Bing He

This paper introduces USM-AA, an alignment-aware genomic foundation model that improves long-range DNA modeling by incorporating reference genome coordinates and reverse-complement masked learning.

The paper develops a large language model to predict SARS-CoV-2-specific TCRs and shows their correlation with antibody levels post-vaccination.

This paper introduces InterAb and InterAb-Opt, computational frameworks that predict and optimize broadly neutralizing antibodies using all-atom modeling and pre-trained language models, showing significant improvements in binding affinity and neutralization activity.

The paper constructs a single-cell GEX-TCR reference atlas and introduces TCR-DeepInsight, a deep-learning framework to characterize disease-associated TCRs.

The paper benchmarks TCR reconstruction methods from scRNA-seq data, finding TRUST4 performs best, while sequencing depth limits accuracy.

The paper uses single-cell imaging features to predict breast cancer patient survival and identifies seven survival subtypes with distinct cellular profiles.

This study investigates how Alzheimer's risk SNPs affect chromatin functions and how their genomic context influences these effects.

The paper introduces ERTransformer, a large language model that generates T-cell receptors with desired epitope-binding properties, showing high functional similarity to natural TCRs.

This paper introduces PALM, a pre-trained generative large language model for de novo antibody CDRH3 design, and A2binder for predicting antigen-antibody binding, showing effectiveness against SARS-CoV-2 variants.

The paper introduces MoFNet, a deep neural network that integrates multi-omic data to identify functional subnetworks associated with Alzheimer's Disease, showing replication in an independent cohort.

This study identifies maternal plasma lipid metabolites associated with preterm birth using metabolomic analysis and machine learning in the Puerto Rico PROTECT cohort.

This study benchmarks integration methods for spatial transcriptomics data, showing that combining highly variable and spatially variable genes improves clustering performance.

The paper develops a two-stage neural network model for liver cancer prognosis using histopathology and transcriptomic data, showing improved accuracy compared to existing methods.

This study identifies lipidomic changes in maternal blood linked to severe preeclampsia, showing altered lipid species and their potential as biomarkers.

The paper identifies a pre-hematopoietic (pre-HE) stage in mouse embryos and shows RUNX1 regulates the transition to hemogenic endothelium, leading to distinct HSPC populations in intra-arterial clusters.

The paper examines 3D genome organization, epigenome, and transcriptome changes in fetal and adult mouse HSCs, finding stage-specific chromatin interactions and transcription factor roles.

The paper defines the transcriptomes and epigenomes of key developmental stages of HSC ontogeny in mice and constructs transcriptional regulatory networks using computational methods.