Kyoto University

Peptide language model trained on HELM notation, a DeBERTa encoder for property prediction on macrocyclic and non-canonical medium-sized peptides.

Peptide collision cross section predictor for ion-mobility proteomics, built on frozen protein language model features plus ion charge and mass.

Protein half-life prediction from sequence and AlphaFold-derived structure tokens, separating short- from long-lived proteins and nominating degrons.

Compound-protein interaction prediction coupling a chemical language model to a protein language model via a cross-attention block.

Generative RNA design model that samples family sequences from a VAE latent space constrained by Rfam covariance models and consensus structure.