Advancing regulatory genomics with machine learning

TL;DR

This paper reviews machine learning methods in regulatory genomics, highlighting techniques from linear models to deep learning, and discusses how to extract and validate biological hypotheses from these models.

Contribution

It provides a comprehensive overview of machine learning approaches in regulatory genomics and discusses methods for assessing confidence in biological insights.

Findings

Deep learning models enable complex gene regulation predictions

Different models offer varying confidence measures for hypotheses

Review of strategies for hypothesis extraction and validation

Abstract

In recent years, several machine learning approaches have been proposed to predict gene expression and epigenetic signals from the DNA sequence alone. These models are often used to deduce, and, to some extent, assess putative new biological insights about gene regulation, and they have led to very interesting advances in regulatory genomics. This article reviews a selection of these methods, ranging from linear models to random forests, kernel methods, and more advanced deep learning models. Specifically, we detail the different techniques and strategies that can be used to extract new gene-regulation hypotheses from these models. Furthermore, because these putative insights need to be validated with wet-lab experiments, we emphasize that it is important to have a measure of confidence associated with the extracted hypotheses. We review the procedures that have been proposed to measure this confidence for the different types of machine learning models, and we discuss the fact that they do not provide the same kind of information.