Thomas Schiex

This paper introduces EffieDes, a neuro-symbolic AI framework for protein sequence design that combines deep learning with automated reasoning to generate sequences with complex functional properties.

This study identifies structural cross-compatibilities between BMC-H paralogs in Klebsiella pneumoniae, revealing hetero-hexamer formation and reinterpreting existing mass spectrometry data.

The paper improves hifiasm haplotype contigs for autopolyploid genomes by adding protein distribution constraints, enhancing balance and BUSCO scores.

This paper introduces a hybrid generative AI approach to design multi-component proteins by combining deep learning and automated reasoning, showing improved performance over existing methods in creating heteromeric assemblies.