JOSS Review

[Vini2]

Hello! I'm creating this issue as part of the review of your submission to JOSS at openjournals/joss-reviews#9066. This is a work-in-progress so I will keep adding comments. Please don't close this issue until the points have been addressed.

The authors present YACHT, a tool to accurately determine the presence/absence of genomes (from a reference database) in a metagenomic sample by incorporating average nucleotide identity (ANI) and sequencing coverage. It uses a k-mer based sketching technique, which avoids having to align samples to reference genomes makinf fast and efficient to process large datasets. The tool is well-implemented and I believe that after addressing the following comments, the paper can be accepted for publication in JOSS.

Major comments

1. The title should be changed so it won't be confused with the original Bioinformatics publication.
2. Could the authors please expand a bit more on the problems the software is designed to solve and who the target audience is in the documentation?
3. The installation of YACHT is straightforward but the documentation should include a clearly-stated list of software dependencies.
4. The manuscript mentions "FrachMinHash" which I believe should be "FracMinHash" introduced in sourmash. Also, it is better to cite the relevant paper for FracMinHash on line 79.
5. Since the paper focuses more on the applications of YACHT, it would be helpful if the authors can expand on the workflow section and describe what inputs are needed for YACHT and what outputs are generated. I see that there is a section called "Output examples" on line 99, similarly you can add a section describing the inputs.
6. Please add what data was used to generate the results in Table 1.
7. Please add figure number and caption for the workflow diagram.
8. Please refer the figure and table in the main text where relevant.

Minor comments

1. The documentation link on anaconda directs to a different software. Please update it in the conda recipe.
2. I noticed later that YACHT does not currently support MacOS after trying it out on an M4 Mac. However, the quick start code worked after I installed wget (not installed by default on macOS). Also there are osx installers for YACHT on anaconda. Perhaps the documentation could be updated clearing these discrepancies. Just want to make sure that YACHT can be installed and works as intended on different OS architechtures.

Because of these points I cannot check all the boxes in the checklist. I'm happy to take a look once they have been addressed.

[bioinfwithjudith]

We thank the reviewers for their insightful and constructive feedback. Below, we have attempted to address each of their concerns in detail:

1. The title should be changed so it won't be confused with the original Bioinformatics publication.
We have changed the title from “YACHT: an ANI-based statistical test to detect microbial presence/absence in metagenomic sample” to “YACHT: Software for an ANI-based statistical test to detect microbial presence/absence in a metagenomic sample” with a slight modification.

While we believe the same title is appropriate given the different publication journals, we are willing to add the term 'Software' to the title to prevent confusion. The original manuscript focuses on the statistical method and algorithm for detecting microbial presence/absence, while the JOSS submission describes the software implementation and real-world use cases.

We believe this approach will best serve the community by maintaining consistency while making the relationship between the two publications transparent. We will also provide explicit citation guidance in our README directing users to cite either or both papers as appropriate:

• Cite the JOSS paper when using the software implementation
• Cite the Bioinformatics paper when referencing the statistical methodology

2. Could the authors please expand a bit more on the problems the software is designed to solve and who the target audience is in the documentation?
We have revised the README file to provide a clear description of the problems YACHT addresses and its intended target audience.

3. The installation of YACHT is straightforward but the documentation should include a clearly-stated list of software dependencies.
We have updated our README file describing the software dependencies in the manual installation section.

4. The manuscript mentions "FrachMinHash" which I believe should be "FracMinHash" introduced in sourmash. Also, it is better to cite the relevant paper for FracMinHash on line 79.
We have revised FracMinHash to FracMinHash and we have added the relevant citation that refers to FracMinHash when first introduced at line 79 (Irber LC, Brooks PT, Reiter TE, Pierce-Ward NT, Hera MR, Koslicki D, Brown CT. 2022. Lightweight compositional analysis of metagenomes with FracMinHash and minimum metagenome covers. bioRxiv 10.1101/2022.01.11.475838 [DOI]).

5. Since the paper focuses more on the applications of YACHT, it would be helpful if the authors can expand on the workflow section and describe what inputs are needed for YACHT and what outputs are generated. I see that there is a section called "Output examples" on line 99, similarly you can add a section describing the inputs.
We have added a concise paragraph following the workflow figure to clarify the expected library inputs. Please understand that the explanation is brief due to the word limit constraint.

6. Please add what data was used to generate the results in Table 1.
The caption for Table 1 has been updated by adding what data was used to generate these results. The data used was the MBARC-26 (SRA: SRR6394747 by Singer et al, 2016) and the YACHT results presented in this table were generated by using the following parameters: k-size of 31, minimum coverage of 0.05 with an ANI threshold of 0.95. For more information on these results, please refer to the following use case examples: https://github.com/KoslickiLab/YACHT/tree/main/use_case_examples/synthetic_metagenome.

7. Please add figure number and caption for the workflow diagram.
We have added the figure number and caption to the diagram, which can be found in the attached PDF draft. Find here: paper-2.pdf

8. Please refer to the figure and table in the main text where relevant.
We have added captions to the figures and tables and referenced them within the main text. Please note that the numbering and formatting are displayed in the rendered PDF version, which we have attached in this response, generated by the JOSS build engine.

Minor comments
1. The documentation link on anaconda directs to a different software. Please update it in the conda recipe.
We have released YACHT v1.3.2 and updated conda. The link should correctly take you to the latest version. We apologize for the inconvenience.

2. I noticed later that YACHT does not currently support MacOS after trying it out on an M4 Mac. However, the quick start code worked after I installed wget (not installed by default on macOS). Also there are osx installers for YACHT on anaconda. Perhaps the documentation could be updated clearing these discrepancies. Just want to make sure that YACHT can be installed and works as intended on different OS architechtures.
We have verified YACHT is compatible and supported on Linux and macOS, which represent the standard operating systems for the majority of bioinformatics workflows.