USA 111:13445-13450. Degn SE, 2014

(5) Forsdyke DR (1978) Role of complement in the toxicity of dietary legumes. Medical Hypothesis 4:97-100. Forsdyke DR, 1978

(6) Swanson MD, et al. (2015) Engineering a therapeutic lectin by uncoupling mitogenicity from antiviral activity. Cell 163:746-758. Swanson MD, 2015

(7) Forsdyke DR (1967) Quantitative nucleic acid changes during PHA-induced lymphocyte transformation in vitro: dependence of the response on the PHA/serum ratio. Biochemical Journal 105:679-684. Forsdyke DR, 1967

(8) Forsdyke DR (1991) Programmed activation of T-lymphocytes. A theoretical basis for short term treatment of AIDS with azidothymidine. Medical Hypothesis 34:24-27. Forsdyke DR, 1991

On 2017 Jun 28, Randi Pechacek commented:

P Ji, the first author of this paper, wrote a beautiful blog on microBEnet giving some background information.

On 2016 Dec 15, Victoria MacBean commented:

Plain English summary:

Opioids are a class of drug which act by attaching to opioid receptors, found in the brain and spinal cord, reducing the perception of pain. For this reason, opioids are often prescribed for pain relief. When people misuse opioids, they are often unaware of the dangerous side effects that come with them. For example, they are respiratory depressants, meaning they can reduce the breathing rate, which can be fatal. Many of the current methods of measuring respiratory depression under-estimate the true effect these drugs have on the body (especially the breathing rate), which is why this study was undertaken. Respiratory depression is a major cause of overdose and if you cannot detect when it is happening effectively, you have less chance of helping someone suffering from it.

The participants in this study were monitored over a course of 150 minutes, after they had been given their usual opioid dose. This was done using EMGpara (a tool which assesses how hard the breathing muscles are working), pulse oximetry (measuring the blood’s oxygen levels), and measurement of carbon dioxide levels in exhaled breath. The participants were asked to rate how much they felt the drug’s effect at three minutes prior to the drug being given, and then at regular intervals afterwards. Staff ratings of intoxication and level of consciousness were also given. Pulse oximetry and observer ratings are the more commonly used methods of observing patients’ breathing currently.

However, this study found that there was an increase in the level of carbon dioxide per breath in eight of the ten participants and a low blood oxygen level in only four out of the ten patients. The difference in results shows that the traditional approach of measuring the blood’s oxygen level is not as sensitive a method to detect respiratory depression after taking an opioid. There were varying degrees of respiratory depression found in all patients. However, the pulse oximetry only picked up four of these. The study also found that just talking to a patient helped to mask episodes where they were breathing unusually slowly. This means that it is very easy to miss a potentially dangerous symptom.

The findings of this study therefore suggest that we should change the way we test for respiratory depression in clinical settings, to help identify, treat and prevent it in patients taking opioids.

This summary was produced by Lily Groom, Year 13 student from Graveney School, London as part of the investigators' departmental outreach programme.

On 2015 Dec 16, Lawrence Kane commented:

I took a great interest in this paper, given the ongoing focus of my lab on the mechanisms of Tim-3 signaling and its implications for T cell and mast cell activation. I was encouraged to see that the authors also observed association of Tim-3 with one or more tyrosine kinases in T cells, consistent with our past data Lee J, 2011. However, by contrast with our previous study in Molecular and Cellular Biology Lee J, 2011, they reach a very different set of conclusions regarding the functional effects of ectopic Tim-3 expression. While we consistently observe(d) enhancement of T cell activation after ectopic expression of Tim-3, via either transient transfection or tet induction of a stable cell line, the authors of this study observe inhibition of T cell activation.

What prompted me to write this comment, however, was the fact that the authors were not completely accurate in their representation of our previous data and why it might conflict with theirs. The authors suggest that we used primarily murine Tim-3, expressed in human T cells; however, we have performed many experiments with murine Tim-3 in murine T cells (e.g. Fig. 2D in Lee J, 2011), as well as some experiments with human Tim-3 in human T cells (Fig. 1C in ref. Lee J, 2011). In addition, we had initially generated T cell lines with stable ectopic expression of Tim-3, but these cells invariably lost expression of Tim-3 over time, despite the maintenance of drug selection (unpublished data). This is why we derived the tet-inducible cells used in the latter part of the study (ref. Lee J, 2011).

Thus, in my opinion, it remains to be seen why conflicting conclusions have been reached by different groups regarding the intrinsic effects of Tim-3 on T cell activation and TCR signaling. This is important, given the rapid push to translate findings with immune checkpoint receptors like Tim-3 Mahoney KM, 2015.

On 2016 Feb 05, Daniel Schwartz commented:

As the abstract mentions, the IPI24 can be calculated online or used in a mobile app:

https://qxmd.com/calculate/diffuse-large-b-cell-lymphoma-prognosis-ipi24

On 2016 Dec 15, Victoria MacBean commented:

Plain English summary:

Patients with Sickle Cell Disease (SCD) are often assumed to have asthma because they have 'airflow obstruction', which is when airways become narrowed and air is not able to move out of the lungs as quickly or easily as in healthy lungs.

Inflammation in the airways is one of the main features of asthma. Nitric Oxide (NO) is a substance that is produced by the airways when they are inflamed, so therefore can be used to measure the severity of asthma in a patient and find out how inflamed their airways are. As well as being produced in the airways, NO is also produced in blood vessels, and helps to widen blood vessels.

People with SCD are anaemic, meaning they have less haemoglobin (a protein in the body that carries oxygen) so their cells cannot carry as much oxygen. In order to compensate for this, the heart beats faster and with more power to make sure enough oxygen is picked up from the lungs and delivered to the body.

This study looked at measurements of NO from the airways (representative of asthma) and the alveoli (representative of blood vessel widening), and compared it to lung function tests (to look at airway narrowing) and measures of pulmonary blood flow (how fast blood was circulating around the lungs).

The results showed that the airway NO was not raised, but that there was still airway narrowing occurring. There was a relationship between how fast blood was circulating through the lungs and NO from the alveoli. This might suggest that previous findings of high NO and airway narrowing resulted in a false assumption that SCD patients' airway narrowing was down to asthma. This study suggests that the changes in heart and blood vessel function in SCD may have an effect on the airways.

This study was relevant in terms of contributing to medical research because it shows the airway narrowing in SCD patients may not always be down to asthma, so it therefore allows us to target other root causes of the problem.

This summary was produced by Ashleigh Francis, Year 13 student from Harris City Academy Crystal Palace, London, as part of the investigators' educational outreach programme.

On 2016 Jan 14, Brandon George commented:

When preparing the data and analysis code for public release, we noticed minor errors in the values reported in the paper. The errors and corrected values are detailed below, which do not substantially alter the conclusions. We apologize for the discrepancies and note that the data from the study and code for the appropriate analyses is available at ICPSR (https://www.openicpsr.org/repoEntity/show/53793?versionId=53793V1).

-In Table 2, the Mean(SD) for peak power in the HIIT group should be 957.7(209.7), and the p-value for the baseline difference between intervention groups should be 0.3875. This was due to a programming error.

-In Table 3, the p-value for the test of an overall change over time in QUICKI (complete case analysis) should be 0.3670, not 0.0254. This was due to a transcription error. Although the nominal significance of QUICKI changed, the overall pre-post change of insulin sensitivity (Si) was still significant and as Si is considered a more sensitivity measure of glucose tolerance and is the basis of our conclusion that MIT or HIIT intervention may improve glucose tolerance.

-In Table 3, the ITT analysis was done using a slightly different definition of subject compliance than was used in the complete case analysis, where subjects’ proportion of completed workouts was used as a covariate. The change in the p-values of the ITT analyses using the same covariate is small and there was no change in nominal significance. The exact values can be calculated using the available code and data at ICPSR.

On 2015 Nov 24, Lydia Maniatis commented:

Pullout quote: "As long as authors are (mostly) rewarded for publishing many articles and editors are (mostly) rewarded for publishing them rapidly, new ways of gaming the traditional publication models will be invented more quickly than new control measures can be put in place."

Everyone knows it, but what can/should be done about it?

On 2015 Nov 17, ANDRES TIRADO-SANCHEZ commented:

Peer-review fraud as a “new” type of misconduct. No politics, no committees, no reports, no referees, no interviews – just highly motivated people picked by a few men of good judgement. Sir James Black. When I read the perspective published by Haug,1 I found an alarming issue with regard to the review process of scientific articles, but a déjà vu. For more than three decades, 2 it has been said that peer-review is not an objective method for evaluating scientific work, as it lends itself to many abuses by authors, reviewers and also the editors (mainly guess-editors). The acceptance or rejection of an article is mainly based on the reviewers’ comments, and the final verdict according the editor´s criteria; this is as irresponsible as peer-reviewing the work of colleagues or their own. On the other hand, peer-review is a useful tool for irrelevant articles, but it is unlikely to detect plagiarism or misconduct, which represents a major weakness; furthermore, it also generates favorable or unfavorable criticism of an article by reviewers´ conflicts of interest. This is not a problem for a country or region, a specific journal or editors, but the entire scientific community; it´s not about the money earned or fast track publications, it´s all about ethics. The process of submitting, receiving, reviewing, editing, and acceptance of an article should be amended to impose increasingly stringent filters, in order to prevent this kind of misconduct. References. Haug CJ. Peer-review fraud – Hacking the Scientific Publication Process. N Engl J Med 2015 Oct 21. Buchele JP. The new malpractice. Peer review. Kans Med 1986; 87(9): 233-6.

On 2015 Nov 02, George McNamara commented:

Could the authors please add scale bars to their light microscope image data?

With respect to figure 3, "dendrites of the NK cells penetrating into the target cells", the word penetrating implies piercing the target cell plasma membrane. SEM cannot show this. Even multicolor fluorescence microscopy would not show this. I suggest Transmission electron microscopy and am surprised the reviewer did not ask for this and the authors did not volunteer it.

Morphology and mechanism by which real Natural Killer cells kill is by immune synapse formation and directed secretion of perforin and granzymes -- not "piercing" target cells with sword like filopodia. As the authors point out in their discussion, the induced cell type may not be either a dendritic cell or a natural killer cell, so these "AMLiK" (my suggested acronym) do not need to follow normal cell's playbooks.

On 2015 Nov 03, Mihir Shah commented:

This paper delves into a hitherto non-highlighted concept on VLC and LED technologies. It is also interesting because the authors span 3 countries, Pakistan, India and the United Kingdom. A commendable paper by Shoaib Muhammad and team.

On 2016 Apr 05, Marko Premzl commented:

The third party data gene data set of eutherian lysozyme genes HG931734-HG931849 was deposited in European Nucleotide Archive under research project "Comparative genomic analysis of eutherian genes" (https://www.ebi.ac.uk/ena/data/view/HG931734-HG931849). The 116 complete coding sequences were curated using tests of reliability of eutherian public genomic sequences included in eutherian comparative genomic analysis protocol including gene annotations, phylogenetic analysis and protein molecular evolution analysis (RRID:SCR_014401).

Project leader: Marko Premzl PhD, ANU Alumni, 4 Kninski trg Sq., Zagreb, Croatia

E-mail address: Marko.Premzl@alumni.anu.edu.au

Internet: https://www.ncbi.nlm.nih.gov/myncbi/mpremzl/cv/130205/

On 2016 Dec 18, Laura Williams commented:

Our online journal club discussed this paper on November 10, 2015. https://www.youtube.com/watch?v=sMfHZ1zdXvk

On 2014 Dec 10, Kath Wright commented:

Other search filters are available from the InterTASC Information Specialists' Sub-Group Search Filter Resource at https://sites.google.com/a/york.ac.uk/issg-search-filters-resource/home

On 2015 Oct 21, GEANNCARLO LUGO commented:

Should you not have access to this publication, please feel free to contact me at lugo@ipbs.fr

On 2015 Oct 21, Jan Egger commented:

A video demonstrating the interactive RFA Volumetry can be found here: https://www.youtube.com/watch?v=GcEwwY2rG-c

For further information please visit: http://www.clinicimppact.eu/

On 2015 Dec 03, Benedict Nwomeh commented:

In pediatric surgery, as in all aspects of medicine, teamwork is critical to achieving the best results for kids and their families. I'm excited to be a part of #IGSJC and I invite everyone - surgeons, non-surgeons, non-physicians - to join the tweetchat on Dec 14 & 15 at 5-9pm EST.

On 2015 Dec 03, Benedict Nwomeh commented:

None

On 2015 Dec 03, Christian Jones commented:

Dr. Gusani (@nirajgusani on Twitter) is certain to be an outstanding moderator, and I'm really excited about this IGSJC session. We're only starting to realize how teamwork matters in surgery, and it's great to see such concepts being formally studied. Kudos to Dr. Greenberg and her entire team.

On 2015 Dec 03, Niraj Gusani commented:

I am delighted to moderate the International General Surgery Journal Club (#IGSJC) Twitter discussion of this manuscript. We hope to have a robust discussion about the effect of surgeon leadership styles and their effects. We are honored to be joined during the chats by several of the authors, including Drs. Sarah Parker, Caprice Greenberg, and Steven Yule. Join us by following the hashtag #IGSJC during the discussion - Dec 14th & 15th, 5-9pmET for live discussion with the paper open for comments/discussion throughout both days.

On 2015 Dec 03, IGSJC - The International General Surgery Journal Club commented:

In addition, a lay article about this paper is available at http://www.reuters.com/article/2015/10/29/us-health-surgeons-leadership-skills-idUSKCN0SN26L20151029#Al3U1ELlyhGcZiq0.97

On 2015 Dec 03, IGSJC - The International General Surgery Journal Club commented:

We're proud to announce that this article will be discussed online in a session of the International General Surgery Journal Club. The conversation will happen on Twitter with the hashtag #IGSJC on December 14 & 15, 2015, from 5-9pm EST each evening. For more information about IGSJC and how to use Twitter to participate in the conversation, visit https://igsjc.wordpress.com/guide-for-new-twitter-users/. Join us!

On 2017 Nov 14, Egon Willighagen commented:

There is a new paper on recent updates for WikiPathways: http://dx.doi.org/10.1093/nar/gkx1064

On 2016 May 22, Egon Willighagen commented:

Read about the integration of Reactome pathways into WikiPathways is this new paper: http://dx.doi.org/10.1371/journal.pcbi.1004941

On 2016 Mar 24, Egon Willighagen commented:

FYI, there is an R package to interact with the WikiPathways API: http://chem-bla-ics.blogspot.nl/2015/12/using-wikipathways-api-in-r.html and https://github.com/wikipathways/rwikipathways

I appreciate feedback!

On 2017 Feb 07, Michelle Fiander commented:

Systematic Reviews (SRs) are a common publication format, but manuscripts frequently do not follow national or international guidance in the conduct or reporting of these reviews. This is not entirely surprising considering the time it takes to become fully (or moderately?!) conversant SR methodologies. With that said, journal editors and authors should engage methodologists to assess SR manuscripts in light of conduct standards (e.g. Cochrane, AHRQ or other) and reporting standards (PRISMA) to ensure systematic reviews fulfill their role of providing reliable, synthesized evidence to inform practice, policy and further research in health care.

This manuscript addresses an interesting and important topic--the impact of state regulations on nurse practitioners' contribution to health care in the US. Unfortunately, it is not clear what systematic review methodology was followed for the conduct or reporting of the review. Conduct standards include Cochrane Handbook; Finding What Works in Health Care (National Academy of Science, etc), among others. Reporting standards include PRISMA, and Cochrane MECIR. Below are 4 areas where this manuscript does not seem to adhere to standards.

1. Risk of Bias: Tool not described or referenced; data of RoB assessment for each study is not provided; seven (7) studies were excluded from the review based on risk of bias--which is not common practice.

Examples of RoB Tools: Cochrane RoB tool-- https://sites.google.com/site/riskofbiastool/ ROBINS (RoB for non-randomized studies)-- https://sites.google.com/site/riskofbiastool/

About RoB: http://methods.cochrane.org/bias/assessing-risk-bias-included-studies. Reporting RoB: http://prisma-statement.org

1. Search Methods: incomplete; not reproducible. A search strategy is described, but is not presented; thus there is no way for readers or methodologists to verify how the evidence base was identified, or if the searches identified the included studies. PRISMA and Cochrane MECIR Standards suggest including at least 1 search strategy "as run", e.g. showing number of results per line as well as total results of the combined concepts within the strategy.

2. No evidence of an a priori protocol.

3. Inclusion criteria are folded into a paragraph on search methods. While inclusion criteria can impact a search strategy--e.g. study design (an inclusion criteria) will determine the type of methodological filter or hedge employed--they are distinct from methods used to identify the evidence base.

On 2016 Feb 01, Lu Wang commented:

Thanks for Cattoretti’s comment, including the literatures presented. Certainly, we agree with the information that specificity of any antibodies used in research should be standardized to ensure the accuracy of research (Bradbury A and Plückthun A). We also agree with the opinion that commercial antibody producers should test their antibodies more rigorously before selling them to scientists who often lack the resources and expertise to evaluate acquired antibodies (Holmseth S, et al). For example, no matter how to explain the result of a preadsorption test, it needs the enough free antigen to perform the test (Holmseth S, et al), hence PRDM1 antibody is also the case. As to the detection system of PRDM1 antibody in our study, please notice that both the final dilution and the secondary antibody were clearly stated in the paper. Meanwhile, cytoplasmic staining of PRDM1 is a question really puzzling us, because this expression pattern was different from all the previous reports. As you said that while cytoplasmic localization of PRDM1 needs to be proven and founded by some methods (for instance, preadsorption test, etc), however, our research design is scientific and rigorous. Firstly, we have established strict controls during beginning of IHC assay, including blank control. When the detecting system excluded PRDM1 antibody, no any signals were found in both cytoplasmic and nuclear of normal follicular epithelial cells. Secondly, we got same results in immunofluorescence detection. There is no exogenous or endogenous biotin participated in the response as everyone knows. Thirdly, we detected PRDM1 mRNA in whole cell mRNA, it exists in cytoplasms, too. So, we trust that there must be some unclear mechanisms to make this nuclear transcription factor expressed in cytoplasms. According the reasons mentioned above, we trust our results and decided to submit our research paper at last. Thanks again and best regards.

On 2015 Dec 04, Giorgio Cattoretti commented:

The claim that the cytoplasmic PRDM1 staining (a nuclear transcription factor) is indeed specific relies on the sole evidence of staining with an antibody known to produce non-specific cytoplasmic staining, whose concentration in the final dilution is not stated, counterstained with an unknown secondary antibody in a tissue known to be rich in endogenous biotin. As the sole first evidence of cytoplasmic localization of PRDM1, this is unproven, unfounded and false until proven otherwise with tools nowadays accepted for showing antibody specific staining in situ (see Bradbury A, Plückthun A (2015). Reproducibility: Standardize antibodies used in research. Nature 518:27-29. and Holmseth Set al (2012). Specificity controls for immunocytochemistry: the antigen preadsorption test can lead to inaccurate assessment of antibody specificity. J Histochem Cytochem 60:174-187.). Respectable Journals are increasingly called to enforce strict quality controls for these type of often flawed results. Best regards Cattoretti (see PMID:11232338, 12204275 and 15772984)

On 2015 Oct 27, Bastian Fromm commented:

the affiliations look little odd...

On 2016 Apr 09, Lydia Maniatis commented:

"This result suggests that the effect of pattern masking is not the integration of mask and target (Breitmeyer, 1984), but instead the complete suppression of the target at threshold."

As always, it depends on the structure of the mask/stimulus, something these authors don't seem to appreciate.

On 2015 Nov 18, Yuri Lazebnik commented:

Editorial or Advertorial?

An apparently objective article that is sponsored by a manufacturer to promote its products is known in the advertisement business as an advertorial, a portmanteau of ‘advertisement’ and ‘editorial’. We are concerned that this editorial has too many similarities with an advertorial, giving an impression that this genre of advertisement has found home in a venerated scientific journal.

Indeed, the editorial, which introduces the Nature Insight on precision medicine, is sponsored by AstraZeneca, a company whose financial interest depends on implementing the concept of precision medicine. As the company explains: “[AZ] is proud to support a new Insight by Nature on precision medicine. Currently, 80% of our pipeline takes a precision medicine, or personalized healthcare approach.” https://www.astrazeneca.com/our-company/media-centre/articles/astrazeneca-is-proud-to-support-a-new-insight-by-nature-on-precision-medicine-14102015.html. In other words, promoting precision medicine would promote AstraZeneca’s future products. As would also be expected from an advertorial, the editorial overlooks the voices of skeptics (1,2). Likewise, the papers commissioned by the Editor for the Insight are overwhelmingly optimistic. Although each acknowledges the enormous changes in health care that would be required for implementing precision medicine, none considers that these changes may be infeasible, unaffordable, and distract us from finding inexpensive practical alternatives. The appearance of bias is not helped by finding that three of the four articles declare competing financial interests.

Although the Nature editorial has acknowledged AstraZeneca sponsorship, this disclaimer fails to dispel the impression that the editorial promotes the products of its sponsor and can lead one to suggest that one of the most revered of scientific journals has abandoned its role as an unbiased moderator. Even a hint of advertorial practices would hardly be beneficial to anyone at a time when the influence of industry on the integrity of research and clinical practices is increasingly becoming a public concern.

Michael J. Joyner, MD, Caywood Professor of Anesthesiology,Distinguished Mayo Investigator, Mayo Clinic

Nigel Paneth, MD, MPH, University Distinguished Professor, Departments of Epidemiology & Biostatistics and Pediatrics & Human Development, College of Human Medicine, Michigan State University

Yuri Lazebnik, PhD, Lerna Consulting, LLC

The authors thank Professors Ana Soto, Carlos Sonnenschein, Sandro Galea, Richard Cooper, and Arturo Casadevall for their helpful comments

REFERENCES

1. Bayer, R. & Galea, S. Public Health in the Precision-Medicine Era. The New England Journal of Medicine 373, 499-501 (2015), doi:10.1056/NEJMp1506241.

2. Joyner, M. J. & Paneth, N. Seven Questions for Personalized Medicine. JAMA 314, 999-1000, (2015) doi:10.1001/jama.2015.7725 .

On 2017 Mar 02, Prajak Barde commented:

A study by Kaul U successfully concluded that paclitaxel-eluting stents failed to demonstrate non-inferiority against everolimus-eluting Stents that in diabetic patients undergoing PCI.

The trial was powered to detect non-inferiority on the basis of Spirit IV trait assuming that 1 year rate of target vessel failure of 8.4% with each stent, with non-inferiority margin of 4 %. However, documented failure rate in Spirit IV was 4.2 % in CAD patients and 6.4% in diabetic patients. Author did not explain the reason for these differences which potentially may lead to under powering of trial.

Author choose the margin (4%) based on Absolute risk difference that can result in under powering of the trial due to lower than expected event rates. Author might have foreseen the differences in the event rate however this was not justified why the sample size adjustment was made.

In the current the relative risk was inflated substantially increasing the risk of accepting the hypothesis of non-inferiority, which was not demonstrated due to clear differentiation of the effect.

On 2015 Nov 12, Sameer Al-Abdi commented:

Result of this outstanding trial can be summarized metaphorically as “The operation (Budesonide) was successful, but the patient died”

On 2017 Nov 18, Christopher Southan commented:

This publication is now updated in our 2018 version https://www.ncbi.nlm.nih.gov/pubmed/29149325

On 2015 Dec 22, Christopher Southan commented:

The new Concise Guide to PHARMACOLOGY (Dec 2015) has been compiled from this database content http://www.ncbi.nlm.nih.gov/pubmed/26650438

On 2015 Oct 15, Christopher Southan commented:

While this supercedes the 2014 equivalent (http://www.ncbi.nlm.nih.gov/pubmed/24234439) the information in both papers is complementary

On 2017 Jan 14, Boris Barbour commented:

I have tried to clear up my confusion about the simulation conditions and more specifically the boundary conditions, which must often represent the physical context of the sphere. The Perspective seems to offer several versions, none of which is entirely satisfactory.

1) The images in Fig. 3 might be taken to suggest that the spheres are surrounded by nothing. In consequence, my calculation in the previous comment assumed a vacuum. However, one should of course read the equations and not put one's faith in Nature illustrations. I apologise to the authors and readers for initially jumping to this probably wrong conclusion.

2) Box 1 states that there is "no flux of the electric field" at the spine boundary. I think this would imply no net charge within and no net charge outside the spine, which would appear to be inconsistent with the case where the spine contains ions. (There are other logical possibilities: a grounded boundary or some exotic distributions with very specific inside-outside symmetries. But these are implausible in the present context). I therefore believe this statement refers to some special case or is an error. Or maybe it represents that well known and remarkably accurate approximation of strict electroneutrality?

3) Box 2 equation 5 gives a different boundary condition, this time with a non-zero electric field

E = Q/(4 * pi * er * e0)

However, I believe this contains an error. If the denominator also included a factor of R² , the equation would be dimensionally correct and also compatible with an infinite aqueous medium. The physical model would then be water everywhere, with a thin, "electrically invisible" barrier enclosing the ions. There would be no ions outside the spine. (It should be understood that we are dealing with "idealised" water that contains no ions, not even from the spontaneous dissociation of water molecules that occurs in reality.)

The corrected option 3) and therefore the infinite aqueous milieu seems the most plausible to me. In this case the arguments in the previous comment about the large potentials required in the "vacuum" case are still relevant, except that the values would be attenuated 80-fold by the relative permittivity of water. Voltages would range from about 20mV for 1000 ions to about 20V for 1000000. Figs. 3b and 3c would still require clearly unphysiological voltages.

In Fig. 4, the authors argue that the model spine has a "logarithmic capacitance" (i.e. V is proportional to log(Q)). However, they define V as the voltage difference between the centre and the rim of the spine. I see two problems with this definition.

1) A minor issue is that, in the model, most ions added to the spine distribute to radii other than zero (the centre), attaining a different potential in the process. An alternative might be the potential averaged over all charges.

2) A more significant issue is that the reference voltage is taken at the rim of the spine, which is well within the potential field of the spine, thereby excluding some if not most of that potential. All practical measurements and most neuronal current loops involve much larger distances outside the spine, and for these a reference at infinity is a reasonable approximation. This alternative of an external reference would require inclusion of the large voltages mentioned above. These are proportional to the charge and are likely to dominate the small intra-spine nonlinearities highlighted by the authors.

In the end, however, spines exist neither in a vacuum nor even in distilled water. As discussed in my earlier comments, most if not all of the electrical behaviour in insulators (dielectrics) described by the authors is likely a poor guide for what happens in conductors. The phenomena could thus be radically altered by the inclusion of a conducting external solution, ions of opposite charges at realistic (near equal) concentrations and a membrane.

On 2017 Jan 07, Boris Barbour commented:

[This comment has been edited as a result of uncertainty about the exact conditions for the simulations. Follow up in next comment.]

A simple calculation shows how unrealistic assumptions of significant deviations from electroneutrality are.

Imagine the sphere is in a vacuum, such that there is no external solution and the membrane capacitance is therefore absent. Then from Gauss' Law and symmetry it can be shown that the potential at the rim of the sphere with respect to infinity (i.e. a distant reference) will be:

V = Q/(4 * pi * e_0 * r),

where V is the potential, Q the charge, r the radius of the sphere and e_0 the vacuum permittivity. (This is equivalent to the classical isolated spherical conductor.)

Taking e_0 = 8.85E-12F/m, r = 0.5E-6m and an electronic charge of 1.6E-19C, we can calculate that just a single ion produces a potential of about 2.8mV. The 1000 ions in the simulation of Fig. 3a will therefore create a voltage of 2.8V (in my slightly smaller sphere), while the 1E6 charges of Fig. 3c would generate a whopping 2.8kV. Thus, the largest physiological potential could not generate a deviation from electroneutrality greater than the equivalent of a few tens of ions in the spine.

In practice a few thousand ions of the same sign can be accumulated (calculated in my first comment), but this is only possible because the charge-compensating mechanisms of the membrane capacitance and polarisation of the medium are present, which largely preserve electroneutrality.

On 2016 Dec 26, Boris Barbour commented:

Despite this article appearing in a journal dedicated to reviews, I appreciate it is tagged as a "Perspective", which presumably indicates that it contains some speculation. However, I still think it is worth pointing out to readers that electrostatic screening has in effect been omitted from the modelling, which could have an extremely strong influence on the phenomena illustrated, particularly as this is not made clear in the article.

This problem will only be addressed by analytical or numerical approaches that include at least two oppositely charged species at approximately physiological concentrations. I insist on this point because the cited preprint (like this Perspective) models the charge alone, which means only a single charged species is considered. Additional manuscripts treating this non-biological situation do not help address the biologically relevant questions. The standard approach to simulating the two-species system would be to include distinct electrodiffusion equations for the concentrations of cations and anions, whose difference yields the net charge distribution.

In the spirit of not jumping to conclusions or trusting to intuition in a complex situation, I would equally encourage readers (and the authors) to exercise due care before extrapolating from simulations with low concentrations of a single charged species to the situation with high concentrations of oppositely charged species.

EDIT: The effective omission of the membrane capacitance should also be rectified in any realistic modelling. This is because most if not all of the net charge in the spine will accumulate below the membrane, where it is quasi-neutralised by a symmetrical accumulation of opposite net charge the other side of the membrane. Without this mechanism, the actual capacitance of the spine/membrane will presumably be lower than observed in situ. In other words, biologically relevant voltages could not drive many charges into a spine without a membrane capacitance. The present modelling ignores the membrane capacitance because there is no extracellular saline in which an oppositely charged accumulation of ions can occur.

On 2016 Dec 26, Rafael Yuste commented:

We appreciate the concerns of Dr. Barbour and thank him for his frankness. But we alert the reader that this publication is not a proper experimental study but an opinion piece with a very limited set of data, which were added strictly to illustrate the point, and did not represent a comprehensive analysis of electrodiffusion in spines. Our review should be viewed precisely as what it is, a "Perspective", highlighting new and potentially controversial topics. The role of electrodiffusion in nanophysiology has been traditionally ignored by cable theory because of the difficulty to analytically solve the corresponding joint equations. Moreover, the interaction between electric fields and diffusion and geometry are not easy to dissect because they are highly nonlinear and sometimes counter intuitive and consequently difficult to discuss without adequate computational simulations. We, and others, are exploring this potentially important phenomenon with simulations and theoretical studies, some of which have already been published (https://arxiv.org/pdf/1612.07941.pdf), and which will hopefully address all the concerns of Dr. Barbour. We encourage the reader not to jump to quick dismissive conclusions but to be patient and wait for further work.

On 2016 Dec 26, Boris Barbour commented:

In this "perspective", the authors build up the importance of using full electrodiffusion simulations without an electroneutrality constraint to describe current flow and ion distributions within dendritic spines, rather than treating the cytoplasm as an Ohmic conductor. Although an electrodiffusion model is in principle more accurate and may in some cases be necessary, the authors do little to make their case here, because their illustrative model in Fig. 3 is quite divorced from physiological conditions.

They illustrate the equilibrium spatial distribution of ions within a small and isolated sphere of water, a situation loosely representing the head of a dendritic spine. When electrical interactions between the ions are considered at high concentrations, a highly nonuniform spatial distribution emerges. However, there are both trivial and fundamental problems with this apparently dramatic illustration.

It is worth fixing some numbers. Empirical measurements of membrane capacitance typically yield values of ~1microF/cm^2. So the capacitance of a typical spine head with diameter 0.5 microns will be about 10fF. Charging this capacitance to the most extreme membrane potentials observed in neurones (of the order of 100mV) will therefore require about 1fC, a charge that represents about 5000 ions. The volume of such a spine head would be about 70aL (attoLitres) and, assuming a 150mM saline of monovalents, would contain about 600000 each of cations and anions. The authors use a rather oversized spine head compared to this, but the numerical differences will be unimportant to what follows.

The authors simulate 1000, 100000 and 1000000 charges, but already the second number exceeds the largest net concentration of charges that will occur physiologically.

The concentrations they report are numerically incorrect at least in Fig. 3b (where the mean concentration should be 40 microM) and in Fig. 3c (where the mean concentration should be 400 microM).

A true representation of these gradients would include the existing 150mM saline. Thus, for Fig. 3a, the ~100nM gradient would be superimposed upon 150mM, about one part in a million, which may not be that significant.

Another important problem arises from the fact that the authors only include charges of one sign in their model, while physiological saline contains both cations and anions, in roughly if not exactly equivalent concentrations (we calculated above that the net charge represents of the order of 1% of ions present, and even this number is only made possible by the membrane capacitance). This unrealistic situation is likely to have a very strong influence on the behaviour illustrated, because the existence of ions of the opposite sign would allow electrostatic screening, which the authors have in effect banished from their model by only including a single ionic species.

The ionic gradients illustrated by the authors arise from collective mutual Coulombic repulsion (note that the submembrane ionic concentration shown is unrelated to the concentrations of charges normally occurring on either side of the membrane capacitance, because the external saline is absent in this model). Such Coulombic forces are usually annihilated over all but the shortest distances and times by electrostatic screening. The habitual approximation considers electrostatic shielding to cause an exponential attenuation of Coulombic interactions with a length constant of the Debye length. In 150mM monovalent saline the Debye length is about 0.6nm (using an approximate formula given on wikipedia), so, even over a fraction of a micron, Coulombic interactions should be attenuated to a cosmic degree and the forces generating the interesting concentration gradients of Fig. 3 are unlikely to operate.

In conclusion, although I'm not in a position to describe exactly the distribution of net charges in the spine head under physiological conditions in the absence of strict electroneutrality, neither are the authors. However, ignoring electrostatic screening seems to be an extremely unrealistic approximation.

On 2015 Nov 19, Cecile Janssens commented:

It is important to distinguish the method from how it was investigated. We applied the method to an unselected sample of meta-analyses rather than limiting to high-quality meta-analyses, because we wanted to show when the method works and when it doesn’t. Our results show that the method does not work that well when the topic of the meta-analysis is heterogeneous, and when there is a need to include grey literature or articles from non-English languages, which is no surprise. The poor performance of the method for these meta-analyses should not be interpreted as a shortcoming of the method, but as guidance for when and when not to use the method.

We did not suggest that the method should be used standalone. We mention in the first paragraph of page 8 that “additional strategies like these can be used to complement our search method–either to find more eligible studies or to increase confidence in the results of the search method when no other studies are found.”

Regarding the specific meta-analyses: The method does not improve the efficiency of meta-analyses that were already efficient. We screened more than 100% of the articles in the studies of Kumar, Stevens and Shan because they screened 573, 400 and 243 articles, compared to 1013, 536 and 289 using our method. More than half of the studies in the meta-analysis of Kumar were case studies, which often have no or a limited number of citations, and several of the other studies were published in non-English language. The topics of both other studies were quite heterogeneous. Shan investigated quality of life, covering quality of life instruments and assessments of functional status, and Stevens summarized six studies on a post-operative drug interaction, which investigated six entirely different surgeries. As said, the method does not work well when topics are heterogeneous.

Finally, this is the first paper on this method, describing two pilot studies. More work is certainly needed. Given the high efficiency of the method and the impressive accuracy, we are most interested in a direct comparison of the method with keyword searching in Pubmed as a first search. The method may well be a more efficient start of literature searching than keyword searching.

On 2015 Nov 10, Wichor Bramer commented:

This article investagates the effectiveness of citation based searching for systematic reviews, because the gold standard keyword-based searching is inefficient. The authors propose a new method of co-citation and bibliographic coupling. Their conclusion is that Citation searching is a reasonably accurate method. But what is reasonably accurate? One should not look at the median or mean or overall recall, but at the minimum recall. A researcher is not performing 20 reviews for which the median recall is accurate enough, researchers decide on which method they want to use for their single review. They want to minimize the chances of missing relevant articles in their N=1 case.

What was the question that the authors tried to answer? Can it be used as a standalone method for indetifying included references in systematic reviews? The answer has te be 'No!'. The most complete method in study 1 (screening all co-citations) retrieves and appropriate median of 94% of all included references, but the minimum of 75% is unacceptable. And that method is only occasionally more efficient than the traditional method. The method that is more efficient than traditional (all co-cited > 1) retrieves sometimes only 50% of all included references. In study 2 the best method (indirect and direct citations) retrieves a minimum of 10% of all included references. In seven out of 42 studies 50% or less of all included references were retrieved, and sometimes not even by reducing the number of hits needed to screen, such as Kumar, where screening 177% of the traditional method resulted in a recall of only 31% of the originally included studies. In their discussion the authors give reasons why their method does not work for the 5 lowest scoring articles, concluding that if these are not taken into account 89% of all references is found. But that still does not explain Stevens [68] in which 138% of hits have to be screened to find 50% of all includes, and Shan [65], where 142% is needed to find 58%.

An interesting question would be whether the described method can add additional references that were not found with the traditional keyword-based approach or traditional citation screening? This cannot be concluded from these results, as the investigators only tried to find the articles that were already included (and thus had been found by the traditional methods).

The conclusion of the authors that: "Researchers … [have] … the advantage of being able to screen only half of the number of articles compared to keyword-base literature search …[which]… is an efficient strategy for finding key articles related to one or more “known” articles" might be correct, but it cannot replace traditional keyword based searching, but only complement it, but that was not investigated.

On 2016 Jun 10, Ali Poormohammadi commented:

If it is possible please mention the experimental conditions in each step.

On 2016 Jun 09, Ali Poormohammadi commented:

None

On 2016 May 10, Morten Oksvold commented:

Please pay attention to the following report from ORI (Office of Research Integrity) before reading this article:

https://ori.hhs.gov/content/case-summary-pastorino-john-g

On 2017 Jun 28, Randi Pechacek commented:

Alex Alexiev briefly described this paper for the public on microBEnet.

On 2016 Mar 01, Ram Dessau commented:

Residual symptoms after Lyme neuroborreliosis. The prevalence is unknown? Comment by Ram B. Dessau

Department clinical microbiology Slagelse Hospital, Region Zealand Ingemannsvej 46, DK4200, Slagelse, Denmark ramd@regionsjaelland.dk

Dersch et al. have performed a systematic review and meta-analysis on the prevalence and spectrum of residual symptoms in patients after treatment for Lyme neuroborreliosis (LNB)[1]. 5579 bibliographic records were screened and 38 studies were included. An overview of treatment studies reporting residual symptoms is given. The follow up time is about one year concerning most of the studies. The conclusion is that 28% of LNB patients may experience symptoms after treatment. This conclusion may be misinterpreted as associated to LNB or caused by LNB. The 38 treatment studies were not designed to elucidate the association of LNB with residual symptoms, as control groups without LNB were not included.

Many of the cited symptoms like sensory disturbances, cognitive disturbances, pain and headache are non-specific and even common in the population at large. Thus, it would be of paramount importance to compare the prevalence of events to a control group using the same methods. Only a higher frequency of residual symptoms in the LNB group may indicate association with LNB. Due to the design of the included studies the residual symptoms could as well be due to other causes such as adverse effects of treatment [2].

Reporting the gross proportion of events may be misleading for other reasons. The load of residual symptoms will depend on not only the number of events for each symptom, but also the number of recorded items and the length of time, where symptoms may occur. The more questions you ask and the higher the gross rate of symptoms will become and eventually the rate of symptoms may approach 100%. Especially if there is also a long time span for the observations.

Along this line of reasoning Dersch et al. also misinterpreted the studies including a control group[1]. The results of these studies do not “remain inconclusive”, but show comparable results, even if there are differences. These differences are not very large and a substantial proportion of the controls also have a similar level of problems. For example the mean physical component FS36 score in LNB was 44(9) compared to 51(6) in controls[3]. A significant difference in mean score was found, but the size of the standard deviation show that the distribution of scores in the two groups have a large overlap. Given that the 95% interval is about 2 SD then FS36 score in LNB is 26-62 and in controls 39-63. Even if LNB patients have “significantly” lower FS36 score than controls, this is not a large difference. Thus a weak or absent association alone could explain that some studies do not report statistical differences and some do due to random variation. Thus, residual symptoms associated with LNB are probably quite rare, if at all.

Grouping together any symptoms without considering a variable relevance to LNB is problematic. For example, it was found in Swedish children that residual symptoms were persistent nerve defects such as facial palsy, but not nonspecific subjective symptoms [4]. Thus, it would be important to distinguish different types of symptoms and their severity. The study by Dersch et al. adds to a large volume of uncontrolled case reports and case series overstating the possible risk of symptoms associated with Lyme borreliosis[5]. This may contribute to unnecessary anxiety about Lyme borreliosis.

It is acknowledged that Dersch et al [1] do state important reservations about especially the "possible" case definitions, as patients with other conditions could contribute to the high load of reported symptomps.

Conclusion The prevalence of residual symptoms associated with LNB after treatment may not be concluded from the study by Dersch et al [1] because control groups are missing. A meaningfull interpretation is not possible. The authors should have focused on the available controlled studies instead.

Conflict of interests: None to declare.

References

1.Dersch R, Sommer H, Rauer S, Meerpohl JJ. Prevalence and spectrum of residual symptoms in Lyme neuroborreliosis after pharmacological treatment: a systematic review. J Neurol 2015 Oct 12.

2.Dersch R, Freitag MH, Schmidt S, Sommer H, Rauer S, Meerpohl JJ. Efficacy and safety of pharmacological treatments for acute Lyme neuroborreliosis - a systematic review. Eur J Neurol 2015 Sep;22:1249-59.

3.Eikeland R, Mygland A, Herlofson K, Ljostad U. European neuroborreliosis: quality of life 30 months after treatment. Acta Neurol Scand 2011 Nov;124:349-54.

4.Skogman BH, Glimaker K, Nordwall M, Vrethem M, Odkvist L, Forsberg P. Long-term clinical outcome after Lyme neuroborreliosis in childhood. Pediatrics 2012 Aug;130:262-9.

5.Baker CJ, et al. Final report of the Lyme disease review panel of the infectious diseases society of America (http://www.idsociety.org).2010.

On 2015 Oct 17, David Keller commented:

Wine should be diluted to reduce epithelial carcinogenic effects in the upper aerodigestive tract

Epidemiologic evidence suggests that people with diabetes are at significantly higher risk for many forms of cancer [1]. It has also been observed that cancer patients with pre-existing diabetes experience higher mortality than cancer patients without diabetes [2]. The randomized controlled trial by Gepner and colleagues confirmed prior observations that moderate ethanol consumption is associated with improved cardiometabolic risk in diabetics and in other groups [3].

A large observational study reported data on cancer and other harms among drinkers whose usual alcoholic beverage was wine, beer or spirits [4]. A dose-response relationship was identified between the concentration of ethanol in the beverage of choice, and the incidence of cancers among those who consumed it [5]. The risk of cancer for drinkers of beer (which contains about 5% ethanol), wine (12% ethanol) and distilled spirits (40% ethanol) were each divided by the cancer risk for teetotalers to yield respective cancer harm ratios (HRs) of 1.2, 1.38 and 1.69; note the dose-response relationship between the ethanol concentration in a beverage, and cancer rates for those who drink that beverage. The cancer HR for beer drinkers was not statistically significantly elevated above that for teetotalers, but its trend is consistent with the proposed dose-response relationship.

Given the elevated risk of cancer in diabetics, and a risk of cancer in drinkers of alcoholic beverages which increases monotonically with the concentration of ethanol in the beverage, it seems prudent for diabetics who consume ethanol to dilute their beverage of choice to a concentration of 5% or less. Wine can be diluted to 4% ethanol by combining one part wine with two parts water or juice.

References:

1: Giovannucci E, Harlan DM, Archer MC, Bergenstal RM, Gapstur SM, Habel LA, Pollak M, Regensteiner JG, Yee D. Diabetes and cancer: a consensus report. Diabetes Care. 2010 Jul;33(7):1674-85. doi: 10.2337/dc10-0666. PubMed PMID: 20587728; PubMed Central PMCID: PMC2890380.

2: Ranc K, Jørgensen ME, Friis S, Carstensen B. Mortality after cancer among patients with diabetes mellitus: effect of diabetes duration and treatment. Diabetologia. 2014 May;57(5):927-34. doi: 10.1007/s00125-014-3186-z. Epub 2014 Mar 15. PubMed PMID: 24633676.

3: Gepner Y, Golan R, Harman-Boehm I, Henkin Y, Schwarzfuchs D, Shelef I, et al. Effects of Initiating Moderate Alcohol Intake on Cardiometabolic Risk in Adults With Type 2 Diabetes: A 2-Year Randomized, Controlled Trial. Ann Intern Med. [Epub ahead of print 13 October 2015] doi:10.7326/M14-1650

4: Smyth A, Teo KK, and the PURE Investigators. Alcohol consumption and cardiovascular disease, cancer, injury, admission to hospital, and mortality: a prospective cohort study. Lancet. 2015 Sep 17. doi:10.1016/S0140-6736(15)00235-4. PubMed PMID: 26386538.

5: Keller DL. Dose-response relationship observed between concentration of ingested alcohol and cancer rate. Comment on PMID 26386538. In: PubMed Commons [Internet]. National Library of Medicine; 2015 Sept 26 [cited 2015 Oct 12]. Available from: http://www.ncbi.nlm.nih.gov/pubmed/26386538#cm26386538_11980

The above comment is also posted on the following Annals of Internal Medicine web page:

http://annals.org/article.aspx?articleid=2456121

On 2015 Oct 15, Andrea Messori commented:

Producing evidence in support of disinvestment: the experience of the Tuscany region in Italy

HTA Unit, ESTAV, 50100 Firenze (Italy)

In June 2013, the regional institution (ESTAV) responsible for drug acquisition in Tuscany (Italy) started a project aimed at identifying a series of pharmacological treatments suitable for disinvestment (i.e. suitable to be replaced by an equi-effective and less costly alternative). One criterion for this initiative was to write a full scientific report focused on comparative effectiveness for each therapeutic issue.<br> Table 1 shows the bibliografic details of the main 10 analyses of disinvestment conducted within this project.

Table 1. Information shown in this table: (a) therapeutic issue; (b) more costly treatment; (c) less costly tretament; (d) references

[1] -----------------------------------------------------------------------

(a) First line treatment of previously untreated patients with genotype-1 HCV infection ; (b) Sofosbuvir + ledipasvir (12 weeks) ; (c) Ritonavir-boosted dual therapy (12 weeks) ; (d) Trippoli S, Fadda V, Maratea D, Messori A. Bayesian network meta-analysis to evaluate interferon-free treatments in naive patients with genotype 1 hepatitis C virus infection. Eur J Gastroenterol Hepatol. 2015 Aug;27(8):983-4.

[2] -----------------------------------------------------------------------

(a) First line treatment of previously untreated patients with multiple sclerosis ; (b) Beta-Interferon ; (c) Azathoprine ; (d) Messori A, Fadda V, Maratea D, Trippoli S. Indirect meta-analytical comparison of azathioprine and of beta interferon effectiveness in all forms of multiple sclerosis pooled together. J Neurol Sci. 2014 Dec 15;347(1-2):408-10.

[3] -----------------------------------------------------------------------

(a) Prevention of Venous Thromboembolism in Major Orthopedic Surgery ; (b) Enoxaparin ; (c) Dalteparin ; (d) Messori A, Trippoli S, Maratea D, Fadda V, Marinai C. Prevention of venous thromboembolism in major orthopedic surgery: Bayesian network meta-analysis of 21 randomized trials evaluating unfractionated heparins, low-molecular weight heparins, and new oral anticoagulants. Int Cardiovasc Res J.2015;9(4):238-242

[4]------------------------------------------------------------------------

(a) Targeted Treatments for Pulmonary Arterial Hypertension: ; (b) Various agents of recent approval ; (c) Sildenafil ; (d) Badiani B, Messori A. Targeted Treatments for Pulmonary Arterial Hypertension: Interpreting Outcomes by Network Meta-analysis. Heart Lung Circ. 2015 Jul 6. pii:S1443-9506(15)01248-2. doi: 10.1016/j.hlc.2015.05.020. [Epub ahead of print]

[5]------------------------------------------------------------------------

(a) Thromboprophylaxis in orthopedic surgery ; (b) Rivaroxaban ; (c) Apixaban ; (d) Messori A, Fadda V, Maratea D, Trippoli S, Marinai C. Testing the therapeutic equivalence of novel oral anticoagulants for thromboprophylaxis in orthopaedic surgery and for prevention of stroke in atrial fibrillation. Int J Clin Pharmacol Ther. 2015 Mar;53(3):211-9. doi: 10.5414/CP202183.

[6]------------------------------------------------------------------------

(a) Treatment of Moderate-to-Severe Psoriasis by Subcutaneous Route ; (b) Ustekinumab 45mg ; (c) Adalimumab ; (d) Messori A, Fadda V, Maratea D, Trippoli S, Gatto R, De Rosa M, Marinai C. Biological drugs for the treatment of moderate-to-severe psoriasis by subcutaneous route: determining statistical equivalence according to evidence-based methods. Clin Drug Investig. 2014 Aug;34(8):593-8.