In Silico Drug Screening for Hepatitis C Virus Using QSAR-ML and Molecular Docking with Rho-Associated Protein Kinase 1 (ROCK1) Inhibitors

Response to Reviewer 1 Comments

1. Summary

Thank you very much for taking the time and effort to review the manuscript. The authors of the following study are pleased to find all of the comments and suggestion in the review process as it helped the study to be improved significantly. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the newly re-submitted revised manuscript and files. All of the revisions done on the manuscript are highlighted in red. Thank you again for the detailed review.

2. Questions for General Evaluation

Reviewer’s Evaluation

Response and Revisions

Does the introduction provide sufficient background and include all relevant references?

Yes

Thank you very much reviewer 1 for reviewing the following paper. Point by point responses in each of the comments are detailed below.

Are all the cited references relevant to the research?

Yes

Is the research design appropriate?

Yes

Are the methods adequately described?

Yes

Are the results clearly presented?

Can be improved

Are the conclusions supported by the results?

Yes

3. Point-by-point response to Comments and Suggestions for Authors

Comments 1: The primary driving force for this research is the steadily rising incidence of drug-resistant HCV genotypes, which is encouraging the investigation of novel compounds that target certain enzymes like ROCK1, which has received little research. This work incorporates a drug screening strategy using molecular docking, Absorption, Distribution, Metabolism, and Excretion predictions, and Quantitative Structure-Activity Relationship Machine Learning approaches. The study's findings are very helpful in identifying substances that inhibit ROCK1.

Response 1: Thank you very much the following research study was greatly motivated for the rising resistance of HCV to direct acting antivirals. Also, as the rising innovations in machine learning and AI technology are increasing in our age using applying their methods in drug discovery can be beneficial and cost-effecting for screening out potential compounds we would want to design.

Comments 2: The subject is unique and pertinent to the field. With the use of this integrated strategy, it is possible to effectively screen for certain compounds and choose drug-like candidates that target ROCK1 inhibition in HCV treatment by using the estimated features and qualities of those compounds as guidance. By inhibiting ROCK1, this novel strategy improves comprehension and facilitates the identification of interesting drugs with potential therapeutic efficacy in the treatment of HCV.

Response 2: Thank you again, ROCK1 inhibition in the context of HCV is under-looked and the study tries to probe compounds in vast amounts and effectively screen only specific structures we would like to know.

Comments 3: The authors ought to keep researching the ligands that were produced and showed the highest scores on C2, matching those of the control.

Response 3: The five ligands found were further subjected to comparative analysis of their structure similarity using the cosine method against other drugs detailed in page 30 paragraph 2 lines 639-643; found in Supplement Materials (Supplement S1 Code and Dataset Repository\IPython Notebooks\cosine_similarity_for_the_compounds_and_drugs). The complete researched details of all of the researched ligands their in depth calculated descriptor values and structures are found in Supplementary Material (Table S2).

Comments 4: The reasoning and supporting data are in line with the conclusions.

Response 4: Thank you, the study tries as much as possible to not deviate with the main objectives.

Comments 5: The citations are suitable.

Response 5: Thank you, there are new appropriate citations added in the following study.

Comments 6: The provided link does not allow you to obtain the supplemental materials.

Response 6: The newly uploaded Supplementary Materials are now added and organized properly in order for the readers to fully understand the research done.

Response to Reviewer 2 Comments

1. Summary

Thank you very much for taking the time and effort to review the manuscript. The authors of the following study are pleased to find all of the comments and suggestion in the review process as it helped the study to be improved significantly. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the newly re-submitted revised manuscript and files. All of the revisions done on the manuscript are highlighted in red. Thank you again for the detailed review.

2. Questions for General Evaluation

Reviewer’s Evaluation

Response and Revisions

Does the introduction provide sufficient background and include all relevant references?

Can be improved

Thank you very much reviewer 2 for reviewing the following paper. Point by point responses in each of the comments are detailed below.

Is the research design appropriate?

Yes

Are the methods adequately described?

Yes

Are the results clearly presented?

Can be improved

Are the conclusions supported by the results?

Can be improved

3. Point-by-point response to Comments and Suggestions for Authors

Comments 1: This article combines QSAR-ML (Quantitative Structure-Activity Relationship Machine Learning) techniques, ADME (Absorption, Distribution, Metabolism, and Excretion) prediction, and molecular docking to present a novel approach for screening ROCK1 inhibitors for HCV (Hepatitis C Virus) therapy. By combining machine learning with molecular docking, the authors demonstrated how to effectively identify potential drug candidates during drug screening, but the following shortcomings remain:

Response 1: Thank you very much the following research study was greatly motivated for the rising resistance of HCV to direct acting antivirals. Also, as the rising innovations in machine learning and AI technology are increasing in our age using applying their methods in drug discovery can be beneficial and cost-effecting for screening out potential compounds we would want to design. ROCK1 inhibition in the context of HCV is under-looked and the study tries to probe compounds in vast amounts and effectively screen only specific structures we would like to know.

Comments 2: The sources and characteristics of the datasets used need to be described in more detail, especially how they were selected and processed.

Although the article uses a variety of machine learning models and selects NuSVR as the best model, it does not detail the basis for model selection and its hyperparameter tuning process.

Response 2: Thank you again, the selection process for all of the datasets used are explained in the newly submitted manuscript at page 28 paragraph 1 lines 521-541.

Parameters used in building the NuSVR model are detailed at 2.6 QSAR-ML page 7-8 lines 312-320.

Comments 3: Some of the figure notes are not detailed enough; it is recommended to add explanations of the elements in the figures for better understanding by the readers.

Response 3: Thank you for pointing this out, the newly submitted manuscript expanded each of the description of figures for better understanding of the readers highlighted in red.

Comments 4: If possible, it is recommended to make the code and dataset public so that other researchers can reproduce and extend the findings.

Response 4: Thank you very much, Supplementary Material (Supplement S1: Code and Dataset Repository) in the newly uploaded zip file documents all of the scripts and datasets used in the following study.

Comments 5: Add background on the specific mechanism of action of ROCK1 in HCV infection so that readers can better understand the basis of the study. In the conclusion section, it is recommended to not only summarize the findings, but also suggest directions for future research and possible improvements.

Response 5: The background for the mechanism of action of ROCK1 in HCV is detailed at page 2 paragraph 1 lines 70-91. Possible improvements in future studies for other researchers are provided at page 31 paragraph 2 lines 674-689.   

Comments 6: The background of the molecular structure diagram needs to be changed to white.

Response 6: Thank you, the updated manuscript changed the background of the molecular structure diagrams to white, it improved it visually for the readers to see.

Response to Reviewer 3 Comments

1. Summary

Thank you very much for taking the time and effort to review the manuscript. The authors of the following study are pleased to find all of the comments and suggestion in the review process as it helped the study to be improved significantly. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the newly re-submitted revised manuscript and files. All of the revisions done on the manuscript are highlighted in red. Thank you again for the detailed review.

2. Questions for General Evaluation

Reviewer’s Evaluation

Response and Revisions

Does the introduction provide sufficient background and include all relevant references?

Can be improved

Thank you very much reviewer 3 for reviewing the following paper. Point by point responses in each of the comments are detailed below.

Is the research design appropriate?

Can be improved

Are the methods adequately described?

Yes

Are the results clearly presented?

Can be improved

Are the conclusions supported by the results?

Can be improved

3. Point-by-point response to Comments and Suggestions for Authors

Comments 1: I found this paper particularly interesting, especially the design of the study. However, I have the following major comments, as designing specific inhibitors for kinases is a challenging task:

Response 1: Thank you very much the following research study was greatly motivated for the rising resistance of HCV to direct acting antivirals. Also, as the rising innovations in machine learning and AI technology are increasing in our age using applying their methods in drug discovery can be beneficial and cost-effecting for screening out potential compounds we would want to design. ROCK1 inhibition in the context of HCV is under-looked and the study tries to probe compounds in vast amounts and effectively screen only specific structures we would like to know.

Comments 2: The authors should perform MD simulations of the protein-ligand complex for at least 100 nanoseconds to demonstrate the stability of the binding.

Response 2: Thank you for the insight on the addition of MD simulation for further improving the stability of the complexes. The following study faces challenging constraints in terms of computational resources and time in order to do MD simulations. The following recommendation is not added onto the study as the scope of the study is limited on only doing QSAR-ML analysis and Docking as referred on its main objectives highlighted on page 3 paragraph 4 lines 134-147.

Comments 3: The binding energy should be calculated using the QM/MM method to provide a more accurate assessment.

Response 3: Again thank you for the insight as there is great value in the suggestion, the study still acknowledged the insight that this added validation method will provide that is why the input parameters for QM/MM simulations are still provided for the system which are in Supplementary Material (Supplement S3), generated using the qm/mm interfacer from CHARMM-GUI.

Comments 4: The molecular interaction of the protein-ligand complex should be presented, highlighting the interacting residues and hydrogen bond interactions. You may use LigPlot+ for this purpose.

Response 4: Thank you very much, the newly updated revised manuscript described all of the contact residues in the following study on page 25-26 and page 30 paragraph 1 line 617-632, the program used is PlayMolecule’s Plexview.