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.
Thank you for your valuable contribution.
The authors have addressed all of the reviewers' comments
[# PeerJ Staff Note - this decision was reviewed and approved by Mehmet Cunkas, a PeerJ Section Editor covering this Section #]
The manuscript is in better shape.
some English minor issues:
-threshold greater => threshold greater
-method encompass => method encompasses
proofread it please!
Design is fine
Black Race validation is implemented
Please check the following comments:
They need to prepare an HQ image for Figure 6.
Validity of the findings
The authors managed to address most of my comments and suggestions; however, the conclusion section still requires improvement.
The authors had also added explainable AI LIME results.
Additional comments
The revised paper is now in a better shape that meets the publishing standards.
General English proofreading should be performed. (For example, on lines 312–314, please ensure you use the plural form for "authors.")
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
The authors addressed the reviewers' comments. The manuscript is in very good shape now
Well designed.
Validated using machine learning performance metrics.
Basic English writing is ok.
The study could be improved with the use of more diverse datasets.
-
• The authors propose to use Gaussian naive Bayes for prostate cancer detection using gene expression data.
• The manuscript is well-written and generally well-organized.
• The introduction is adequate, and the authors now have more references from (2024-2025).
The authors managed to address all my comments and suggestions.
They just need to prepare an HQ image for Figure 6.
The authors managed to address most of my comments and suggestions; however, the conclusion section still requires improvement.
The authors had also added explainable AI LIME results,
The revised paper is now in a better shape that meets the publishing standards in my opinion.
General English proofreading should be performed. (For example, on lines 312–314, please ensure you use the plural form for "authors.")
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
**PeerJ Staff Note:** It is PeerJ policy that additional references suggested during the peer-review process should only be included if the authors agree that they are relevant and useful.
**Language Note:** When you prepare your next revision, please either (i) have a colleague who is proficient in English and familiar with the subject matter review your manuscript, or (ii) contact a professional editing service to review your manuscript. PeerJ can provide language editing services - you can contact us at [email protected] for pricing (be sure to provide your manuscript number and title). – PeerJ Staff
The study aims to improve prostate cancer detection in White patients using gene expression data and an optimized Gaussian Naïve Bayes (GNB) machine learning model, while also exploring racial disparities in diagnosis. I have the following comments:
- The literature may add more related works using machine learning (including standard classifiers such as Bayesian) in prostate cancer. I suggest reading:
https://doi.org/10.1177/1176935119835522
https://doi.org/10.3390/diagnostics9040219
**PeerJ Staff Note:** The PeerJ's policy is that any additional references suggested during peer review should only be included if the authors find them relevant and useful.
- Strong use of DEG analysis and ROC thresholds (AUC > 0.9) to prioritize genes. However, the rationale for log2foldchange thresholds (0.35 vs. 0.4) lacks biological justification.
- Can the authors apply XAI (SHAP and LIME) for interpretability?
- Justification why the upsampling techniques were considered (e.g, why SMOTE, not ADYSYN)?
-I would appreciate it if the manuscript had a criterion to check whether the model is overfitting/underfitting.
-Validation of the selected genes can be done through KEGG pathway analysis and/or GO enrichment.
-
-
-
The language is clear, and the formal results are presented effectively. However, the study lacks novelty, as all methods employed are already well-established in the fields of gene selection and machine learning.
A detailed description of the DEG analysis procedure is essential to ensure reproducibility and to validate the findings.
The conclusion section must be revised to explicitly address the study’s limitations.
The study should be validated on additional cancer types, and a comparative analysis using alternative machine learning models is necessary to demonstrate robustness.
Given the imbalance in the dataset used, it is important to evaluate the proposed methods on both balanced and imbalanced datasets to assess generalizability.
• The authors propose to use Gaussian naive Bayes for prostate cancer detection using gene expression data.
• The manuscript is well-written and generally well-organized.
• The introduction is adequate, but the literature review must include more recent references regarding the topic (only 2 references from the last two years).
• The title says it’s for white patients, but the simulations show both black and white patients’ results.
• Not all the combinations shown in Table 2 are used in simulations. SOMTEEN and Borderline SMOTE, for example.
• Tables 4, 5, 6, and 7 all show different splitting ratios. Be consistent.
• The last 2 rows in Table 5 are the same. It’s a repetition. Delete it.
• Step 12 in Figure 1 shows “AI model “. There is no AI presented in this work (you should!)
• Provide the black race samples’ statistics as in Figure 2.
• While the manuscript is well-written and generally well-organized, I believe that the contribution of this work is very limited. Work is kind of an ablation study comparing the training/testing splitting ratio, augmentation techniques, and hyperparameters.
• The results section requires a more in-depth analysis and justification beyond simply restating the values from the tables. Please provide a more comprehensive interpretation of the findings and discuss their significance.
The contribution of the authors in the manuscript is very limited, and it requires significant revision and modification before it is suitable for publication in this journal.
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.