Computational Biology of Noncoding Genomes

"Introductory sentence"

Context

Although only a small fraction of eukaryotic genomes encodes canonical proteins, recent advances in transcriptomics, translatomics and proteomics have revealed that noncoding regions are far more biologically active than previously anticipated. Thousands of RNAs originating from presumed noncoding regions are expressed and translated across species, giving rise to a large diversity of noncanonical products, including microproteins and alternative peptides.

Long considered as transcriptional or translational noise, these noncanonical products are now increasingly recognized as an important source of molecular innovation. Some have been associated with regulatory functions, stress responses, pathogenic processes, or immune recognition, while others may represent transient evolutionary intermediates participating in the emergence of novel genes and proteins.

The CBNG team develops integrative computational approaches to investigate how noncoding regions contribute to proteome plasticity, genome evolution in eukaryotic genomes, and the emergence of noncanonical molecular signatures associated with human diseases, particularly cancer. Our research combines comparative genomics, k-mer-based and alignment-free approaches, transcriptomics, ribosome profiling, proteomics, structural bioinformatics and machine learning to characterize the diversity, properties and evolutionary trajectories of noncanonical RNAs and protein products.

Research Axes

1/ Noncoding regions and their derived products

We investigate the diversity of products generated from noncoding regions, including noncanonical transcripts, translated small ORFs and microproteins. A major objective is to characterize their sequence, structural and evolutionary properties, as well as their cellular localization and potential functional impact.

Our work notably explores:

the diversity of noncanonical translation products
the structural potential hidden in noncoding regions
the folding and dynamic properties of microproteins
the potential of noncanonical translation products as sources of neoantigens
and the contribution of pervasive expression to proteome diversity

2/ Biological processes underlying pervasive expression

The team develops computational and statistical methods dedicated to the analysis of pervasive transcription and noncanonical translation.

Using transcriptomics, Ribosome Profiling and proteomics, we aim to characterize:

how noncanonical RNAs and translated products are generated
how their expression varies across conditions, species or pathologies
and which genomic or molecular determinants promote their production and persistence

Particular attention is given to the development of reference-free and comparative approaches enabling the detection of transcripts and translation products absent from reference genomes.

3/ From noncoding to coding: emergence, evolution and applications

A central objective of the team is to understand how novel genes and microproteins emerge from ancestrally noncoding regions.

*** We study the transition from transient noncanonical products to evolutionarily established genes by integrating comparative genomics, structural bioinformatics and evolutionary analyses. This includes the identification of sequence and structural features associated with the retention, stabilization or functionalization of emerging products.

*** We also investigate how noncanonical translation contributes to the molecular landscape of pathological contexts such as cancer. In this context, we develop approaches to identify tumor-specific translated products and noncanonical neoantigens that may represent promising targets for immunotherapy-based strategies.