# Predictive Features of Gene Expression Variation Reveal a Mechanistic Link Between Expression Variation and Differential Expression
Clone this repo
<divalign="justify">
This repository contains code to reproduce the results of our paper [Predictive features of gene expression variation reveal a mechanistic link between expression variation and differential expression](https://www.biorxiv.org/content/10.1101/2020.02.10.942276v1) by Olga Sigalova, Amirreza Shaeiri, Mattia Forneris, [Eileen Furlong](http://furlonglab.embl.de/), and [Judith Zaugg](https://www.zaugg.embl.de/).
</div>
![synopsis](/images/synopsis_image_resized.jpg)
<divalign="justify">
We used a machine-learning approach to identify genomic features predictive of gene expression variation across individuals in Drosophila and human and applied this framework to link differential expression upon various perturbations to genomically encoded regulatory features.
* Expression variation across individuals is a robust gene-specific property
* Expression variation is informative of gene regulation
* Promoter architecture and overall gene regulatory complexity are predictive of gene expression variation
* Expression variation features could also predict differential expression upon various perturbations