Ubiquitous microbial contaminants associated with scientific ocean drilling
This article has been Reviewed by the following groups
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Article activity feed
-
-
Thank you for addressing the concerns raised by the reviewers. The work presented is clear and the arguments well formed. This study would be a valuable contribution to the existing literature. This is a study that would be of interest to the field and community.
-
-
Please be careful with the some of the statements made since they might be misinterpreted to not include your own controls when analysing samples (see Line 350 - 354, concern raised by reviewer 2) who also points out the lack of statistical testing in some analyses and is asking for some clarification regarding the data analysis and further details on the methodology is requested by reviewer 3. There are also some further minor comments raised by all of the reviewers that need addressing. Individual Reviewer Comments to Author (Editor's Copy)
-
Comments to Author
The study is both relevant and interesting and has addressed the issue of identifying common microbial contaminants associated with ocean drilling, as the title depicts, by proffering a database of these contaminants for use as polymerase chain reaction (PCR) control. It has also gone further to test these controls for suitability. The use of no template negative controls has been emphasised. 1. Methodological rigor, reproducibility and availability of underlying data. The PCR methodology for six samples was explicit and reproducible. However, the methodology for the four studies (Line 138) needs to be explained. The results state that 313 samples from the four studies were identified as negative controls. Were physical samples retrieved or were sequences from the literature or repository used as …
Comments to Author
The study is both relevant and interesting and has addressed the issue of identifying common microbial contaminants associated with ocean drilling, as the title depicts, by proffering a database of these contaminants for use as polymerase chain reaction (PCR) control. It has also gone further to test these controls for suitability. The use of no template negative controls has been emphasised. 1. Methodological rigor, reproducibility and availability of underlying data. The PCR methodology for six samples was explicit and reproducible. However, the methodology for the four studies (Line 138) needs to be explained. The results state that 313 samples from the four studies were identified as negative controls. Were physical samples retrieved or were sequences from the literature or repository used as negative control? Also, would this be a negative or positive control? This aspect of the methodology should be explained further to aid reproducibility. Was the same PCR methodology used for all samples? The methodology (Line 142) states that six samples were received from the Kochi Repository (Table 1). There is no mention of the six sediment samples from the Japan Trench collected during IODP in the methods. However, it is mentioned in the results in Line 263. Are these samples the same? If not, it would be good to discuss it in the methods. And if the same, it would be good to state it. Overall, the methodology is good. What is required is a chronological explanation based on methodological sequence. It would be good to know the number of samples from each source and how they were analysed prior to the detailed methodology to maintain a good flow of thoughts. 2. Presentation of Results The results have been well presented. However, there are two different tables titled Table 1 which need to be labelled Tables 1 and 2. The table titled ''Table 1: Taxonomic assignment of the 25 ASVs that occurred in the most samples from Control Groups I and II'' has not been referred to in the results text. It could be mentioned in line 224 and renamed Table 2. Table 3 should then be renamed Table 3 and updated in the text. Would have been good to see a list of the order found in Control group 111 if available. 3. How the style and organization of the paper communicates and represents key findings The paper is well organised, and the key findings have been addressed in the discussion. 4. Literature analysis or discussion There are good citations of relevant literature. However, a couple more would be beneficial especially in lines 73 and 364. Overall, the literature analysis is good. The discussion is very good and captures the essence of the study while maintaining clarity and including relevant citations. However, it would be good to have a conclusion. Lines 388 to the end could be captioned Conclusion and Recommendation and a sentence or two included to conclude the discussion. Any other relevant comments Would be good to write names in full for the first time before using abbreviations like PCR and ASVs. Also, Line 25 - Preferable to use 'by' once. ` Lines 25 - 27 -More clarification needed. Do you mean that tracers were used to test drilling fluids to confirm their presence in the MBIO samples? Line 31 - what is referred to as contamination controls? Is it positive control? Clarification would be beneficial for reproducibility. From the graphical abstract, would it be that you have built a database of common contaminants and produced positive controls which were then used for the screening of 16S rRNA gene amplicon datasets from past IODP expeditions? Line 58 - 1.0 x 109 (109) seems high and the referenced literature has an even higher value of 1029 Is this the value you meant to quote? It would be good to note also that sediment depth is an important factor for variations in biomass. Lines 69 - 73 - Would be good to see citations here. Lines 70 - 73 - Same as lines 25 - 27 above Lines 76 - 81 - Words in brackets could be explained as separate sentences for clarity. Line 107 - Do you mean frozen core samples are used mostly for molecular analysis? Line 111 - using negative controls Line 138 - Four studies fulfilled all 4 criteria - would be good to introduce what will happen next regarding these studies before discussing the next 6 samples. Line 191 - Good to include 'respectively for each of the groups' after the word reads. Line 252 - Sentence can be rephrased to exclude 'here'. Line 323 - Do you mean exclude common contaminants or remove them? Line 337 - Here could be replaced with 'In this study'. Line 349 - would read better as - if not all, of the contaminants --- Line 360 - 361 - could read Although it was not possible ----- Line 364 - Citation would be relevant Line 381 - Could clarify how this was demonstrated. Line 384 - Could do without the brackets. For example: sterile tools such as autoclaved spatulas, sterilising caps by spraying with 70 % (v /v) ethyl alcohol before sealing and placing in ----------. Line 384 and 386 - 70 % (v /v) ethyl alcohol is preferable
Please rate the manuscript for methodological rigour
Very good
Please rate the quality of the presentation and structure of the manuscript
Good
To what extent are the conclusions supported by the data?
Strongly support
Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?
No
Is there a potential financial or other conflict of interest between yourself and the author(s)?
No
If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?
Yes
-
Comments to Author
This paper is concerned with examining potential contaminants in samples obtained during ocean drilling. The authors make a strong case for the need to ensure that reported community analysis of sediment samples represent the community in the sediment and not contaminants from sample processing. The authors propose a number of sensible recommendations (Line 389 -394) for sample processing in future studies but it is not clear to what extent these are already being applied in studies. The authors build a database of common contaminants from the various stages in sample processing which is very interesting but I am not sure how this can be used. Just because an organism is typically found as a contaminant doesn't discount its possible presence in an uncontaminated sample. The authors themselves state on …
Comments to Author
This paper is concerned with examining potential contaminants in samples obtained during ocean drilling. The authors make a strong case for the need to ensure that reported community analysis of sediment samples represent the community in the sediment and not contaminants from sample processing. The authors propose a number of sensible recommendations (Line 389 -394) for sample processing in future studies but it is not clear to what extent these are already being applied in studies. The authors build a database of common contaminants from the various stages in sample processing which is very interesting but I am not sure how this can be used. Just because an organism is typically found as a contaminant doesn't discount its possible presence in an uncontaminated sample. The authors themselves state on Line 359 that "it remains possible that these microorganisms are endemic to the subsurface due to their ubiquitous nature" I am particularly concerned by the statement on Line 350 - 354 " While this database should not be used in lieu of controls, it could be used as a complement to controls, or when controls are unavailable, to determine which taxa are potential contaminants from handling and molecular reagents, but also seawater infiltration or drilling fluids" This seems to me that the authors are suggesting that samples can be analysed in the absence of controls which is not appropriate. Other comments It would have been useful to statistically analyse the data. Are the differences in the data in Figs 2-5 significantly different? More detail is needed on how the data from the 4 studies were analysed. For example, two of the studies (refs 11 and 18) refer to ASVs but the studies in refs 12 and 17 refer to OTUs. How did the authors take account of this? Line 288 - More detail is needed on how the authors tested the database on the two published IODP studies that did not have controls Line 376 - Details are needed on how the authors "manipulated" the IODP MBIO Legacy samples Line 194 - More clarity is needed for the statement "In general, the number of sequenced reads did not impact the number of identified ASVs, i.e., the samples with fewer reads did not generate fewer ASVs" Is this a comment on diversity? Line 205 - Were the 2 most commonly detected ASVs found in samples from all 4 studies or were they associated more with samples from 1 of the studies? Line 395 - Are the authors proposing that future studies contain lists of organisms detected in their controls? I am unclear how they propose that "potential contaminants should be labelled as such" The bibliography needs to be checked as many references are incomplete. For example Line 474 ref #17 should have the journal name, volume and pages and not a link to the journal rather than the specific paper. Also Line 476 ref #18 should have the journal name, volume and pages rather than "epub ahead of print" The paper was published in 2016!
Please rate the manuscript for methodological rigour
Poor
Please rate the quality of the presentation and structure of the manuscript
Satisfactory
To what extent are the conclusions supported by the data?
Partially support
Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?
No
Is there a potential financial or other conflict of interest between yourself and the author(s)?
No
If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?
Yes
-
Comments to Author
First of all, my apologies for my slow review (holidays). In any case, I think this is an important paper as increasing awareness even more of the issue of contamination is needed. The authors of this manuscript are expertly aware of the many sources of contamination that may affect sequencing results and the problems associated with low microbial biomass samples. They certainly quickly identify some of the most common well-known contaminant groups (Ralstonia/Burkholderiales/Pseudomonadales/Xanthomonadales/Sphingomonadales). These are certainly contaminants (seen over and over again in a multitude of other studies). I do have a few questions/comments that may improve the manuscript further. 1: In regards to the seawater contaminants (group 3), have they investigated whether when using fluorescent …
Comments to Author
First of all, my apologies for my slow review (holidays). In any case, I think this is an important paper as increasing awareness even more of the issue of contamination is needed. The authors of this manuscript are expertly aware of the many sources of contamination that may affect sequencing results and the problems associated with low microbial biomass samples. They certainly quickly identify some of the most common well-known contaminant groups (Ralstonia/Burkholderiales/Pseudomonadales/Xanthomonadales/Sphingomonadales). These are certainly contaminants (seen over and over again in a multitude of other studies). I do have a few questions/comments that may improve the manuscript further. 1: In regards to the seawater contaminants (group 3), have they investigated whether when using fluorescent microspheres, whether the concentration of those correlate properly with ocean water associated microbial contaminants? 2: "Because a majority of microorganisms in the environmental and contamination control samples belong to the Proteobacteria phylum, more specifically to the Alphaproteobacteria class, we opted for the order level, which allowed for confident taxonomic assignment and separation of likely contaminants from likely endemic microorganisms. " --> While the taxonomic identification down to a species-like level is often not possible for many environmental bacteria (much more so than for human-associated bacteria) you can still accurately perform analysis on the ASV level, and by publishing the ASVs and the numbers of reads per ASV per sample you can future-proof your analyses as such a method is reference library independent. Going to the order level only makes sense if you're comparing different 16S amplicon sequencing results from different studies that utilize different 16S regions (and you can use order-level-summary analyses of the ASV analyses). 3: One can also often recognize reagent contamination from other signals by performing within-batch consistency pattern analyses (See: Recognizing the reagent microbiome Nature Microbiology). Often, one can identify the contaminants without even using negative controls (it is recommended however that you still use as many sorts of negative controls as possible in combination with other methods --> use as many lines of defense as possible --> achieving concordance between methods for identifying contaminants is very useful). I'd recommend trying this approach as well to see how well it matches up with your current contamination identification analyses. 4: What I'm not seeing yet is clear verifications of genuine microbial signals from these cores. Identifying large contaminants is easy, identifying all contaminants is hard (especially for the many small signals) and identifying genuine signals is potentially hard. Can you give examples of signals that are genuine, and the reasons why you think they are genuine (showing the relevant patterns). Can you show an inverse correlation with known contaminants? Can you highlight ecological plausibility & that similar findings have been found in other studies (that have taken contamination into account)? Often reagent contaminants are negatively correlated with the DNA concentration after DNA isolation, though this will not be the case for ocean contaminants (which you can however identify by association using enough samples using fluorescent spheres) and genuine signals. 5: From your discussion: Can you also show that the amount of contamination increases using your data as samples are taken at lower levels below the sea-floor? 6: Discussion: WGS sequencing just like 16S sequencing similarly has contamination problems. Yes, it does not suffer from primer bias but it does suffer from the fact that it is reference library dependent (unlike an ASV-based approach) and a few other problems. Ideally, you perform both 16S and WGS and look for concordance to give you yet another very powerful tool to identify particular sets of contaminants. Of course, doing both costs more money (just like doing proper controls costs money).
Please rate the manuscript for methodological rigour
Very good
Please rate the quality of the presentation and structure of the manuscript
Good
To what extent are the conclusions supported by the data?
Strongly support
Do you have any concerns of possible image manipulation, plagiarism or any other unethical practices?
No
Is there a potential financial or other conflict of interest between yourself and the author(s)?
No
If this manuscript involves human and/or animal work, have the subjects been treated in an ethical manner and the authors complied with the appropriate guidelines?
Yes
-