Review Reports

Round 1

Reviewer 1 Report

Please explain and define " Large scale"  for this study ..line 141 and also other terms used

Elaborate more and quantify on the evaluation section., What's the evaluation benchmark used for this study ?

Elaborate on the AI enabler and details on this study

Need strong justification on the medical laboratory selected for evaluation 

Author Response

Dear Reviewer,

thank you for taking the time to review and comment on our paper. Your suggestions have helped us to improve our paper.

We have addressed your comments as following:

1. Explaining terms like "large scale" and others: We have read through the paper again and have identified several terms that readers might not be familiar with or that lacked a clear definition. We have addressed these shortcomings by adding additional sentences to explain the terms or in some cases by removing the unclear term if it didn't provide additional insights.
2. Evaluation Section: We have added more content to better explain the evaluation approach as well as possible shortcomings.
3. AI enabler: We have added a short explanation why we think that GenDAI could help with AI utilization while also making clear that ultimately more work is needed on this topic in the technical architecture and implementation.
4. Laboratory: We have added an explanation on why the laboratory in question was chosen.

In addition we have done several other improvements like expanding on some topics in the introduction section as well as improving english language and style.

Thank you again for your time!

Reviewer 2 Report

The article provides a conceptual architecture called "GenDAI" which can be used to address the common challenges facing the development of new biomarkers in the context of genomic data. The article is well-written and organized. There are some parts in the background that the author needs to clarify.

1. the omics data not only includes the transcriptomic, and genomics but also includes other types of data modalities like epigenomics, proteomics, etc.

2. The application of Artificial intelligence(AI) depends a lot on the sample size which is usually limited in the clinical field. The author doesn't introduce the challenges from this perspective.

3. Even though deep learning can achieve a high accuracy rate, it works as a black box that can't be approved by regulation. The author doesn't introduce much about this.

Author Response

Dear Reviewer,

thank you for taking the time to review and comment on our paper. Your suggestions have helped us to improve our paper.

We have addressed your comments as following:

1. We have changed the introduction to make it more clear that metagenomics and gene expression analysis are only two examples within the field of omics.
2. The problem of small sample size was only addresses in a half-sentence before. We agree that it is an important issue, so we have expanded this in the introduction.
3. Deep learning is indeed often a black box and we did not really introduce this challenge. We have therefore added a paragraph explaining the problem, upcoming exemplary regulation in this area, and the criterias that need to be met for AI to comply with future regulation.

You have also suggested english language and style improvements, which is why we did a proof-reading with a native speaker to fix several mistakes.

In addition we made further improvements in the introduction section to explain some of the terms used (e.g. PMS/PMPF) and elaborated a bit more in the evaluation section to address feedback of other reviewers.

Thank you again for your time!

Round 2

Reviewer 1 Report

Authors have amend some of the feedbacks.