Tingjun Hou

PartiDesign is a structure-sequence co-design model that generates molecular structures and sequences for proteins, nucleic acids, and small molecules, achieving high design throughput across benchmarks.

LucaCell is a sequence-centric foundation model that improves cross-species single-cell analysis by using mRNA sequence embeddings instead of fixed gene identifiers.

PRIS improves protein-RNA interaction prediction and RNA library screening by combining structure selection and binding preference inference.

IgGM2 is an all-atom foundation model for immune receptor design that jointly predicts CDR sequences and full-atom structures, outperforming existing methods in structure prediction and sequence design benchmarks.

BBBP-Atlas predicts blood-brain barrier permeability for small molecules and peptides using a cross-modal dataset and structure-aware model, achieving high accuracy and outperforming existing methods.

LucaVirus predicts viral evolution and function using a unified nucleotide-protein language model, achieving state-of-the-art results in key virology tasks.

BioClaw enables human-bot collaboration in group chats to execute biomedical analyses using containerized tools and workflows across multiple domains and data types.

ProtFlow generates high-quality protein sequences by learning the full semantic distribution of protein space, outperforming existing methods in generating antimicrobial peptides with broad activity.

Ophiuchus-Ab is a generative foundation model for antibody design that captures inter-chain dependencies and predicts antibody properties like developability and binding affinity.

ECloudGen improves molecular design by using electron clouds as latent variables, generating more potent binders and exploring broader chemical space than existing methods.

This study reveals that PLX5622 affects anesthetic emergence and nicotine-withdrawal anxiety through enhanced drug metabolism, not microglial depletion.

EpiMII improves MHC-II epitope and neoantigen design by integrating structural features with graph neural networks, achieving higher accuracy and in vivo efficacy in tumor models.

The paper introduces PepPCBench, a benchmark framework to evaluate AlphaFold3's performance in predicting protein-peptide complexes using a curated dataset.

The paper introduces MCGLPPI, a geometric representation learning framework that combines graph neural networks with a molecular coarse-grained model to predict PPI properties efficiently.

MultiTAP improves tumor antigen immunogenicity prediction by integrating sequence and structure features with an attention mechanism, outperforming existing models across diverse HLA phenotypes.

This paper introduces ProtET, a method for CLIP-informed protein editing using multi-modality learning, achieving significant improvements in protein stability and functionality.

The paper reports high-resolution structures of MC4R with agonists and reveals insights into ligand recognition and activation mechanisms.