All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
Dear Robert,
As your submission has been initially classified as "major revisions" your revised manuscript was sent for re-evaluation to all reviewers again. As you can see in the comments below, your revised version satisfied two of the reviewers completely and I agree with their opinion. Therefore I decided to accept the manuscript without further revision.
No Comments
No Comments
No Comments
As the author has chosen to completely ignore my main concern regarding the structure of the manuscript (which is still think is a major issue), I have no further comments on the revised manuscript.
No Comments
No Comments
My comments/claims were satisfactorily answered in the revised version
My comments/claims were satisfactorily answered in the revised version
My comments/claims were satisfactorily answered in the revised version
Dear Robert,
Thank you for submitting your manuscript for publication in PeerJ. It has been examined by three expert reviewers who have concluded that the work is potentially sound and suitable for publication as the evolution of MassyPup64 is an important development and the manuscript offers an overview of the improved platform and presents specifically designed workflows that are useful to the community; however, it appears that all three reviewers consider the manuscript not yet clear enough especially for a reader that is unfamiliar with the software and still missing relevant information, which means that a revision will be needed prior to its further consideration for publication.
Please see the enclosed reviewer's reports for details regarding the requested changes and/or additions.
When preparing your revision, please focus mainly on the style and organization of the paper to improve clarity and the understanding what is included and what it can be used for. I consider the suggestion of reviewer 1 and reviewer 3 to include a full page table listing of the software included to be an idea that you should consider.
Please also take very serious the concern of reviewer one, mentioned in the "Validity of findings" about addressing the extremely high false-discovery rate.
Please address either in your revised manuscript the concerns/remarks raised by all three reviewers or in case you have objections to a specific recommendation please present your arguments in a point-to-point response.
Review of: An Evolving Computational Platform for Biological Mass Spectrometry: Work-flows, Statistics and Data Mining with MASSyPup64
Getting a large number of packages to work together on a single operating system is a daunting task. Configuring specific compilers, interpreters, and organization of permissions is often not achievable by typical scientists. The evolution of MassyPup64 is an important development to encourage distribution of tools that can seemlessly work together.
Recommended revisions:
Original data is almost always in a proprietary format from a vendor. Elaborate on line 44 which formats are compatible with msconvert via the MassyPup64 Linux distribution and which are not. Its my naieve understanding that certain vendors have released their dlls in such a way that they are compatible with Wine and other Linux .net emulators. This is an excellent opportunity to identify the current state of affairs for raw data import for each vendor.
imzml/ibd files are mentioned in figure 1, but there is no mention of mass spectrometry imaging other than line 234 "MSI.R for evaluating MSI data." There are numerous software packages for processing imzml, and they aren't mentioned. This should be corrected unless there is a reason that they can't be installed on MassyPup64?
Python is mentioned briefly on line 205 and there is absolutely no mention of the IPython interface. Including the Python packages for mass spectrometry, computational chemistry, and scientific computing packages needed for plotting, statistics, and file import/export seems like a major oversight that could be easily overcome.
It isn't clear from reading the manuscript if an MSMS reference database (ie: from mass bank or Metlin) and a search algorithm for comparing a measured MSMS spectrum to a reference database is included.
It isn't clear if mzmine is included in this release. For example, mzmine is mentioned on line 162. Is it included? Is Java included? Is the java version compatible with other java apps in MassyPupp64?
My general impression is that many valuable tools are bundled together in this linux distribution, but the style and organization of the paper in its current form makes it very difficult to tell exactly what is included and what it can be used for. A full page table listing the software included would make this much clearer.
Not applicable to this manuscript
I was surprised to see the mention of putative-identification on line 503-505. I did not see any mention of the consolidation of features into pure-spectra where adducts, isotopes, and degradation products can be identified. It is a concern that this this manuscript advocates for accurate mass based assignments of features to compounds independent of pure-spectra and MSMS. This may not be the viewpoint of the author, but concluding the paper on this point makes it seem like a highlight. Addressing the extremely high false-discovery rate in this approach is necessary, at a minimum.
The manuscript by Robert Winkler details a computational platform for mass spectrometry analysis called MASSyPup64, which is an evolution on the already published MASSyPup platform. But while the work on providing easy access to open source bioinformatics software is highly commendable, the manuscript in its current form leaves me rather confused.
To me it seems like the author is trying to write multiple manuscripts in one, and I think that readers not familiar with most of the software used will be unable to extract much from the text. I'd recommend restructuring and refocusing the manuscript into more of a technical brief kind of manuscript, with a lot of the current details and examples moved to either a supplementary or the MASSyPup64 web page. I think this will make it much easier for a reader that is unfamiliar with the software to understand how he or she can utilize the significant efforts put into MASSyPup64.
No Comments
No Comments
Abstract is too long and ought to be shortened?
Introduction: Please remove the quotes and capital first letters for Omics, Genomics, Proteomics, etc.
The text ought to be inspected in detail and minor errors corrected. For example:
Abstract: Remove white space in the link: "http://www. bioprocess.org/massypup/" -> "http://www.bioprocess.org/massypup/".
Introduction, line 30, typo: "mass spectrometry experiments follow the all same logic" > "mass spectrometry experiments all follow the same logic"
Figure 1: I don't understand why R has been given such a big focus in the figure? After all this is just one option for the data analysis? And why is the Model Building cogwheel not connected to the others?
Raw Data Import, line 40, typo: "it is recommendable to execute" > "it is recommend to execute"
Raw Data Import, line 49, typo: "it is advisable to convert profile spectra" > "it is advised to convert profile spectra"
The manuscript presents a linux platform that already includes a large number of tools for mass spectrometry data analysis.
The author provides a comprehensive overview of the available software and focuses especially on further downstream ways of data analysis such as predictive models. As there already exists a paper presenting the platform, the author goes here deeper on tools for "data mining" and proposes example workflows for further data analysis.
The manuscript offers an overview of the improved platform and presents specifically designed workflows that are useful to the community. However, it still misses relevant information and clarity. Therefore, I recommend re-submission a revised version.
Issues and comments:
* The manuscript frequently states the potential of data mining for proteomics and metabolomics analysis. I am though missing a clear definition and examples of what the author means by data mining. Data mining is a very broad notion and involves practically everything from data analysis and statistics to data interpretation.
* Is there a way for upgrading to the new releases of MASSyPup? The project seems to be quite vivid with regular changes and improvements and a simple procedure could be very useful.
* line 218: "are exceed" -> "exceed"
* lines 230-231 seem to be obsolete
* line 318: "Association analysis": It is not completely clear what the study is aimed for. I suggest introduction of the biological question at the beginning. In addition, what are the input parameters? Do they include peptide FDRs?
* Data mining would include mining publicly available data. Does the package include software for GO term analysis, network/pathway analysis? Is Cytoscape installed?
* What about post-translational modifications (PTMs)? MS data often comprises large amounts of PTM measurements and they were not discussed in the manuscript. Does MASSyPub contain softwares to calculate false localization rates of PTMs on peptides? I can recommend a recently published review of PTM analysis: http://www.ncbi.nlm.nih.gov/pubmed/26216596
* What about integration with other data sources such as transcriptomics? Data mining would include retrieving additional information from databases such as Uniprot.
* There are many more R libraries applicable to MS data than presented in the manuscript, see https://www.bioconductor.org/packages/release/BiocViews.html#___MassSpectrometry. Are they installed in the platform?
* What about tools for motif search, extraction and discovery?
* I strongly recommend providing a table with all installed tools so the user can check for already available tools.
* The platform contains TPP and OpenMS. What about ProteoSaFe?
* The author applies k-means clustering amongst other techniques. Fuzzy c-means is heavily used in proteomics studies and should be worth a try, maybe yielding better separation of the chicken groups.
* The model building section (targeted metabolomics data) presents nicely different prediction models. I don't understand why one has to call the analysis "based on data mining". What is the relation to data mining here? I would rather call the analysis "predictive data modeling".
Generally, it is not clear what the author specifically means with data mining.
see above
see above
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.