Transfer learning with convolutional neural networks for cancer survival prediction using gene-expression data

Guillermo López-García; José M. Jerez; Leonardo Franco; Francisco J. Veredas

doi:10.1371/journal.pone.0230536

Peer Review History

Original SubmissionNovember 29, 2019
16 Jan 2020 Decision Letter - Ruxandra Stoean, Editor PONE-D-19-33176 Transfer learning with convolutional neural networks for cancer survival prediction using gene-expression data PLOS ONE Dear Mr Lopez-Garcia, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. The reviewers argue about the necessity to address the class imbalance problem, the need for a comparison with conceptually similar works, the desire for more information on the feature selection component, a larger description of the technical aspects of algorithm implementation, a clearer demonstration of the originality and impact of the idea. We would appreciate receiving your revised manuscript by Feb 29 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. We look forward to receiving your revised manuscript. Kind regards, Ruxandra Stoean Academic Editor PLOS ONE Journal Requirements: When submitting your revision, we need you to address these additional requirements. 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes Reviewer #2: Yes ******** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #2: Yes ****** 3. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes Reviewer #2: Yes ****** 4. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes ****** 5. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: This is a very interesting work revealing the use of RNA-Seq data for cancer survival prediction in terms of deep learning and conventional ML techniques. I have no doubts about the work done and the methodology that is proposed in terms of transfer learning and survival prediction. Some concerns that need to be addressed before publication by the authors: The class imbalance problem should be handled in different ways for obtained unbiased results. Under sampling, over sampling and ensemble techniques do exist for performing this task. The figures need to be updated in terms of resolution, they are not in the preferred format I think. A comparison table with similar studies-methodologies should be presented in order to clearly highlight the impact of the current methodology and its usage in another studies. Reviewer #2: The article seems to be sufficiently novel and interesting. The main goal is to propose and apply the DL-with-TL methodology to rearrange RNA sequence data by transforming them into gene-expression images. Based on these images CNN can extract high-level features. Additionally, authors investigate whether extracted knowledge contributes to the extraction of features that improve lung cancer prediction. The results are compared with other selected ML approaches. This approach imply promising results about successful prediction of investigated problem. I think that this paper adhere to the journal standards. Presented paper is well organised. Abstract of the article is relevant to its content. It is formulated correctly and relates directly to the substantive content of the article. The article is divided into appropriate sections like introduction with some theoretical background, some historical steps, presentation of investigated approach, materials and methods, gathered research results and summary. It also contains a detailed and thematically relevant bibliography. The technical content of the article is generally understandable. Authors can add information about the result of the selection of features using applied FS methods. How many original features were considered significant. Maybe also describing the technical aspects of the algorithm implementation might be of interest to readers. Methods must be described in sufficient detail for another researcher to reproduce the experiments described. Authors should provide adequate justification that the methods presented in the paper are significantly different from previously presented. ****** 6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review?** For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: No Reviewer #2: No [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step. https://doi.org/10.1371/journal.pone.0230536.r001
Revision 1
27 Feb 2020 Author Response As indicated by the academic editor, a rebuttal letter responding to each point raised by the academic editor and reviewers has been uploaded as separate file and labeled 'Response to Reviewers'. The content of that file will be copied and pasted here: PONE-D-19-33176 Transfer learning with convolutional neural networks for cancer survival prediction using gene-expression data PLOS ONE We would like to thank the editor and the reviewers for their fair comments and suggestions and for giving us the opportunity to improve and to re-submit our paper. In this revised version of our paper we have addressed all the comments pointed out by the reviewers and emphasised the novel contributions beyond our previous related work. COMMENTS FOR REVIEWER #1: We would like to thank this reviewer for her/his comments and suggestions. The class imbalance problem should be handled in different ways for obtained unbiased results. Under sampling, over sampling and ensemble techniques do exist for performing this task. As it has been fairly suggested by this reviewer, we conducted additional experiments in order to handle the severe class imbalance of the gene-expression datasets. Different techniques―random under-sampling, SMOTE and random over-sampling―were employed to re-sample the data before feeding it to the predictive models. We analyzed the performance of the models using each technique, and finally selected the one that maximized the mean CV AUC metric. In this way, random over-sampling procedure applied both during pre-training and fine-tuning phases of the TL approach gave the best results both using CNN and MLNN models. When applied in combination with the traditional ML approach, SMOTE gave the highest AUC values. The revised version of the paper has been updated with the new results obtained using random-oversampling and SMOTE in combination with DL-with-TL approach and the traditional ML approach, respectively. Figure 8 has also been updated. The figures need to be updated in terms of resolution, they are not in the preferred format I think. Following the submission guidelines, all figures were uploaded as separate files. Besides, we used PACE software, as suggested in Figures guidelines, to ensure all figures met PLOS requirements. A comparison table with similar studies-methodologies should be presented in order to clearly highlight the impact of the current methodology and its usage in another studies. To the best of our knowledge, we are the first authors to explore a TL approach using CNN models on gene-expression data to tackle a cancer-survival predictive task. For this reason, the results obtained in this paper are not directly comparable with the results achieved in previous works, as different methodologies are applied as well as completely distinct cancer prediction tasks are tackled. However, we totally agree with this reviewer on the necessity of performing a thorough comparison between our proposed methodology and similar-studies methodologies to lucidly emphasize the impact of our contribution. In this way, the final paragraphs of the Results and Discussion section in the revised version of the manuscript addresses this concern. COMMENTS FOR REVIEWER #2: We would like to thank this reviewer for her/his comments and suggestions. Authors can add information about the result of the selection of features using applied FS methods. How many original features were considered significant. We considered the number of input features to be selected by the FS method as a hyper-parameter of the traditional ML approach. The range of possible values of this hyper-parameter, as well as the possible numbers of features to be extracted by the feature extraction methods, are described in detail in S3 Table. Maybe also describing the technical aspects of the algorithm implementation might be of interest to readers. Methods must be described in sufficient detail for another researcher to reproduce the experiments described. As stated in the revised version of the paper, for reproducing the results obtained in this work, all the data, code and technical descriptions of the algorithm implementation have been made publicly available at https://github.com/guilopgar/GeneExpImgTL In addition, the Supporting Information file contains a detailed description of the hyper-parameter optimization procedure employed to perform automatic tuning of the hyper-parameters of all the models analyzed in this work. Authors should provide adequate justification that the methods presented in the paper are significantly different from previously presented. As it has been fairly suggested by this reviewer, the Results and Discussion section from the revised version of the manuscript contains a detailed description of the novelties and methodological contributions of the current study, as well as a thorough comparison between our proposed methodology and other similar-studies methodologies. Attachments Attachment Submitted filename: lopez_garcia_et_al_IWANN_Response_Reviewers.pdf https://doi.org/10.1371/journal.pone.0230536.r002
3 Mar 2020 Decision Letter - Ruxandra Stoean, Editor Transfer learning with convolutional neural networks for cancer survival prediction using gene-expression data PONE-D-19-33176R1 Dear Dr. Lopez-Garcia, We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements. Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication. Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org. If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. With kind regards, Ruxandra Stoean Academic Editor PLOS ONE Additional Editor Comments (optional): Reviewers' comments: https://doi.org/10.1371/journal.pone.0230536.r003
Formally Accepted
6 Mar 2020 Acceptance Letter - Ruxandra Stoean, Editor PONE-D-19-33176R1 Transfer learning with convolutional neural networks for cancer survival prediction using gene-expression data Dear Dr. Lopez-Garcia: I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org. For any other questions or concerns, please email plosone@plos.org. Thank you for submitting your work to PLOS ONE. With kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Ruxandra Stoean Academic Editor PLOS ONE https://doi.org/10.1371/journal.pone.0230536.r004

Open letter on the publication of peer review reports

PLOS recognizes the benefits of transparency in the peer review process. Therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. Reviewers remain anonymous, unless they choose to reveal their names.

We encourage other journals to join us in this initiative. We hope that our action inspires the community, including researchers, research funders, and research institutions, to recognize the benefits of published peer review reports for all parts of the research system.

Learn more at ASAPbio .