Shengwen Yang

This paper introduces DTi-pKa, a graph neural network that predicts pKa values with thermodynamic consistency and achieves high accuracy on benchmark datasets.

This paper introduces Kairos-GFM, a model that generates transition states with high physical fidelity and low RMSD, outperforming diffusion-based methods but underperforming transport-based ones.

This paper introduces RF-PepTune, a method that improves cyclic peptide design by preserving binding geometry while optimizing multiple properties, achieving higher success rates in structural and functional criteria.