Junzhou Huang

The paper reports an in vitro system of lipid kinases and phosphatases that self-organize into dynamic spatiotemporal patterns on model membranes.

This study compares two transfer learning approaches for predicting small molecule bioactivity, finding similar performance but different strengths in handling smaller assays.

Cmai predicts antigen-antibody interactions from high-throughput BCR sequencing data, revealing immune responses to tumor antigens and ICI treatment outcomes.

scBERT is a pre-trained deep learning model for cell type annotation in single-cell RNA-seq data, showing improved performance and robustness compared to existing methods.

This study tracks prenatal development of orofacial movements in marmoset monkeys, showing that vocalization-related patterns emerge before birth.

This paper introduces a Graph Enhanced Transformer model for retrosynthesis prediction, achieving better performance than a vanilla Transformer by incorporating molecular graph information.

This paper introduces SE(3)-Fold, an end-to-end protein folding method using SE(3)-equivariant energy-based models, achieving comparable accuracy to state-of-the-art methods with significant speed improvements.

This study uses NMR to analyze metabolic changes in blood and cerebrospinal fluid of Alzheimer's model mice, finding that metabolite correlations are influenced by amyloid plaque formation.

The paper introduces MLA-GNN, a graph neural network that discovers clinically interpretable pathway-level biomarkers by integrating multi-level graph features and a saliency mechanism, achieving state-of-the-art results on transcriptomic and proteomic data.

The paper classifies binding and non-binding intra-protein peptide fragments with high accuracy using a feed-forward neural network.