Statistical Modeling
I see from your code page that you used other stats like Kolmogorov-Smirnov Test and Levene’s Test, for example. I think, these should be described somewhere here in this section as well as assumptions and potential violations to assumptions.
References
Probably interesting for background:
Eklund, A., Frank, J., Nilsson, L., Zetterberg, A., & Mansson, J. 2024. Times of trouble - Seasonal variation in number and severity of attacks on sheep caused by large carnivores and eagles in Sweden. European Journal of Wildlife Research, 70(9): 2-11. DOI: https://doi.org/10.1007/s10344-023-01761-4
Kvalshaug, O.J. 2013. Inter-specific patterns of depredation on domestic sheep and semi-domestic reindeer in Norway, by a large predator guild. Master Thesis, Norwegian University of Life Sciences, 36.
Linnell, J.D.C., Nilsen, E.B., Lande, U., Herfindal, I., Odden, J., & Skogen, K. 2005. Zoning as a means of mitigating conflicts with large carnivores: Principles and reality. Conservation Biology Series-Cambridge, 9: 163-175. DOI: https://doi.org/10.1017/cbo9780511614774.011
Mabille, G., Stien, A., Tveraa, T., Mysterud, A., Brøseth, H., & Linnell, J.D.C. 2015. Sheep farming and large carnivores: What are the factors influencing claimed losses? Ecosphere, 6(5): 1-17. DOI: https://doi.org/10.1890/es14-00444.1
Strand, G., Hansen, I., De Boon, A., & Sandström, C. 2019. Carnivore Management Zones and their Impact on Sheep Farming in Norway. Environmental Management, 64: 537-552. DOI: https://doi.org/10.1007/s00267-019-01212-4
Strand, G. 2020. The combined effects of centralization and carnivore management on sheep farmers and sheep farming in Norway. Human Dimensions of Wildlife, 26(4): 321-336. DOI: https://doi.org/10.1080/10871209.2020.1818895
Seasonal and Regional Variations in Sheep and Reindeer DNA Detection
Brown bear densities: chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://static02.nmbu.no/mina/publikasjoner/mina_fagrapport/pdf/mif82.pdf Region 8 = Troms og Finnmark Region 7 = Nordland Region 6 = Central Norway (Midt-Norge) including Trøndelag og Møre- og Romsdal
For reindeer, take a look here: https://www.reinbase.no/Studer-reindriften/Reintall you may have to sum up number of reindeer in different regions (for example, Nord-Trøndelag and Sør-Trøndelag). This can be discussed.
For sheep, data may be found here: Nordland: https://beitestatistikk.nibio.no/fylker/ny/nordland Trøndelag: https://beitestatistikk.nibio.no/fylker/ny/troendelag Troms: https://beitestatistikk.nibio.no/fylker/ny/troms Finnmark: https://beitestatistikk.nibio.no/fylker/ny/finnmark
The previous NINA report should be discussed somewhere here: https://brage.nina.no/nina-xmlui/handle/11250/3065394?locale-attribute=no
Are the results comparable or not between this study and Kopatz et al.?
significant
non-significant
was not significant
was clearly not significant
Additionally, regional comparisons showed significant differences between Troms og Finnmark and Trøndelag (estimate = 3.02, SE = 1.13, p = 0.020), but no significant differences between Nordland and Troms og Finnmark (p = 0.99) or between Nordland and Trøndelag (p = 0.079). (Figure 1).
I assume that you aim to account for the differences in sample and detection sizes when looking at the regional differences?
is.F
Add space.
Season and Region
season and region (i.e., no capital letters)
2021.
in four Norwegian counties (i.e., Troms and Finnmark, Trøndelag, Nordland, Innlandet).
“In conclusion, the model revealed that reindeer DNA detection was significantly higher in Spring compared to Summer. Among the regions, Trøndelag exhibited significantly lower detection rates overall when compared to Nordland (Figure 2). The high Conditional R² value suggests that individual variability among bears plays a critical role in explaining the variation in reindeer DNA detection. Additionally, the fixed effects of season and region contribute substantially to the model’s overall performance.”
Is this a citation from somewhere/someone? I'm wondering because it is written with quotation marks.
gender
I recommend to use the term sex here instead. In this instance, the sex (i.e., female or male) was determined by genotyping of sex chromosomes and no social-cultural context is looked at here.
Fall (128), followed by Summer (44) and Spring (27).
fall, summer and spring (i.e., no capital letters)
Rangifer tarandus
Rangifer tarandus
Please check that Latin names are written in italics througout the document.
able 1:
Is there something to be shown in the column called False binding (dG)?
Sheep and Reindeer
sheep and reindeer
other mammals
Add a list of used mammal species here or in a table.
DBI, RSQLite, dplyr, lubridate, sp, mapview, glmmTMB, DHARMa, performance, MuMIn, emmeans, car, ggplot2, and kableExtra.
Add versions and relevant references to tese packages, if available. Add references to reference list.
I assume that you will provide the full code as an appendix to your thesis? If so, please add link to repository here.
norway
Norway
norway
Norway
s
Scandinavia
using different extraction kits
and DNA extraction was done using different extraction kits
PrimerBlast
Add version and citation here for the program used.
BioEdit
Add version and citation here.
Sample Selection
This section requires more work. Please provide a clear explanation of your selection criteria for all the samples included in the analysis. Your sampling logic must be understood by the examiner to determine its suitability for answering your research questions. It will also be beneficial to relocate the "Data Management" section below and integrate it here for a more coherent flow of logic.
NetPrimer and PrimerPremier5
Add versions and citations here.
How does seasonality affect the likelihood of detecting sheep and reindeer DNA in brown bear fecal samples?
Expand this section to include a sentence or two that formulates a testable hypothesis based on previous research. The same goes for the second question below.
Aims and hypothesis:
Jumping directly to these research questions without first addressing assay development and validation seems strange, as the assays haven't undergone prior testing in an independent, peer-reviewed study.
Specifically, this study seeks to answer two key questions:
Addressing these questions largely depends on the precision of newly developed qPCR assays, which should also be stated here, at least as an objective.
BioEdit
Include references for the software you used.
Tissue samples from other mammals were collected opportunistically, with some, such as moose or reindeer, donated by hunters, while others were used in previous studies. These tissue samples were utilized for target and non-target experiments during the testing phase and served as positive controls in fecal analysis.
Please provide a sentence or two explaining the selection of these species used to test the cross-reactivity of the primer/probe systems, along with the actual names of the panel of mammals used in this study.
Although the TaqQuest mastermix is not specified as inhibitor-resistant, positive results from previous microsatellite analyses suggested minimal inhibition.
Indeed, you did conduct a test beforehand prior to large-scale work, correct?
Primer triplets
"primer/probe systems" - Hopefully, the Introduction section will cover the TaqMan assay's principle.
The only primer pair retained from the literature was the sheep primer set from Dalsecco (2018), which was slightly modified.
I find this situation quite unusual. Please clarify your selection criteria and also state exactly how you modified this primer.
The reverse primer was reverse complemented due to the G at the 3’ end and used as a probe in the TaqMan assay. New priming sites were manually identified, and potential primers were evaluated using NetPrimer and PrimerPremier5 for Tm, secondary structures, and binding sites. Primers were designed to avoid overlap, targeting the cytB gene.
This information is very difficult to understand. It is not clear which assay or primer/probe system this refers to.
The PCR run included a hold stage at 95°C for 10 min, followed by 40-45 cycles at 60°C for 90 s (annealing and extension) and 95°C for 15 s (denaturation). Data was collected at the end of each annealing step.
Also state the name of the qPCR instrument that was used to run and collect the data here.
Primer Validation
More detail is needed here, include how you determined the limit of detection of your assay. State that you used standard curves to estimate limit of detection (LOD), but see Klymus et al. (2020). Given that your assays are for qualitative purposes, the limit of quantification (LOQ) is likely not relevant in your case. Please verify this to clarify in the main text why the qPCR efficiency may be irrelevant for your assays, but the LOD is.
Depending on who you will get as an examiner, it may be worthwhile to also mention that you did the testing according to the MIQE guidelines, which I think were incorporated into this paper (see thier Appendix S1 for the checklist):
Thalinger, B., Deiner, K., Harper, L. R., Rees, H. C., Blackman, R. C., Sint, D., ... & Bruce, K. (2021). A validation scale to determine the readiness of environmental DNA assays for routine species monitoring. Environmental DNA, 3(4), 823-836.
Bustin, S. A. (2024). Improving the quality of quantitative polymerase chain reaction experiments: 15 years of MIQE. Molecular aspects of medicine, 96, 101249.
Klymus, K. E., Merkes, C. M., Allison, M. J., Goldberg, C. S., Helbing, C. C., Hunter, M. E., Jackson, C. A., Lance, R. F., Mangan, A. M., Monroe, E. M., Piaggio, A. J., Stokdyk, J. P., Wilson, C. C., & Richter, C. A. (2020). Reporting the limits of detection and quantification for environmental DNA assays. Environmental DNA, 2, 271–282. https://doi.org/10.1002/edn3.29
Most testing assays were normalized to a DNA concentration of 5 ng/µl. The sheep-specific primer set Ovi-JOE showed cross-talk into the ROX channel above 0.8 µM. Based on similar amplification curves for concentrations from 0.5 µM to 0.7 µM, a final concentration of 0.5 µM (forward/reverse) and 0.25 µM (probe) for the Ovi-JOE primer set was chosen to minimize cross-talk. The reindeer-specific primer set Ran-ROX delivered consistent curves across all concentrations, differing only in intensity. Given ROX’s lower intensity compared to JOE, the highest tested concentration of 1.0 µM for forward/reverse and 0.5 µM for the probe in the final reaction was used.
This should be included in a separate section on primer/probe system optimization and validation. I believe there are differing perspectives here on how to test for fluorescence crosstalk or bleedthrough, and the definition thereof. Please clearly define the experimental setup you used to test for crosstalk. Did you follow the approach outline in Grohmann et al, (2021: Guidance document on multiplex real-time PCR methods, EUR 30708 EN, Publications Office of the European Union, Luxembourg, 2021, ISBN 978-92-76-37820-4, doi:10.2760/243914, JRC125188: https://publications.jrc.ec.europa.eu/repository/handle/JRC125188), page 17?
predation
I propose a different title - probably something like this (but this can be further refined): Brown bear (Ursus arctos) predation on sheep (Ovis aries) and reindeer (Rangifer tarandus), two culturally and economically important species in northern Norway: a molecular detection approach
Samples were pipetted from and to 96-well plates, including two negative and two positive controls (5 ng/µl sheep and reindeer DNA mix).
Please include more detail here, including the number of technical replicates used. The consensus is that a sample should be run with a minimum of three technical replicates, although more are generally preferred. For each sample, the number of positive hits would be reported over the number of replicates run. Replicates are essential to your dataset because you are engaged in modeling. The below section is confusing for me in that regard. How do you model a dataset where, in the case of absences, there is a 50% chance of missing a hit due to the limit of detection (LOD) of your assays? Please explain this to the reader. I believe that running your samples once without replication is a fundamental flaw that limits the application of GLMMs on your current dataset.
DNA extracts were retrieved from a minus eighty-degree freezer, previously used for population genetics studies.
This may be seen as a weakness if not properly explained, as the DNA extraction from scats was mainly focused on obtaining DNA from the source species rather than the dietary DNA itself, no? Please provide a report on the subsampling process for DNA extraction from scats, including whether the approach differed for different types of scats and the specific DNA extraction method employed. This is crucial for determining the extent of random effects that can be incorporated in GLMMs, if one could model the data in its current form.
Primer Design
This section requires more details. For a detailed overview of the workflow and inclusion criteria for primers from the literature, please refer to van der Pouw Kraan et al. (2024; https://doi.org/10.1111/1755-0998.13945) and their Figure 1 and Table 1. It would be beneficial if you could also report this section accordingly. Currently, this section assumes too much from the reader and is overly generalistic. Also reference Table 1 below here.
Data Production
A summary table on number of hits and replicates per individual and season and year may be interesting here to understand whether there was variation in detection across time or according to season.
Statistical Modeling
I see from your code page that you used other stats like Kolmogorov-Smirnov Test and Levene’s Test, for example. I think, these should be described somewhere here in this section as well as assumptions and potential violations to assumptions.
References
Probably interesting for background:
Eklund, A., Frank, J., Nilsson, L., Zetterberg, A., & Mansson, J. 2024. Times of trouble - Seasonal variation in number and severity of attacks on sheep caused by large carnivores and eagles in Sweden. European Journal of Wildlife Research, 70(9): 2-11. DOI: https://doi.org/10.1007/s10344-023-01761-4
Kvalshaug, O.J. 2013. Inter-specific patterns of depredation on domestic sheep and semi-domestic reindeer in Norway, by a large predator guild. Master Thesis, Norwegian University of Life Sciences, 36.
Linnell, J.D.C., Nilsen, E.B., Lande, U., Herfindal, I., Odden, J., & Skogen, K. 2005. Zoning as a means of mitigating conflicts with large carnivores: Principles and reality. Conservation Biology Series-Cambridge, 9: 163-175. DOI: https://doi.org/10.1017/cbo9780511614774.011
Mabille, G., Stien, A., Tveraa, T., Mysterud, A., Brøseth, H., & Linnell, J.D.C. 2015. Sheep farming and large carnivores: What are the factors influencing claimed losses? Ecosphere, 6(5): 1-17. DOI: https://doi.org/10.1890/es14-00444.1
Strand, G., Hansen, I., De Boon, A., & Sandström, C. 2019. Carnivore Management Zones and their Impact on Sheep Farming in Norway. Environmental Management, 64: 537-552. DOI: https://doi.org/10.1007/s00267-019-01212-4
Strand, G. 2020. The combined effects of centralization and carnivore management on sheep farmers and sheep farming in Norway. Human Dimensions of Wildlife, 26(4): 321-336. DOI: https://doi.org/10.1080/10871209.2020.1818895
Seasonal and Regional Variations in Sheep and Reindeer DNA Detection
Brown bear densities: chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://static02.nmbu.no/mina/publikasjoner/mina_fagrapport/pdf/mif82.pdf Region 8 = Troms og Finnmark Region 7 = Nordland Region 6 = Central Norway (Midt-Norge) including Trøndelag og Møre- og Romsdal
For reindeer, take a look here: https://www.reinbase.no/Studer-reindriften/Reintall you may have to sum up number of reindeer in different regions (for example, Nord-Trøndelag and Sør-Trøndelag). This can be discussed.
For sheep, data may be found here: Nordland: https://beitestatistikk.nibio.no/fylker/ny/nordland Trøndelag: https://beitestatistikk.nibio.no/fylker/ny/troendelag Troms: https://beitestatistikk.nibio.no/fylker/ny/troms Finnmark: https://beitestatistikk.nibio.no/fylker/ny/finnmark
The previous NINA report should be discussed somewhere here: https://brage.nina.no/nina-xmlui/handle/11250/3065394?locale-attribute=no
Are the results comparable or not between this study and Kopatz et al.?
significant
non-significant
was not significant
was clearly not significant
is.F
Add space.
Season and Region
season and region (i.e., no capital letters)
Additionally, regional comparisons showed significant differences between Troms og Finnmark and Trøndelag (estimate = 3.02, SE = 1.13, p = 0.020), but no significant differences between Nordland and Troms og Finnmark (p = 0.99) or between Nordland and Trøndelag (p = 0.079). (Figure 1).
I assume that you aim to account for the differences in sample and detection sizes when looking at the regional differences?
gender
I recommend to use the term sex here instead. In this instance, the sex (i.e., female or male) was determined by genotyping of sex chromosomes and no social-cultural context is looked at here.
“In conclusion, the model revealed that reindeer DNA detection was significantly higher in Spring compared to Summer. Among the regions, Trøndelag exhibited significantly lower detection rates overall when compared to Nordland (Figure 2). The high Conditional R² value suggests that individual variability among bears plays a critical role in explaining the variation in reindeer DNA detection. Additionally, the fixed effects of season and region contribute substantially to the model’s overall performance.”
Is this a citation from somewhere/someone? I'm wondering because it is written with quotation marks.
2021.
in four Norwegian counties (i.e., Troms and Finnmark, Trøndelag, Nordland, Innlandet).
Fall (128), followed by Summer (44) and Spring (27).
fall, summer and spring (i.e., no capital letters)
DBI, RSQLite, dplyr, lubridate, sp, mapview, glmmTMB, DHARMa, performance, MuMIn, emmeans, car, ggplot2, and kableExtra.
Add versions and relevant references to tese packages, if available. Add references to reference list.
I assume that you will provide the full code as an appendix to your thesis? If so, please add link to repository here.
Rangifer tarandus
Rangifer tarandus
Please check that Latin names are written in italics througout the document.
able 1:
Is there something to be shown in the column called False binding (dG)?
other mammals
Add a list of used mammal species here or in a table.
Sheep and Reindeer
sheep and reindeer
norway
Norway
norway
Norway
s
Scandinavia
using different extraction kits
and DNA extraction was done using different extraction kits
BioEdit
Add version and citation here.
NetPrimer and PrimerPremier5
Add versions and citations here.
PrimerBlast
Add version and citation here for the program used.
Aims and hypothesis:
Jumping directly to these research questions without first addressing assay development and validation seems strange, as the assays haven't undergone prior testing in an independent, peer-reviewed study.
Sample Selection
This section requires more work. Please provide a clear explanation of your selection criteria for all the samples included in the analysis. Your sampling logic must be understood by the examiner to determine its suitability for answering your research questions. It will also be beneficial to relocate the "Data Management" section below and integrate it here for a more coherent flow of logic.
Specifically, this study seeks to answer two key questions:
Addressing these questions largely depends on the precision of newly developed qPCR assays, which should also be stated here, at least as an objective.
How does seasonality affect the likelihood of detecting sheep and reindeer DNA in brown bear fecal samples?
Expand this section to include a sentence or two that formulates a testable hypothesis based on previous research. The same goes for the second question below.
Tissue samples from other mammals were collected opportunistically, with some, such as moose or reindeer, donated by hunters, while others were used in previous studies. These tissue samples were utilized for target and non-target experiments during the testing phase and served as positive controls in fecal analysis.
Please provide a sentence or two explaining the selection of these species used to test the cross-reactivity of the primer/probe systems, along with the actual names of the panel of mammals used in this study.
Primer triplets
"primer/probe systems" - Hopefully, the Introduction section will cover the TaqMan assay's principle.
BioEdit
Include references for the software you used.
The PCR run included a hold stage at 95°C for 10 min, followed by 40-45 cycles at 60°C for 90 s (annealing and extension) and 95°C for 15 s (denaturation). Data was collected at the end of each annealing step.
Also state the name of the qPCR instrument that was used to run and collect the data here.
Although the TaqQuest mastermix is not specified as inhibitor-resistant, positive results from previous microsatellite analyses suggested minimal inhibition.
Indeed, you did conduct a test beforehand prior to large-scale work, correct?
The only primer pair retained from the literature was the sheep primer set from Dalsecco (2018), which was slightly modified.
I find this situation quite unusual. Please clarify your selection criteria and also state exactly how you modified this primer.
The reverse primer was reverse complemented due to the G at the 3’ end and used as a probe in the TaqMan assay. New priming sites were manually identified, and potential primers were evaluated using NetPrimer and PrimerPremier5 for Tm, secondary structures, and binding sites. Primers were designed to avoid overlap, targeting the cytB gene.
This information is very difficult to understand. It is not clear which assay or primer/probe system this refers to.
Primer Validation
More detail is needed here, include how you determined the limit of detection of your assay. State that you used standard curves to estimate limit of detection (LOD), but see Klymus et al. (2020). Given that your assays are for qualitative purposes, the limit of quantification (LOQ) is likely not relevant in your case. Please verify this to clarify in the main text why the qPCR efficiency may be irrelevant for your assays, but the LOD is.
Depending on who you will get as an examiner, it may be worthwhile to also mention that you did the testing according to the MIQE guidelines, which I think were incorporated into this paper (see thier Appendix S1 for the checklist):
Thalinger, B., Deiner, K., Harper, L. R., Rees, H. C., Blackman, R. C., Sint, D., ... & Bruce, K. (2021). A validation scale to determine the readiness of environmental DNA assays for routine species monitoring. Environmental DNA, 3(4), 823-836.
Bustin, S. A. (2024). Improving the quality of quantitative polymerase chain reaction experiments: 15 years of MIQE. Molecular aspects of medicine, 96, 101249.
Klymus, K. E., Merkes, C. M., Allison, M. J., Goldberg, C. S., Helbing, C. C., Hunter, M. E., Jackson, C. A., Lance, R. F., Mangan, A. M., Monroe, E. M., Piaggio, A. J., Stokdyk, J. P., Wilson, C. C., & Richter, C. A. (2020). Reporting the limits of detection and quantification for environmental DNA assays. Environmental DNA, 2, 271–282. https://doi.org/10.1002/edn3.29
Most testing assays were normalized to a DNA concentration of 5 ng/µl. The sheep-specific primer set Ovi-JOE showed cross-talk into the ROX channel above 0.8 µM. Based on similar amplification curves for concentrations from 0.5 µM to 0.7 µM, a final concentration of 0.5 µM (forward/reverse) and 0.25 µM (probe) for the Ovi-JOE primer set was chosen to minimize cross-talk. The reindeer-specific primer set Ran-ROX delivered consistent curves across all concentrations, differing only in intensity. Given ROX’s lower intensity compared to JOE, the highest tested concentration of 1.0 µM for forward/reverse and 0.5 µM for the probe in the final reaction was used.
This should be included in a separate section on primer/probe system optimization and validation. I believe there are differing perspectives here on how to test for fluorescence crosstalk or bleedthrough, and the definition thereof. Please clearly define the experimental setup you used to test for crosstalk. Did you follow the approach outline in Grohmann et al, (2021: Guidance document on multiplex real-time PCR methods, EUR 30708 EN, Publications Office of the European Union, Luxembourg, 2021, ISBN 978-92-76-37820-4, doi:10.2760/243914, JRC125188: https://publications.jrc.ec.europa.eu/repository/handle/JRC125188), page 17?
Primer Design
This section requires more details. For a detailed overview of the workflow and inclusion criteria for primers from the literature, please refer to van der Pouw Kraan et al. (2024; https://doi.org/10.1111/1755-0998.13945) and their Figure 1 and Table 1. It would be beneficial if you could also report this section accordingly. Currently, this section assumes too much from the reader and is overly generalistic. Also reference Table 1 below here.
Samples were pipetted from and to 96-well plates, including two negative and two positive controls (5 ng/µl sheep and reindeer DNA mix).
Please include more detail here, including the number of technical replicates used. The consensus is that a sample should be run with a minimum of three technical replicates, although more are generally preferred. For each sample, the number of positive hits would be reported over the number of replicates run. Replicates are essential to your dataset because you are engaged in modeling. The below section is confusing for me in that regard. How do you model a dataset where, in the case of absences, there is a 50% chance of missing a hit due to the limit of detection (LOD) of your assays? Please explain this to the reader. I believe that running your samples once without replication is a fundamental flaw that limits the application of GLMMs on your current dataset.
Data Production
A summary table on number of hits and replicates per individual and season and year may be interesting here to understand whether there was variation in detection across time or according to season.
predation
I propose a different title - probably something like this (but this can be further refined): Brown bear (Ursus arctos) predation on sheep (Ovis aries) and reindeer (Rangifer tarandus), two culturally and economically important species in northern Norway: a molecular detection approach
DNA extracts were retrieved from a minus eighty-degree freezer, previously used for population genetics studies.
This may be seen as a weakness if not properly explained, as the DNA extraction from scats was mainly focused on obtaining DNA from the source species rather than the dietary DNA itself, no? Please provide a report on the subsampling process for DNA extraction from scats, including whether the approach differed for different types of scats and the specific DNA extraction method employed. This is crucial for determining the extent of random effects that can be incorporated in GLMMs, if one could model the data in its current form.
Statistical Modeling
I see from your code page that you used other stats like Kolmogorov-Smirnov Test and Levene’s Test, for example. I think, these should be described somewhere here in this section as well as assumptions and potential violations to assumptions.
References
Probably interesting for background:
Eklund, A., Frank, J., Nilsson, L., Zetterberg, A., & Mansson, J. 2024. Times of trouble - Seasonal variation in number and severity of attacks on sheep caused by large carnivores and eagles in Sweden. European Journal of Wildlife Research, 70(9): 2-11. DOI: https://doi.org/10.1007/s10344-023-01761-4
Kvalshaug, O.J. 2013. Inter-specific patterns of depredation on domestic sheep and semi-domestic reindeer in Norway, by a large predator guild. Master Thesis, Norwegian University of Life Sciences, 36.
Linnell, J.D.C., Nilsen, E.B., Lande, U., Herfindal, I., Odden, J., & Skogen, K. 2005. Zoning as a means of mitigating conflicts with large carnivores: Principles and reality. Conservation Biology Series-Cambridge, 9: 163-175. DOI: https://doi.org/10.1017/cbo9780511614774.011
Mabille, G., Stien, A., Tveraa, T., Mysterud, A., Brøseth, H., & Linnell, J.D.C. 2015. Sheep farming and large carnivores: What are the factors influencing claimed losses? Ecosphere, 6(5): 1-17. DOI: https://doi.org/10.1890/es14-00444.1
Strand, G., Hansen, I., De Boon, A., & Sandström, C. 2019. Carnivore Management Zones and their Impact on Sheep Farming in Norway. Environmental Management, 64: 537-552. DOI: https://doi.org/10.1007/s00267-019-01212-4
Strand, G. 2020. The combined effects of centralization and carnivore management on sheep farmers and sheep farming in Norway. Human Dimensions of Wildlife, 26(4): 321-336. DOI: https://doi.org/10.1080/10871209.2020.1818895
Seasonal and Regional Variations in Sheep and Reindeer DNA Detection
Brown bear densities: chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://static02.nmbu.no/mina/publikasjoner/mina_fagrapport/pdf/mif82.pdf Region 8 = Troms og Finnmark Region 7 = Nordland Region 6 = Central Norway (Midt-Norge) including Trøndelag og Møre- og Romsdal
For reindeer, take a look here: https://www.reinbase.no/Studer-reindriften/Reintall you may have to sum up number of reindeer in different regions (for example, Nord-Trøndelag and Sør-Trøndelag). This can be discussed.
For sheep, data may be found here: Nordland: https://beitestatistikk.nibio.no/fylker/ny/nordland Trøndelag: https://beitestatistikk.nibio.no/fylker/ny/troendelag Troms: https://beitestatistikk.nibio.no/fylker/ny/troms Finnmark: https://beitestatistikk.nibio.no/fylker/ny/finnmark
The previous NINA report should be discussed somewhere here: https://brage.nina.no/nina-xmlui/handle/11250/3065394?locale-attribute=no
Are the results comparable or not between this study and Kopatz et al.?
was not significant
was clearly not significant
significant
non-significant
Season and Region
season and region (i.e., no capital letters)
is.F
Add space.
Additionally, regional comparisons showed significant differences between Troms og Finnmark and Trøndelag (estimate = 3.02, SE = 1.13, p = 0.020), but no significant differences between Nordland and Troms og Finnmark (p = 0.99) or between Nordland and Trøndelag (p = 0.079). (Figure 1).
I assume that you aim to account for the differences in sample and detection sizes when looking at the regional differences?
“In conclusion, the model revealed that reindeer DNA detection was significantly higher in Spring compared to Summer. Among the regions, Trøndelag exhibited significantly lower detection rates overall when compared to Nordland (Figure 2). The high Conditional R² value suggests that individual variability among bears plays a critical role in explaining the variation in reindeer DNA detection. Additionally, the fixed effects of season and region contribute substantially to the model’s overall performance.”
Is this a citation from somewhere/someone? I'm wondering because it is written with quotation marks.
gender
I recommend to use the term sex here instead. In this instance, the sex (i.e., female or male) was determined by genotyping of sex chromosomes and no social-cultural context is looked at here.
2021.
in four Norwegian counties (i.e., Troms and Finnmark, Trøndelag, Nordland, Innlandet).
able 1:
Is there something to be shown in the column called False binding (dG)?
Fall (128), followed by Summer (44) and Spring (27).
fall, summer and spring (i.e., no capital letters)
DBI, RSQLite, dplyr, lubridate, sp, mapview, glmmTMB, DHARMa, performance, MuMIn, emmeans, car, ggplot2, and kableExtra.
Add versions and relevant references to tese packages, if available. Add references to reference list.
I assume that you will provide the full code as an appendix to your thesis? If so, please add link to repository here.
Rangifer tarandus
Rangifer tarandus
Please check that Latin names are written in italics througout the document.
Sheep and Reindeer
sheep and reindeer
other mammals
Add a list of used mammal species here or in a table.
norway
Norway
norway
Norway
s
Scandinavia
using different extraction kits
and DNA extraction was done using different extraction kits
BioEdit
Add version and citation here.
NetPrimer and PrimerPremier5
Add versions and citations here.
predation
I propose a different title - probably something like this (but this can be further refined): Brown bear (Ursus arctos) predation on sheep (Ovis aries) and reindeer (Rangifer tarandus), two culturally and economically important species in northern Norway: a molecular detection approach
Data Production
A summary table on number of hits and replicates per individual and season and year may be interesting here to understand whether there was variation in detection across time or according to season.
PrimerBlast
Add version and citation here for the program used.
Primer Design
This section requires more details. For a detailed overview of the workflow and inclusion criteria for primers from the literature, please refer to van der Pouw Kraan et al. (2024; https://doi.org/10.1111/1755-0998.13945) and their Figure 1 and Table 1. It would be beneficial if you could also report this section accordingly. Currently, this section assumes too much from the reader and is overly generalistic. Also reference Table 1 below here.
DNA extracts were retrieved from a minus eighty-degree freezer, previously used for population genetics studies.
This may be seen as a weakness if not properly explained, as the DNA extraction from scats was mainly focused on obtaining DNA from the source species rather than the dietary DNA itself, no? Please provide a report on the subsampling process for DNA extraction from scats, including whether the approach differed for different types of scats and the specific DNA extraction method employed. This is crucial for determining the extent of random effects that can be incorporated in GLMMs, if one could model the data in its current form.
Bears in Norway: A molecular approach to detect predation
The title reads odd. Please contemplate rephrasing it.
Primer Validation
More detail is needed here, include how you determined the limit of detection of your assay. State that you used standard curves to estimate limit of detection (LOD), but see Klymus et al. (2020). Given that your assays are for qualitative purposes, the limit of quantification (LOQ) is likely not relevant in your case. Please verify this to clarify in the main text why the qPCR efficiency may be irrelevant for your assays, but the LOD is.
Depending on who you will get as an examiner, it may be worthwhile to also mention that you did the testing according to the MIQE guidelines, which I think were incorporated into this paper (see thier Appendix S1 for the checklist):
Thalinger, B., Deiner, K., Harper, L. R., Rees, H. C., Blackman, R. C., Sint, D., ... & Bruce, K. (2021). A validation scale to determine the readiness of environmental DNA assays for routine species monitoring. Environmental DNA, 3(4), 823-836.
Bustin, S. A. (2024). Improving the quality of quantitative polymerase chain reaction experiments: 15 years of MIQE. Molecular aspects of medicine, 96, 101249.
Klymus, K. E., Merkes, C. M., Allison, M. J., Goldberg, C. S., Helbing, C. C., Hunter, M. E., Jackson, C. A., Lance, R. F., Mangan, A. M., Monroe, E. M., Piaggio, A. J., Stokdyk, J. P., Wilson, C. C., & Richter, C. A. (2020). Reporting the limits of detection and quantification for environmental DNA assays. Environmental DNA, 2, 271–282. https://doi.org/10.1002/edn3.29
The PCR run included a hold stage at 95°C for 10 min, followed by 40-45 cycles at 60°C for 90 s (annealing and extension) and 95°C for 15 s (denaturation). Data was collected at the end of each annealing step.
Also state the name of the qPCR instrument that was used to run and collect the data here.
Samples were pipetted from and to 96-well plates, including two negative and two positive controls (5 ng/µl sheep and reindeer DNA mix).
Please include more detail here, including the number of technical replicates used. The consensus is that a sample should be run with a minimum of three technical replicates, although more are generally preferred. For each sample, the number of positive hits would be reported over the number of replicates run. Replicates are essential to your dataset because you are engaged in modeling. The below section is confusing for me in that regard. How do you model a dataset where, in the case of absences, there is a 50% chance of missing a hit due to the limit of detection (LOD) of your assays? Please explain this to the reader. I believe that running your samples once without replication is a fundamental flaw that limits the application of GLMMs on your current dataset.
Most testing assays were normalized to a DNA concentration of 5 ng/µl. The sheep-specific primer set Ovi-JOE showed cross-talk into the ROX channel above 0.8 µM. Based on similar amplification curves for concentrations from 0.5 µM to 0.7 µM, a final concentration of 0.5 µM (forward/reverse) and 0.25 µM (probe) for the Ovi-JOE primer set was chosen to minimize cross-talk. The reindeer-specific primer set Ran-ROX delivered consistent curves across all concentrations, differing only in intensity. Given ROX’s lower intensity compared to JOE, the highest tested concentration of 1.0 µM for forward/reverse and 0.5 µM for the probe in the final reaction was used.
This should be included in a separate section on primer/probe system optimization and validation. I believe there are differing perspectives here on how to test for fluorescence crosstalk or bleedthrough, and the definition thereof. Please clearly define the experimental setup you used to test for crosstalk. Did you follow the approach outline in Grohmann et al, (2021: Guidance document on multiplex real-time PCR methods, EUR 30708 EN, Publications Office of the European Union, Luxembourg, 2021, ISBN 978-92-76-37820-4, doi:10.2760/243914, JRC125188: https://publications.jrc.ec.europa.eu/repository/handle/JRC125188), page 17?
Although the TaqQuest mastermix is not specified as inhibitor-resistant, positive results from previous microsatellite analyses suggested minimal inhibition.
Indeed, you did conduct a test beforehand prior to large-scale work, correct?
The only primer pair retained from the literature was the sheep primer set from Dalsecco (2018), which was slightly modified.
I find this situation quite unusual. Please clarify your selection criteria and also state exactly how you modified this primer.
The reverse primer was reverse complemented due to the G at the 3’ end and used as a probe in the TaqMan assay. New priming sites were manually identified, and potential primers were evaluated using NetPrimer and PrimerPremier5 for Tm, secondary structures, and binding sites. Primers were designed to avoid overlap, targeting the cytB gene.
This information is very difficult to understand. It is not clear which assay or primer/probe system this refers to.
Primer Design
This section requires more details. For a detailed overview of the workflow and inclusion criteria for primers from the literature, please refer to van der Pouw Kraan et al. (2024; https://doi.org/10.1111/1755-0998.13945) and their Figure 1 and Table 1. It would be beneficial if you could also report this section appropriately. Currently, this section assumes too much from the reader and is overly generalistic. Also reference Table 1 below here.
BioEdit
Include references for the software you used.
Aims and hypothesis:
Jumping directly to these research questions without first addressing assay development and validation seems strange, as the assays haven't undergone prior testing in an independent, peer-reviewed study.
Primer triplets
"primer/probe systems" - Hopefully, the Introduction section will cover the TaqMan assay's principle.
How does seasonality affect the likelihood of detecting sheep and reindeer DNA in brown bear fecal samples?
Expand this section to include a sentence or two that formulates a testable hypothesis based on previous research. The same goes for the second question below.
Sample Selection
This section requires more work. Please provide a clear explanation of your selection criteria for all the samples included in the analysis. Your sampling logic must be understood by the examiner to determine its suitability for answering your research questions. It will also be beneficial to relocate the "Data Management" section below and integrate it here for a more coherent flow of logic.
Tissue samples from other mammals were collected opportunistically, with some, such as moose or reindeer, donated by hunters, while others were used in previous studies. These tissue samples were utilized for target and non-target experiments during the testing phase and served as positive controls in fecal analysis.
Please provide a sentence or two explaining the selection of these species used to test the cross-reactivity of the primer/probe systems, along with the actual names of the panel of mammals used in this study.
Specifically, this study seeks to answer two key questions:
Addressing these questions largely depends on the precision of newly developed qPCR assays, which should also be stated here, at least as an objective.
Bears in Norway: A molecular approach to detect predation
The title reads odd. Please contemplate rephrasing it.
DNA extracts were retrieved from a minus eighty-degree freezer, previously used for population genetics studies.
This may be seen as a weakness if not properly explained, as the DNA extraction from scats was mainly focused on obtaining DNA from the source species rather than the dietary DNA itself, no? Please provide a report on the subsampling process for DNA extraction from scats, including whether the approach differed for different types of scats and the specific DNA extraction method employed. This is crucial for determining the extent of random effects that can be incorporated in GLMMs, if one could model the data in its current form.
Primer Validation
More detail is needed here, include how you determined the limit of detection of your assay. State that you used standard curves to estimate limit of detection (LOD), but see Klymus et al. (2020). Given that your assays are for qualitative purposes, the limit of quantification (LOQ) is likely not relevant in your case. Please verify this to clarify in the main text why the qPCR efficiency may be irrelevant for your assays, but the LOD is.
Depending on who you will get as an examiner, it may be worthwhile to also mention that you did the testing according to the MIQE guidelines, which I think were incorporated into this paper (see thier Appendix S1 for the checklist):
Thalinger, B., Deiner, K., Harper, L. R., Rees, H. C., Blackman, R. C., Sint, D., ... & Bruce, K. (2021). A validation scale to determine the readiness of environmental DNA assays for routine species monitoring. Environmental DNA, 3(4), 823-836.
Bustin, S. A. (2024). Improving the quality of quantitative polymerase chain reaction experiments: 15 years of MIQE. Molecular aspects of medicine, 96, 101249.
Klymus, K. E., Merkes, C. M., Allison, M. J., Goldberg, C. S., Helbing, C. C., Hunter, M. E., Jackson, C. A., Lance, R. F., Mangan, A. M., Monroe, E. M., Piaggio, A. J., Stokdyk, J. P., Wilson, C. C., & Richter, C. A. (2020). Reporting the limits of detection and quantification for environmental DNA assays. Environmental DNA, 2, 271–282. https://doi.org/10.1002/edn3.29
Most testing assays were normalized to a DNA concentration of 5 ng/µl. The sheep-specific primer set Ovi-JOE showed cross-talk into the ROX channel above 0.8 µM. Based on similar amplification curves for concentrations from 0.5 µM to 0.7 µM, a final concentration of 0.5 µM (forward/reverse) and 0.25 µM (probe) for the Ovi-JOE primer set was chosen to minimize cross-talk. The reindeer-specific primer set Ran-ROX delivered consistent curves across all concentrations, differing only in intensity. Given ROX’s lower intensity compared to JOE, the highest tested concentration of 1.0 µM for forward/reverse and 0.5 µM for the probe in the final reaction was used.
This should be included in a separate section on primer/probe system optimization and validation. I believe there are differing perspectives here on how to test for fluorescence crosstalk or bleedthrough, and the definition thereof. Please clearly define the experimental setup you used to test for crosstalk. Did you follow the approach outline in Grohmann et al, (2021: Guidance document on multiplex real-time PCR methods, EUR 30708 EN, Publications Office of the European Union, Luxembourg, 2021, ISBN 978-92-76-37820-4, doi:10.2760/243914, JRC125188: https://publications.jrc.ec.europa.eu/repository/handle/JRC125188), page 17?
Samples were pipetted from and to 96-well plates, including two negative and two positive controls (5 ng/µl sheep and reindeer DNA mix).
Please include more detail here, including the number of technical replicates used. The consensus is that a sample should be run with a minimum of three technical replicates, although more are generally preferred. For each sample, the number of positive hits would be reported over the number of replicates run. Replicates are essential to your dataset because you are engaged in modeling. The below section is confusing for me in that regard. How do you model a dataset where, in the case of absences, there is a 50% chance of missing a hit due to the limit of detection (LOD) of your assays? Please explain this to the reader. I believe that running your samples once without replication is a fundamental flaw that limits the application of GLMMs on your current dataset.
The PCR run included a hold stage at 95°C for 10 min, followed by 40-45 cycles at 60°C for 90 s (annealing and extension) and 95°C for 15 s (denaturation). Data was collected at the end of each annealing step.
Also state the name of the qPCR instrument that was used to run and collect the data here.
Although the TaqQuest mastermix is not specified as inhibitor-resistant, positive results from previous microsatellite analyses suggested minimal inhibition.
Indeed, you did conduct a test beforehand prior to large-scale work, correct?
The reverse primer was reverse complemented due to the G at the 3’ end and used as a probe in the TaqMan assay. New priming sites were manually identified, and potential primers were evaluated using NetPrimer and PrimerPremier5 for Tm, secondary structures, and binding sites. Primers were designed to avoid overlap, targeting the cytB gene.
This information is very difficult to understand. It is not clear which assay or primer/probe system this refers to.
Primer Design
This section requires more details. For a detailed overview of the workflow and inclusion criteria for primers from the literature, please refer to van der Pouw Kraan et al. (2024; https://doi.org/10.1111/1755-0998.13945) and their Figure 1 and Table 1. It would be beneficial if you could also report this section appropriately. Currently, this section assumes too much from the reader and is overly generalistic. Also reference Table 1 below here.
The only primer pair retained from the literature was the sheep primer set from Dalsecco (2018), which was slightly modified.
I find this situation quite unusual. Please clarify your selection criteria and also state exactly how you modified this primer.
BioEdit
Include references for the software you used.
Primer triplets
"primer/probe systems" - Hopefully, the Introduction section will cover the TaqMan assay's principle.
Sample Selection
This section requires more work. Please provide a clear explanation of your selection criteria for all the samples included in the analysis. Your sampling logic must be understood by the examiner to determine its suitability for answering your research questions. It will also be beneficial to relocate the "Data Management" section below and integrate it here for a more coherent flow of logic.
Tissue samples from other mammals were collected opportunistically, with some, such as moose or reindeer, donated by hunters, while others were used in previous studies. These tissue samples were utilized for target and non-target experiments during the testing phase and served as positive controls in fecal analysis.
Please provide a sentence or two explaining the selection of these species used to test the cross-reactivity of the primer/probe systems, along with the actual names of the panel of mammals used in this study.
How does seasonality affect the likelihood of detecting sheep and reindeer DNA in brown bear fecal samples?
Expand this section to include a sentence or two that formulates a testable hypothesis based on previous research. The same goes for the second question below.
Specifically, this study seeks to answer two key questions:
Addressing these questions largely depends on the precision of newly developed qPCR assays, which should also be stated here, at least as an objective.
Aims and hypothesis:
Jumping directly to these research questions without first addressing assay development and validation seems strange, as the assays haven't undergone prior testing in an independent, peer-reviewed study.
Here, we use non-invasive fecal DNA analysis combined with generalized linear mixed models (GLMMs) to explore the factors influencing the presence of sheep and reindeer DNA in bear feces.
Testing this hypothesis largely depends on the precision of newly developed qPCR assays, which should also be stated here, at least as an objective.