Exhaustive Variant Interaction Analysis Using Multifactor Dimensionality Reduction

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a clearly written paper that makes a software contribution in the sense of offering a parallel platform for the analysis and a scientific contribution by identifying promising pairs of gene variations that could indicate diabetes. Similar methods could apply to other datasets of diseases or other phenotypes.

My main problem with the work is that the authors are fixed on pairs of genes, but there are many phenotypes that involve scores or even hundreds of genes. For that reason I would recommend a comparison of the results with random forests, boosted random forests, and BART (Bayesian Additive Regression Trees). The authors can continue to use the parallel platform for the construction of random forests and for cross-validation and for the final processing of the test data.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Innovative part should be better described.

What are clinical benefits of proposed method?

How about comparison of the MDR results with other techniques for same problem.

HPC is using for reduce the speed of calculation, but not clear can this method apply on single computer and how much time it will be need to solve same problem.

Limitations of the study should be more explored.

Comments on the Quality of English Language

Some typos should be fixed.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

1.     Can the interactions between genomic variants be generalized across different populations or are they population-specific?

2.     How do environmental factors interact with genomic variants to modulate the risk of developing T2D?

3.     How does the identified subset of variant pairs associated with T2D compared to previously reported associations, and what insights does it provide into the genetic architecture of T2D?

4.        Can the findings from genomic variant interaction studies be translated into personalized medicine approaches for T2D prevention or treatment?

5.        What computational methods or algorithms beyond Multifactor Dimensionality Reduction (MDR) can enhance the detection of variant interactions in large genomic datasets? Does MDR have any advantages over them？

6.        How do epigenetic modifications influence the interactions between genomic variants and their impact on T2D susceptibility?

7.        What additional computational and experimental validation steps are needed to confirm the functional relevance of the identified variant pairs and their interactions in T2D pathophysiology?

8.        Can the methodology developed in this study be adapted to investigate variant interactions associated with other complex diseases beyond Type 2 Diabetes?

Comments on the Quality of English Language

 English language fine, Just be aware of a few minor mistakes

•  

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

All my comments have been addressed.

Reviewer 2 Report

Comments and Suggestions for Authors

Authors responded to my comments.