The joint IGMM/LIRMM/IMAG “Computational Regulatory Genomics” develops machine learning and statistical methods to integrate and interpret diverse types of genomics data, delineate relevant genomic regions and identify novel regulatory elements, with the ultimate goal of uncovering genetic features relevant to medical genomics. As an example, I will present our recent work aimed at deciphering the contribution of low complexity regions, in particular microsatellites, in gene regulation. Microsatellites, also called Short Tandem Repeats (STRs), correspond to repeated DNA motifs of 2 to 6 bp and constitute one of the most polymorphic and abundant repetitive elements. STRs are known to widely impact gene expression and to contribute to expression variation. Working in the frame of the international FANTOM consortium, we discovered widespread transcription initiation at STRs. To probe this transcription, we developed fully interpretable modular neural networks (MNNs), which combines learning and interpretation in one single step preserving automatic feature extraction from DNA sequence. MNN is a fully explainable deep neural network that identifies motifs correlated with sequence-based expression. We further leveraged GTEx genetic and expression data to revisit expression(e)STR computations considering SNPs located around STRs and using the prediction of our models as regressors. This new eSTR catalog complements existing eQTLs and eSTRs based solely on STR length variation. Together, our findings contribute to the understanding of molecular regulatory mechanisms orchestrated by non-coding RNAs in human tissues and their impact on complex traits. Our work constitutes a useful resource for the interpretation of thousands of genetic variants located at the vicinity of microsatellites.

website: https://www.igmm.cnrs.fr/en/service/joint-igmm-lirmm-imag-computational-regulatory-genomics-team/