Jiaqing Li

MolMAE is a multimodal masked autoencoder that learns molecular representations by integrating surface point clouds, 3D graphs, and SMILES-derived tokens, achieving strong performance on molecular property prediction tasks.

ChatDIA uses a zero-shot large language model to automate targeted DIA proteomics analysis with high accuracy and transparent rationales.

The paper validates helical symmetry parameters in EMDB by identifying and correcting inconsistencies between parameters and density maps.