Publications
Generalizable and scalable protein stability prediction with rewired protein generative models
Nature Communications 2026
We developed SPURS (stability prediction using a rewired strategy), the SOTA protein stability predictor, to benefit protein function understanding, such as functional site discovery, fitness prediction, and pathogenicity.
Deconvolving mutation effects on protein stability and function with disentangled protein language models
RECOMB 2026
DETANGO disentangles functional and stability constraints in protein function modeling by integrating evolutionary and thermodynamic predictors, and evaluates this relationship following the SPURS functional site identification task.
Rewiring protein sequence and structure generative models to enhance protein stability prediction
RECOMB 2025
RECOMB version of the SPURS.
Double-ended synthesis planning with goal-constrained bidirectional search
NeurIPS 2024 Spotlight
Toward universal cell embeddings: integrating single-cell RNA-seq datasets across species with SATURN
Nature Methods 2024
Rethinking the Setting of Semi-supervised Learning on Graphs
IJCAI 2022 Oral
We highlight the issue of over-tuning in GNNs on the validation set.
Modeling Protein Using Large-scale Pretrain Language Model
Pretrain@KDD 2021
The early research on Protein Language Models featured the largest Protein Language Model at the time.