There are specific guidance on metadata collection provided by the Genomic Standards Consortium https://genomicsstandardsconsortium.github.io/mixs/HumanVaginal/

Xiao L; Sonne SB; Feng Q; Chen N; Xia Z; Li X; Fang Z; Zhang D; Fjære E; Midtbø LK; Derrien M; Hugenholtz F; Li J; Zhang J; Liu C; Hao Q; Vogel UB; Mortensen A; UR Kleerebezem MW; Licht TR; Li Y; Arumugam M; Wang J; Madsen L; Kristiansen K (2017): Data and analysis of diet-induced and obesity-associated alterations of gut microbiota of 129S6/Sv and C57BL/6J mice GigaScience Database. DOI:10.5524/100271

GigaByte journal is now accepting and encouraging data paper submissions including GBIF datasets of >5000 occurences. This was begun as a WHO initiative to encourage GBIF submissions for vector datasets, but is being continued and expanded to any biodiversity datasets that are of suitable significance to warrent a data paper. https://gigabytejournal.com/articles/series/GIGABYTE_SERIES_0002

You can annotate and highlight specific sections of text using Hypothes.is

This tree has been re-assembled with some new data and a new assembly tool, please see dataset DOI:10.5524/100773

You can even add Hypothes.is annotations to this blog post.

test comment

Hilmar Lapp - whats the OBF? non-profit volunteer run advocate group for open science, in particular bioinformatics. run many projects including Bio* and EMBOSS.

Nomi Harris- talking about the importance of OPEN data and the sessions structures etc. Quote from Madeleine Ball about open data sharing "its like donating my body to research but I don't have to die"

last speaker before lunch on last day. talking about low input MG. bridge the gap between single cell and MG. 1 cell = 1fantogram, >1ng for MG DNA. low input MG is the region in the middle ~~100fg to 100picograms Some environmental samples are low in microbial biomass so the levels of DNA are low.

postdoc in Hugenholtz lab.

how do we classify bacterial species. obviously cant use sexual reproduction species concept. now a days we tend to use a polyphasic approach to include multiple pheno/genotype to group "similar" cells into a species group. but infact ~85% of species there is sufficient recombination to allow the true sexual reproduction definition of the biological species classification. This work has shown that sequence identity does NOT nessasarily deliniate species. i.e. some bacterial "species" have upto 30% sequence divergence, but are still the same biological species and capable of recombination within species.

from QingTao (where the beer comes from :-) ) Raman-activated single cell genomics - imaging data combined with genomics data = Ramanome. Raman Spectra gives indication of particular chemical composition of cell, the theory is that particular cellular state has a specific spectral profile that can be identified and used to isolate cells of interest. Can be used to do RWAS (raman wide association studies) spectral imaging of raman stuff can be done in hours, so much faster than seq. RAMDB - Ren Lh Su XQ etal.

unifying taxonomy: UniEuk. primarily interested in unicelluar/microbial eukaryotes - commonly called Protists. very diverse in physiology and phylogenetics. cover both botanical and animal taxonomy.

talking about viral taxonomy and the "virosphere" . Viruses are an important part of life on earth but little catered for in the archives and research.

protein families. used in metagenomics via: CheckM, IGM, MEGAN, MG-RAST, WebMGA, FANTOM, InterProScan. so What make virus protein families different? have mostly polyproteins, which can reshuffle to create different multi-domain proteins. They have a faster evolution than many cellular genomes, which means while they maintain structural similarities they often have very differnt nucleotide sequence in the genes.

seq data getting bigger still, should be associated with opportunities rather than "deluge" or other pesimistic words. IMG system is trying to integrate all 5billions genes. estimated 100billion in 10yrs time.

Gave a 10min talk on Island sampling day at last years GSC meeting in Crete. participants of meeting went out to collect samples on the island with extensive metadata. the island has a diversity of habitats ranging from 0m to 2500m in elevation. Agricultural (inc palm groves, orchards etc), beaches, national parks, urban. clay, silt, sand, and more.in total took 435 samples. used GIS app for data collection, it includes GPS and photos and GSC metadata. backed up with traditional datasheets.

Antartica - 95% covered by ice, 70%of earths fresh water. taken~100 samples from various locations over antartica. lots of DNA extractions on location, and oxford nanopore seqs on location.big problem is low battery life at such low temps! ~45mins and laptops die. Managed to get 10k+ reads. DNA is currently being seqs on pacbio now in labs.

Australian microbiome initiative: network of ~40 partners including gov and academic and industrial. ~3000microbiomes data in the database already. core data is stuff they have paid for the production of, BASE project - soils project started in 2013 highly standardised. ~1200 sample locations in Aus. soils samples are archived by CISRO. 2nd set of core data is marine microbes. time series of samples.

mining genomes from public metagenomes. >1500MGs from >150SRA proects. resulted in ~8000 "complete" genomes of microbes. ~3500 of "high" quality. this dataset increases the phylogenetic branch length of the tree of bacterial life by ~33%, even in well covered phylum like firmicutes. Data for this was submitted prior to 2016.

Part Two: Quality of RefSeq genomes (r80) 75k genomes, >90%complete, only ~400genomes of questionable completeness and quality.

CWL - now funded 50% by Elixir.

slides

CWL being used to optimise and shrink EMG pipeline, from >9000 lines of code to ~350 lines in CWL. Human freindlly, and machine executble. MGRast and EMG are standardising their workflows using CWL. to allow swap compute and benefit anyone wanting to run either or both systems on their own systems. CWL homepage

exhisting terminology is insufficient . models capture with their own terminology . a standard will enable across model .<br> break genetic interaction down into 3 controlled vocabs (1) Perturbation tyep. (2)phenotype (3)interaction type . interaction type broken down into 3 modules (a) (b)& (c) .

presenter

why work on kinome? - kinases are key therapeutic targets

presnter

changed title to "EC numbers for Dummies"

presenter

presenter

presenter

presenter

https://twitter.com/drsbello/status/846155714124931073

presenting

Patrick Ruch presenting

presenter

this is what EFO is. it links multiple ontologies together and unifies different data using different ontologies

focus on pre-competitive drug targets.

speaker

http://repurpos.us/#/

presenter

integrated knowledgebases of cancer information

presenter

Jodi Hirschman presents.

change of title to "taming the (flying) spaghetti monster of prot-prot interactions" Bravely doing a live demo of the website

This is going to be awesome! <br> Join the Giga-Curation Challenge at www.annotometer.com

http://141.61.102.18/phosida/index.aspx

Title of talk= DataScience@NIH: Pivoting to the future

presenting on PDB (Protein data bank)

ncbi taxID=3311

join the challenge!

just wanted to add an annotation somewhere to check what I see on the admin page

https://usegalaxy.org/

https://en.wikipedia.org/wiki/Notch_signaling_pathway

not going to talk about pigs today. Homology Independant Targeted Intergration (HITI). methodology to help select for specific inserts of a specified DNA insert in the correct orientation at a specific genomic loci. based on non-homologuos End repair pathway, which works in all stages of cell cycle (unlike other pathways which are only active in cell dividion stages of cell cycle). Used for knock-in experiments. shown to work in rat model using retinal gene knockin. i.e. a rat strain with 1.8kb deletion in retinal gene, they can show recovery of full vision with knockin of the missing exon 1.8kb fragment using this method.

using pig models of human disease. and possible future xenotransplantation organs. Intro to CAS9 CRISPR editiing. chromotin structure has effect on specificity of CRISPR. Open structure has higher specificty, so is a better target. Tested some drugs to open the chromotin structure. Most of these drugs also effect methylation and cell proliferatoin, so had to use low conc of drugs to test effect on genome editing specificty of crispr. Also shown that the gRNA structure is also important for efficacy, so making sure the gRNA doesn't form high energy secondary structures (hairpin loops), increases efficacy.

Where Richard see's genomics taking us. historical technology overview. combination of long read with short data is opening up a new era of discovery in genomics. Lake Malawi Cichlids - massive radiation of cichlids species happened within the 1million years. >500 soeices in different lakes in same geographic region. so far have sequenced 80 species, Entire radiation of species have only slightly more variation between genetics than would expected within species, i.e. ~0.2% Looking to use to define phenotype-genotype relations. Genome sequences show evidence that there may have been horizontal gene flow between species. backing up claims of horizontal gene transfer between humans and denisovian and neanderthals. Specieation may preceed genetic separation rather than vice versa. Malawi cichlid project was made possible by G10K doing 1 fish first giving a suitable genetic map. The G10K project is ready for the next big push, with 100's this year and 1000's of genomes in 2017. The hangover from sequencing the "reference genome" is that we kind of expect a single sequence for an organism, but the reality is that diploid genomes have 2 slightly different genomes. (and to a lesser extent each individual cell has its own genome,) this hetrozygosity needs to be taken into account when doing genome assembly. Long read data helps with this, it can provide information on hetrozygosity. using a phaser like "WhatsHap" you can align pacbio reads to hi-fidelity illumin contigs (i.e. an assembly that doesn't collapse hetrozygous regions) the long reads can then idntify those short contigs as individual hapoltypes. G10K aims to cover all 75 orders of fish within the next year.

Slightly jetlagged. First action on stage; to destroy the decorations :-D PGD and PGS - intro & controversy PGD - Pre-implantation genetic diagnosis. The ability to select the embryo before implantation, therefore ensuring no (known) diseases, particularly important for carriers of recessive disease. historically extracting cells from invitro fertilised eggs anywhere from day1 to day6. Use of arrayCGH from a single cell you can detect large scale chromosomeal abnoralities (trisomy etc) PGS-contreversy was that non sequencing methods of testing dont show the full story. ESHRE have set up an RCT case control study to hopefully rectify this. Chromosomal mosaicism at blastocyst stage causes an obvious problem when doing bio-opsy! if 1in 5 cells show different chromosomal it may not give relevant results. ending with ethical dilemas 17 different companies offering Expanded carrier screening of parents, in total >1100 possible things to be screened for, but only 9 of those were the same thing being screened in all 17 tests! i.e. each company offers a different set of screens. Aim is to offer free (in UK, paid by NHS) NGS seq of parents. Looking at using CRISPR to eliminate disease from genome. Ethical issues of that need to be thought about. already certain places allow selection of child based on Sex, but its the same principal to select the exact traits the parents wants for their child! Conclusions. Preconception carrier screening is becoming the norm. PGS for IVF Everyone will being born will soon already have the genome sequenced before they are even born.

Risk of developmental delay is not even throughout the world. Environmental factors, including social, family, economic and physical environment all effect a childs development. Stress produces hormones that are toxic to brain development. The Raine Study. many areas of research carried out, one in particular was the study of stressful events during pregnancy. it showed 1 or 2 stressful events during preg didn't show any significant effect of development by age 14, but 3 or more did show negative effect. Mechanisms for this involved gene-environment interactions. e.g. epigenetics. Effects during pregnancy on the epigenome could effect not only mother and unborn child, but also the DNA within the feotus that will become the sperm/egg for the granchild, therefore effecting the next generation. The effect of the microbiome can also be seen on child development. maternal diet can effect the microbiome, which in term can effect the child, both during pregnancy and during breastfeeding stage. mouse study shows that breastfeeding mitigates certain genetic predispositions to weight gain. Conclusion - there are multiple genomes and multiple env factors that need to be teased apart to help understanding of child development. Early identification of risk and targeted intervention will be the most effective way to mitigate possible detremental effects.

The Alliance for Human development, Uni Toronto.

Use this page to make notes using the hypothes.is app

including video introductions from Sydney Brenner and Francis Collins

grand finale...... drum roll please..... <br> committee = Mark Liles, Sara Sjoling, Gunter Brader. Winner is Aline Pushkarev. Congratulations .

phd student under Marla Trindada at IMBM, UWC. looking at surfice active molecules (surfacants). <br> assay for surfacants is to look for reduced surface tension, and can cause emulsions of non-miscible liquids (e.g. oil and water).<br> massive commercial uses, market estimated at ~30billion USD.<br> until june 2016, no bio-surfactant from metagenome had been described.<br> acyltransferase is required to join hydrophilic to hydrophobic ends of surfactant molecules, but acyltransferases are virtually infinately diverse. which is why a functional metagenomics approach is hoped to provide a useful tool to discover new ones. <br> library created in e.coli. expression done in psudomonas matida. <br> functional screen done by spraying kerosine over surface of colonies! colonies expressing surfactants show up as "halo" around the colony.

creating large numbers of clones capable of holding >100kb which he suggests is large enough to contain a complete functional operon. the clones are unbias, and ammeneable to expression.<br> Also able to make Fungal Artificial Chromosomes (FAC) clones for cloning and expression in fungus. (aspergillus).

polyketides synthases (PKS) are essential for synthesis of things like rapamycin and other antimicrobials.

looking at a variety of extreme environments, with the hope of answering the question "is life possible on other planets in our solar system?" <br>

inventer of TA cloning<br> looking for phage DNAP "Replicases"<br>

SINTEF - not industrial, not academic, somewhere in between.~2000 staff, mostly in trondhiem, ~70different nationalities.<br>

creating large libraries of fosmid clones and screening for polysaccharide activity using a novel butterfly wing method. i.e. the width of the polysaccharide layer effects the wavelength of light reflected, so the amount of actively is shown in the assay by different reflective pattern when shining light onto the plate.

ArcticZymes = bioprospecting company<br> based at the most northern university in the world. within the arctic circle in northern Norway. <br> new project is for looking at viruses trying to annotate more proteins from viruses, currently this group is largely underrepresented in the databases and difficult to predict functionally.

454 shotgun MG of composter containing shit from 3000 exotic animals in San Paulo Zoo. Results show ~90bacteria, some archea, some virus, but no fungi. Several time points over 99day time line. MG, MT and 16S<br> MT -functional profile shanges over time, but interestingly the point directly after turning of heap closely resembles the early time point samples, i.e. the addition of airation "resets" the microbiome activity to some extent.<br> using mytaxa binning they reconstructed top 5 most abundant genomes. top = 2.48% of classified reads Rhodothermus marinus.<br> From the unclassified reads they constructed a variety of novel microbial genomes.<br>

CLC Bio - is now Qiagen. CLC workbench is software for all things genomics, including metagenomics analysis. including assembly.

3 types of vineyard being explored, chemical pest control, bio-dynamic and conventional

very little data to show yet, just in process of setting up. Looking at the rhizosphere to help increase yield of crops without increasing the need for extra land (i.e. food security).<br> Much lost yield is due to "stress" of various sorts (drought, flood, etc).

currently working towards metagenomics version 3.0 - which is generating a method to screen, clone, express and functionally characterize within a few months instead of a few years.

METAFLUIDICS - a project to elucidate unknowns within metagenomic datasets currently trying to publish 96 gut MG datasets from a variety of zoo animals.

bacteriorhodopsin is actually found primerily in archea. Alina presents evidence for presence in bacteria

currently a MG assembly graph is a massive ball of tangled wool! EGE food waste reactor - Oslo, capacity of ~50,000tonnes of food waste, generates both methane (used for busses and taxis) and slurry fertilizer for farmers/land use. binning done by MaxBin, assessment of these bins suggests bins containing high coverage longer reads are OK with single copy marker genes, but the lower the coverage and contig length the worse the binning. A developement version of MetaSpades was used to make the assemblies as it out performed any individual assembler, using pacbio CCL and Illumina reads

making functional metagenomics libraries with different hosts (i.e. not E.coli) e.g Gluconobacter oxydans 621H -

CM2BL.org canadian metaMicrobiome library, is an effort to make a resource of shared mirobiome libraries. made in Cosmid clones

Guest editor for metaFUNC series http://tinyurl.com/gpl3nc7

the eukaryotic DNA in a sample (particular focus on soil) is 10-100x less than bacterial, so how to effectively sample it? Also, eukaryotes have introns. Best approach is to use MetaTranscripomics., makes selective use of polyA tail on euk genes and removes the problem of introns.

nifty use of mini pH meteres in conjunction with 96well plates on a robot to allow fairly high throughput screening

2nd yr PhD student, a pool of 1,250,000 clones! 50um cell solution required to make ~10million droplets (compared t. ~1000 96well plates would take>1000litres of cell solution). FADS = similar to FACS but for droplets that may contain 1 cell (~40% have 0 cells, ~20% have 2 or more).

Only has undergrads, no MSC, no PhDs or post-doc as his institute doesn't have any post-grad school. This means he can focus on the more repetitive manual tasks, e.g. plate-out these 1000 clones.

sample north sea water through multiple filters, 10,3 and 2um filters. for 16s analysis, they generated the Tax4Fun software to analyse >300samples worth of data. it takes 16s data and predicts functional profile from it. benchmarked it against various MG sequenced datasets and showed Spearmans rho values of >0.8 (I'm not sure what that means? but I know I don't believe 16s analysis as a proxy for function!)

bio-reactors are very similar to various guts, but one major difference is the build-up of intermediary compounds (in particular Ammonia, NH3) , which in organisms are often incorporated into the host or excreted, but in reactors can build up and cause instability of the system. See Adrian Naas -poster P5

Metaexplore project: metagenomics for bioexplorations, tools and collaborations

Kiritimati = Christmas islands Christmas islands microbial biomats are so thick you can sample with a spade! 10-20cm thick. 16S diversity of mat layers used to demonstrate "types of bacteria" in each layer, e.g. anaerobic, phototrophic etc... Using Illumina "long read" to look at functional MG of different layers of a biomat from Lake 21 (Christmas Island) read length average~5KB, FIRST TIME ever that I have seen evidence of 16S taxonomic profile giving something similar to MG taxonomy!!

exploring plant - microbiome interactions

microbial community metabolism drives ecosystem function. studying low temp waters around the world, including Baltic Sea

talking about glacial cryconite, 24 x 16S samples, 12 of those also underwent full MG seq. half of samples from each of Baltoro (Pakistani Karakoram) and Forni (Italian Alps). 1.6m predicted genes in total. Genes assigned to taxonomy based on homology marked difference between species present in these 2 very similar environments Hypothesis is that pollutants have major effect of these ecosystems, so set up experiments in controlled microcosm. The experiment show differences but not statistically significant. Light & anaerobic with the addition of cryconite sample appeared to have largest effect of CPF degredation.

Nice outdoors meal, and later a natural light show! https://twitter.com/Ronny_King_/status/780153457764368384

Why not use hypothes.is to add your notes directly to the program, linked to the appropriate speaker section, AND share your notes with the world?!

A talk from someone that has done 16S and MT and MG, hopefully they will be able to describe the functional differences between the cohorts

Prof. Glenn R. Gibson from the @UniofReading

@trinebrounge

@jac_ravel

Dr. Majdi Osman @OpenBiome

@DrAilsaHart

@reid_probiotics

@gbgloor

@gbgloor https://github.com/ggloor

@dbkell

16S analysis used as primary sub-grouping of samples, then did DESI-MS metabolite analysis of another swab taken at same time as 16s extraction swab. nothing clear in results presented.

why care about it? preterm birth, aquisition and transmission of HIV and STIs are both effected by it. in contrast to most sites, LOW diversity of microbiome is considered healthy, mostly lactobacillus.

neil williams = exercise physiologist GOS = galacto-oligosaccharide asthma = inflammatory disease of the airways rise in prevelence of astma associated with life style (including antibiotics) and diet changes.

http://www.cfmatters.eu/

Defined the term "microbiota transplant" hopefully this can be seen in the video after thats published.

"no-one likes talking about Pooh more than us"

~2.5% of potential donors are accepted into stool bank, >250questions asked in pre-donnation screening

While FMT for C.diff is recognised as a good treatment, there is a relatively low uptake by the UK NHS trusts. In IBD, there are many publications with a variety of outcomes of use of FMT for IBD, with no conclusion as yet, but there is evidence that certain fractions of the IBD spectrum may perform better in responce to FMT.

to be summarized in 1 line at the end apparently!

the wiki for the workshop: https://github.com/ggloor/CoDa_microbiome_tutorial/wiki the data required to work through the course: https://github.com/ggloor/CoDa_microbiome_tutorial

nice talk on working towards causation over correlation by looking at functions of microbes, particularly with respect to IgA. Discussed previous work (http://www.cell.com/cell/abstract/S0092-8674(14)01001-0) and eluded to continuing work with more data.

The Giga-Curation Challenge 2016 ends HERE

a VSM curation interface

Self proclaimed research parasite

nice description of "buckets" of data being made shareable in association with a particular publication (e.g. like Dryad and GigaDB

Wow, i'm looking forward to this one ;-)

BioCuration participants- You can add comments about each talk here if you want!

this is the most important part

https://twitter.com/drsbello/status/719791714735665152

interesting data for mining potential new drugs

MatLab tools

python.org

Breast cancer

https://en.wikipedia.org/wiki/Notch_signaling_pathway

6 first authors!

the test in HK

Chinese (Asian) human cancer samples

109 Danish, 11 Chinese and 140 Spanish samples

123 non-obese and 169 obese Danish individuals

MetaHIT, 396 samples

MetaHIT, 124 samples

These data should be cited as: Fang, X; Huang, Z; Li, L; Luo, R; Wang, X; Xu, F; Yang, L; Zhu, Y (2012): Genomic data from the Pacific oyster (Crassostrea gigas). GigaScience. http://dx.doi.org/10.5524/100030 The FTP server location given here will shortly be replaced with a new one meaning these links will no longer work.

Join the Giga-Curation Challenge 2016! at www.annotometer.com

Soon to be announced, THE curation competition of the BioCuration2016 conference. Organised by GigaScience and featuring Hypothes.is and iCLiKVAL.

the addition of information(curation) to biological data(bio).

This was the first GigaDB dataset, and so deserves the first hypothes.is note being added!

so does it work on the carosel thing then?

see wikipedia for definition https://en.wikipedia.org/wiki/Battle-axe_(woman) (nb, sorry Rob for using your blog to try out this tool with random junk!)