A dynamical system model for predicting gene expression from the epigenome

TL;DR

This paper introduces a novel stochastic dynamical systems model that predicts gene expression from epigenetic methylation data within a gene regulatory network, offering a mechanistic approach beyond traditional statistical models.

Contribution

The paper presents a new dynamical systems model for gene expression prediction from epigenetic data, integrating gene regulatory networks for mechanistic insights.

Findings

Model evaluated on real patient data.

Software tools provided for data processing and prediction.

Demonstrates potential for mechanistic understanding of gene regulation.

Abstract

Gene regulation is an important fundamental biological process. The regulation of gene expression is managed through a variety of methods including epigenetic processes (e.g., DNA methylation). Understanding the role of epigenetic changes in gene expression is a fundamental question of molecular biology. Predictions of gene expression values from epigenetic data have tremendous research and clinical potential. Despite active research, studies to date have focused on using statistical models to predict gene expression from methylation data. In contrast, dynamical systems can be used to generate a model to predict gene expression using epigenetic data and a gene regulatory network (GRN) which can also serve as a mechanistic hypothesis. Here we present a novel stochastic dynamical systems model that predicts gene expression levels from methylation data of genes in a given GRN. We provide an evaluation of the model using real patient data and a GRN created from robust reference sources. Software for dataset preparation, model parameter fitting and prediction generation, and reporting are available at \verb|https://github.com/kordk/stoch_epi_lib|.