Genome Wide Epistasis Study of On-Statin Cardiovascular Events with Iterative Feature Reduction and Selection

Round 1

Reviewer 1 Report

In this manuscript titled, " Genome Wide Epistasis Study of On-Statin Cardiovascular Events with Iterative Feature Reduction and Selection ", Solomon M. Adams et al., authors aimed to stratify the risk of on-statin MACE based on polygenic epistatic predictors. Overall, the manuscript is written clearly.  However, the manuscript appears preliminary.

1. In the abstract, line 6 “Controlled-access data for 5,980 subjects taking a statin collected from…..”, there have 5,980 subjects but in results, line 82 the numbers of subjects are 5,890. Which one is correct?
2. In figure 2, the authors need to note what does LD mean?
3. The authors showed six unique networks of genes and variants in this study. Do you confirm these target genes in vitro? If not, how can you apply them to risk estimation for on-statin MACE?

Author Response

Dear Reviewer, 

Thank you for your kind review of our paper. Please find our responses to your critiques below. 

In the abstract, line 6 “Controlled-access data for 5,980 subjects taking a statin collected from…..”, there have 5,980 subjects but in results, line 82 the numbers of subjects are 5,890. Which one is correct?

Please accept my apologies for this error. The correct number is 5,890. This has been corrected in the manuscript. 

In figure 2, the authors need to note what does LD mean?

Figure 2 has been altered to spell out Linkage Disequilibrium in the legend. 

The authors showed six unique networks of genes and variants in this study. Do you confirm these target genes in vitro? If not, how can you apply them to risk estimation for on-statin MACE?

Thank you for this comment - we agree that this needs to be better explained in the manuscript. These networks have not been confirmed in vitro. Text was added to the discussion, figure legend, and conclusion that clarifies that the ultimate goal of this research should be clinical translation of the findings. However, next steps will be 1) replication of findings in independent (and diverse) cohorts of patients and 2) mechanistic study of interactions in vitro. 

Reviewer 2 Report

Adams et al convincingly based their work on a genome-wide epistasis study in order to reveal epistatic risk factors associated with on-statin Major Adverse Cardiovascular Events (MACE). All the proposed methodology is in accordance with the standards and there is no issue with the interpretations of the results. The highlighted association show that

1- the usual low effect-sizes observed in genome wide studies can be overcome with this method

2- the vascular physiology probably have a distinguishable regulation pattern

3-the Random Forest Iterative Feature, Reduction and Selection method was succesful to validate some previous observations, and to bring new insights/targets.

As such, the manuscript is a very good piece of work combining the state-of-the-art biomarkers discovery method, and is applied for a relevant topic (on-statin MACE).

However, it would have been a plus if the authors could discuss more about the originality of hteir work, and bring to the reader attention the expected impact (and limitation) to generalize the application of the methodology in the general population (at a large scale).

Lane 83, the control and cases numbers do not correspond to the method ones.

Author Response

Dear Reviewer, 

Thank you for your kind review and suggestions. Please find our response to your critiques below. 

However, it would have been a plus if the authors could discuss more about the originality of hteir work, and bring to the reader attention the expected impact (and limitation) to generalize the application of the methodology in the general population (at a large scale).

Thank you for this comment - we agree that additional information was needed to clarify the originality of this work and the context of the field. A new section was added to the discussion that describes the significance of this to the field, and especially focuses on the need for downstream research into the specific interactions found to determine 1) their reproducibility in independent populations and 2) in vitro determination of mechanisms. 

Lane 83, the control and cases numbers do not correspond to the method ones.

We apologize as this was an oversight. This has been fixed to the correct numbers. 

Round 2

Reviewer 1 Report

The manuscript has been significantly
improved and enough to publish.