Frédéric Cazals

The paper extends Ramachandran diagrams by modeling amino acid and pair torsion-angle distributions conditioned on secondary structure elements, enabling more accurate conformational predictions.

The paper compares binding modes and thermodynamic properties of single domain and double domain antibodies using an integrated interface model.

The paper evaluates AlphaFold predictions using a new scoring function and identifies three structural populations in proteomes.

The paper introduces new quantitative measures to assess AlphaFold predictions by analyzing pLDDT values along protein sequences.

The paper introduces SPECTRALDOM, an improved method for protein domain identification that outperforms existing approaches without using evolutionary information.

The paper analyzes AlphaFold predictions across whole genomes, examining packing properties, pLDDT scores, and disordered regions to improve confidence assessment.

This paper analyzes geometric constraints in tripeptide loops and shows how relaxing constraints increases the configuration space volume for protein loop sampling.

This paper introduces a new method for sampling protein loop conformations using a global parameterization of backbone geometry, achieving better diversity and lower energy compared to existing approaches.

This paper introduces Genetrank, a method to prioritize genes regulating drug pharmacodynamics using random walks on protein interaction networks, applied to single-cell TRAIL sensitivity data.

The paper studies tripeptide loop closure by analyzing solutions from Ramachandran distributions and shows their geometric diversity and potential energy accuracy.

The paper introduces MISA, a tool for visualizing protein interface statistics, and applies it to analyze coronavirus spike proteins.

The paper introduces combined RMSD as a method to better characterize molecular flexibility by combining local RMSD measures, demonstrating its utility in analyzing conformational changes, structural phylogenies, and quaternary structure assignment.

The paper introduces a hybrid sequence-structure HMM method to identify homologous proteins with low sequence identity and structural conservation, demonstrating its effectiveness in detecting class II fusion proteins.

This paper introduces a framework for multiscale structural analysis of proteins undergoing conformational changes, improving structural comparison and homology detection.