You are here

    • You are here:
    • Home > Events > PRBB Computational Genomics Seminars Mathys Grapotte

PRBB Computational Genomics Seminars Mathys Grapotte

PRBB Computational Genomics Seminars Mathys GrapottePRBB Computational Genomics Seminars Mathys Grapotte

Add to Calendar

PRBB Computational Genomics Seminars Mathys Grapotte


30/06/202212:00R_473.10_AULAPRBB Computational Genomics SeminarsMathys GrapotteSanofi Chilly-Mazarin / Institut of Molecular Genetics (CNRS) Montpellier, FranceDeep Neural Network to assess the functional consequences of microsatellite-associated transcription on gene expressionHost: Garrido Enamorado, RominaAbstract:link:

The human genome is scattered with repetitive sequences, in particular microsatellites, also called short tandem repeats (STRs), which correspond to repeated DNA motifs of 2–6¿bp and constitute one of the most polymorphic and abundant repetitive elements1. STRs have been shown to widely impact gene expression through various molecular mechanisms2 ,and some constitute genuine expression Quantitative Trait Loci (eQTLs), called eSTRs3 .The eSTRs traditionally regress gene expression with STR length and do not consider
Single Nucleotide Polymorphisms (SNPs) located at the vicinity of STRs. Within the framework of the international FANTOM consortium, we discovered widespread transcription initiation at STRs4. We found that
genetic variants linked to human diseases are preferentially detected at STRs with high transcription initiation level, supporting the idea that transcription plays key functions in STR-mediated regulations. We developed fully interpretable Deep Neural Network (DNN) able to accurately predict transcription initiation at STRs and to specifically assess the impact of SNPs located around STRs. We are now combining GTEx data and the
output of our models to revisit eSTR computation, and to evaluate the functional consequences of STR-initiating transcription on gene expression. We anticipate that these analyses will help us interpret thousands of genetic variations that have been linked to human diseases by statistical analyses (GWAS or eQTLs) without any molecular mechanism to support these regulations

1. Willems, T. et al. The landscape of human STR variation. Genome Res. 24, 1894–1904 (2014).

2. Bagshaw, A. T. M. Functional Mechanisms of Microsatellite DNA in Eukaryotic Genomes. Genome
Biol. Evol. 9, 2428–2443 (2017).

3. Gymrek, M. et al. Abundant contribution of short tandem repeats to gene expression variation in humans. Nat. Genet. 48, 22–29 (2016).

4. Grapotte, M. et al. Discovery of widespread transcription initiation at microsatellites predictable by sequence-based deep neural network. Nat. Commun. 12, 3297 (2021).