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.
This is a well-written manuscript, the authors tried to use deep learning method to solve the problem, the proposed deep network architecture seems to work well on the prediction, achieved a better accuracy than the peer method.
[# PeerJ Staff Note - this decision was reviewed and approved by Paula Soares, a PeerJ Section Editor covering this Section #]
This is a well written manuscript, the authors tried to use deep learning method solve the problem, the proposed deep network architecture seems working well on the prediction, achieved a better accuracy than peer method.
no comment
no comment
no comment
All my concerns and questions have been addressed, and I recommend the acceptance of the paper。
no comment
no comment
no comment
Authors provide a manuscript presenting their works about methylation site prediction for proteins. This is a well written manuscript, the authors tried to use deep learning method solve the problem, the proposed deep network architecture seems working well on the prediction, achieved a better accuracy than peer methods. It would be helpful to improve the following researches of the functional and conformational changes of a protein. However, there still have a few suggestions:
1, As we all know, the position-specific scoring matrix (PSSM) is widely used as an effective representation of protein feature for a variety of protein-related problems. The authors should explain why this feature is not used.
2. The authors have to create a web server or submit all the source codes to github for reproducing the results.
3. The LSTM layer often followed by self-attention mechanism，I think this can help to improve the predictive performance.
4. The draft needs proofread.
Comparative experiments are not sufficient. Please give more results compared with other methods on more datasets.
[# PeerJ Staff Note: Please ensure that all review comments are addressed in a response letter and any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate. It is a common mistake to address reviewer questions in the response letter but not in the revised manuscript. If a reviewer raised a question then your readers will probably have the same question so you should ensure that the manuscript can stand alone without the response letter. Directions on how to prepare a response letter can be found at: https://peerj.com/benefits/academic-rebuttal-letters/ #]
1) Please attention to some word mistakes, such as “swift and effectively” and “performers” in the abstract. Please check the words and grammar of the whole paper again.
2) The Fig. 3 can be more professional and informative drawn by specialized tools, such as PlotNeuralNet.
Comparative experiments are not sufficient. Please give more results compared with other methods on more datasets.
１）I think that fusing CNN and LSTM branch with a weighted operation is more rational. Please give reasons for simple summation.
２）DeepRMethylSite uses one hot encoding, where each amino acid is defined as a 20 length vector, with only one of the 20 bits as 1. And SSMFN uses an embedding layer with 21 neurons to encode different amino acids. What is the difference of those two representations and advantages of SSMFN?
This paper proposes a neural network model which combines CNN and LSTM to extract spatial and sequential information from amino acid sequences, respectively. The writing structure is reasonable, but there are some problems. Please see the detailed comments above.
no comment
no comment
no comment
1. We note that your model achieves good results, but the explanation of the structure of the model is not sufficient, and we hope it can be improved.
2. For Figure2, we would like you to have more visual representations and less textual representations.
3. You can try to center the content in Table1 to make it look more beautiful.
4. The literature review of the manuscript need to be improved by comparing similar papers. Thus, the readers can compare the results of various works and the main novelty of this paper can be easily identified.
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.