Barthélémy Caron

AtlasOT improves multimodal integration of disease atlases by addressing unpaired data constraints through a novel optimal transport framework.

PAGAN predicts digenic interactions by learning network representations of gene pairs from single-gene data in biological networks.

NCBoost v2 improves pathogenicity scoring for non-coding variants in Mendelian diseases using updated conservation features and splice-altering scores, outperforming existing methods.

The paper identifies novel genetic and non-genetic factors associated with plasma protein levels in healthy donors, highlighting their impact on immune response variability.

The paper develops a supervised learning method to score pathogenic non-coding variants in Mendelian diseases using purifying selection signals, achieving improved predictive performance over existing methods.