Antoine Taly

The paper tests if SMILES can be fragmented into ordered blocks for retrosynthetic analysis, finding that it improves physico-chemical property reproduction but degrades LLM performance.

This paper evaluates limitations of deep learning in biomolecular structure prediction, highlighting the benefits of explicit ion modeling for improving conformational accuracy.

The paper uses AlphaFold to generate full-length structural models of mitofusins, revealing a novel cross-type dimerization mode and a new hypothetical mechanism of outer mitochondrial membrane fusion.

This study evaluates SMILES and SELFIES as molecular representations for generative chemistry, finding that neither ensures both viability and fidelity, and introduces ClearSMILES to improve both.

This paper introduces Statistical Molecular Interaction Fields (SMIF) to characterize RNA and protein binding pockets, showing they are fast, informative, and compatible with pharmacophoric models.

The paper introduces Menger_Curvature, a MDAKit implementation for calculating Menger Curvatures and Local Flexibilities in polymers, showing it is faster than RMSF and effective for analyzing protein dynamics.

The paper shows the ER stabilizes plasmodesmata during incomplete cytokinesis in plants by preventing fenestrae fusion and forming communication bridges between daughter cells.

The paper models the structure of the active NOX2 complex using AlphaFold2 and experimental data, revealing the role of disordered regions in subunit assembly.

The paper identifies Lys716 in yeast mitofusin Fzo1 as a residue modulating dimer stability and fusion through molecular dynamics simulations and experimental validation.

This study uses integrative modeling to predict the structure of the PAR1-TRAP6 complex, revealing a novel T-shaped pi-stacking interaction.

The paper shows plasmodesmata function as ER-PM tubular membrane contact sites that regulate intercellular molecular exchange in plants through PI4P and MCTP proteins.

The paper predicts the structure of Arabidopsis MCTP4's transmembrane region using deep learning and explores its dynamics with molecular simulations, revealing conformational complexity not captured by deep learning alone.

This study demonstrates that the GluD2 receptor functions as an ion channel using genetically-engineered photopharmacology, showing light-controlled modulation of its activity.