Denovo-GCN: De Novo Peptide Sequencing by Graph Convolutional Neural Networks

Round 1

Reviewer 1 Report

The manuscript by Wu et. al. entitled “Denovo-GCN: De Novo Peptide Sequencing by Graph Convolutional Neural Networks” describes a de novo sequencing method, Denovo-GCN. The authors argue that their results indicate Denovo-GCN significantly outperforms DeepNovo with a relative improvement of 13.7-25.5% in terms of the peptide-level recall.

Points

1. An earlier paper as cited by the authors (ref 17) details how deep neural network model can enable de novo sequencing with high accuracy. The data presented in Fig 3 does not indicate a significant or spectacular improvement over DeepNovo. It is certainly comparable and moderately better in certain species.

2. Data in Fig 4 also indicate only moderate improvement over DeepNovo.

3. The authors should show with specific examples as done in ref 17 (supp Figs S 5,6) that the method described in the paper is significantly better than the one described in ref 17.  

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The article "Denovo-GCN: De Novo Peptide Sequencing by Graph Convolutional Neural Networks" proposes a method based on Graph Convolutional Neural Networks and Convolutional Neural Networks for de novo peptide sequencing. 

The paper's title is informative and matches with the main topics listed in the abstract section. The keywords fit with the article's topic. The references are also appropriately cited. However, authors are invited to perform minor revisions in the final version of the manuscript in order to get it accepted for publication.

- There are some typos. Please use “spelling check tool” before you submit the paper.

- The author couldn't use abbreviation in abstract section.

- Authors should add name of evalation metrics under the title of Data sets and Evaluation metrics. They should add equation. In addition, they should evaluate the success of the model with at least 3 performance criteria and present it in a table.

- Authors should be expanded references list via reference list from this journal.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript " Denovo-GCN: De Novo Peptide Sequencing by Graph Convolutional Neural Networks" by Wang et al. is an excellent example of applying GCN and CNN in peptide sequencing. There is a minor comment: As the computational and data processing have too many steps, following the details is difficult for the readers. Therefore authors must add a box-wise protocol to the manuscript at the end of the presentation of the results.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The response by the authors is OK. The method is an improvement, though not significant, as argued by the authors.