Qiang Huang

ParSeek is a deep learning model for cryo-EM particle picking trained on synthetic data, achieving high accuracy and outperforming existing methods on multiple datasets.

This paper develops a computational method using graph neural networks and Metropolis Monte Carlo sampling to design high-affinity nanobodies, achieving a Pearson correlation of 0.62 on benchmark.

This paper introduces AiDTA, a deep reinforcement learning method for de novo design of protein-binding aptamers, achieving high-affinity DNA aptamers for disease-related proteins.

This study designs nanobodies against SARS-CoV-2 variants using a deep generative model, identifying two neutralizing candidates.

This study identifies how directed-evolution mutations in ABE8e enhance DNA-binding affinity and protein stability, improving its base editing activity.

Monte Carlo protein-protein docking estimates that the 2019-nCoV spike receptor-binding domain binds human ACE2 with about 73% of the affinity of SARS-CoV.

This study uses glycan arrays and single-virus force spectroscopy to show that influenza A virus receptor binding preference does not always match cell specificity.