Predicting Gene Expression Between Species with Neural Networks

TL;DR

This paper presents a neural network model trained on paired human and rat gene expression data to accurately predict human gene expression levels from rat data, facilitating cross-species genomic analysis.

Contribution

It introduces a neural network approach trained on paired data to predict human gene expression from rat data, advancing cross-species gene expression prediction methods.

Findings

The neural network accurately predicts human gene expression levels.

Predicted differentially expressed genes closely match actual human data.

Model performs well on unseen compounds, demonstrating robustness.

Abstract

We train a neural network to predict human gene expression levels based on experimental data for rat cells. The network is trained with paired human/rat samples from the Open TG-GATES database, where paired samples were treated with the same compound at the same dose. When evaluated on a test set of held out compounds, the network successfully predicts human expression levels. On the majority of the test compounds, the list of differentially expressed genes determined from predicted expression levels agrees well with the list of differentially expressed genes determined from actual human experimental data.