Shaoliang Peng

This paper introduces HomoDSP, a method for DNA-binding specificity prediction using homologous protein-DNA structures, showing improved accuracy and generalization over existing methods.

CAPTAIN is a multimodal foundation model trained on single-cell RNA and protein data, improving multi-omics integration and uncovering protein-driven intercellular dynamics.

This paper introduces a symmetric self-play framework for protein inverse folding that improves sequence design self-consistency by decoupling structural objectives.

The paper introduces EiRA, a generative model for designing proteins that bind biomolecules, achieving state-of-the-art performance across multiple metrics and demonstrating DNA-conditioned binder design.

GeSite identifies nucleic acid-binding residues using a domain-adaptive protein language model and explainable geometric deep learning, showing superior performance on benchmark datasets.

SpaCCC uses a fine-tuned LLM and gene network to infer cell-cell communications from spatial transcriptomic data, outperforming existing methods and revealing hidden ligand-receptor pairs.

This paper introduces ESM-DBP, a domain-adapted protein language model for DNA-binding proteins, showing improved prediction performance and interpretability.

This paper introduces scDCA, a method to identify dominant cell communication assemblies from scRNA-seq data, validated in renal cell carcinoma to reveal key immune gene expression patterns and therapeutic responses.

This study expands the number of complete temperate phage genomes by over 100-fold using a novel computational method applied to bacterial NGS data.

HyperVR predicts virulence factors and antibiotic resistance genes in microbial data using gene sequences, achieving high precision and recall without strict thresholds.

selfRL improves link prediction in heterogeneous biomedical networks by using self-supervised learning, outperforming 25 methods and identifying drugs that inhibit cytokine storms in COVID-19.