Data-driven methods to unravel hidden layers of genome regulation

Gene regulation involves both transcriptional and translational regulation.

At the translational level, codon usage has emerged as a ‘code inside the code’ that determines mRNA stability and influences protein folding. Synonymous mutations that were thought to be silent, instead have potential to directly regulate gene expression. Despite their importance, the mechanisms behind these processes remain obscure.

I propose to systematically study codon usage through data-driven methods, enabling a more complete understanding of gene regulation.

My proposed research will focus on development of open source tools and databases, in particular by systematically processing all published Ribo-seq data, that will subsequently be used to carry out large-scale analyses of translational dynamics (Aim 1).

Moreover, I will improve sequence- to-function prediction by integration of codon usage into the deep learning models and use modelling to test specific hypotheses (Aim 2).

Finally, I propose to establish an evolutionary Drosophila model to understand codon usage in the context of tRNA competition between mRNA and viruses or transposable elements (Aim 3).

Altogether, this work will lead to a significant shift in our understanding of codon usage and provide a computational toolbox widely applicable also beyond these immediate questions.