Yi Xiong

RNASTOP predicts and optimizes mRNA chemical stability using deep learning, improving accuracy by 13% and reducing free energy in a vaccine sequence.

RNARL is a reinforcement learning framework for RNA codon design that generates high-performance sequences across multiple species and RNA types.

Distance-AF improves AlphaFold2 predictions by incorporating user-specified distance constraints to refine protein structures and reduce RMSD.

ProStab predicts protein stability changes upon mutations using a deep learning framework combining sequence and structure information, showing strong performance and experimental validation.

GRATCR generates epitope-specific TCR sequences with fewer training data than state-of-the-art models, showing higher specificity and biological functionality.

This paper introduces Hot2Mol, a deep learning framework for designing drug-like PPI inhibitors by generating molecules that mimic hot-spot residues, showing improved performance over baselines.

PLMSearch uses a protein language model to perform fast and accurate homologous protein sequence search, improving sensitivity over traditional methods while matching structure search performance.

This paper introduces MultiPPIMI, a deep learning framework that predicts PPI-modulator interactions without requiring target structure or reference modulators, achieving high AUROC scores in benchmark tests.

NetGO 3.0 integrates a protein language model with NetGO 2.0 to improve functional annotations by leveraging unannotated proteins.

scCobra is a deep learning tool that integrates and harmonizes single-cell data by mitigating batch effects and enabling online label transfer.

This study introduces T4SE-XGB, an XGBoost-based model for predicting type IV secreted effectors from protein sequences, achieving high accuracy and interpretability.

This study introduces DTI-MLCD, a multi-label learning method with community detection for predicting drug-target interactions, showing improved performance over existing methods.

DTI-CDF improves drug-target interaction prediction by integrating hybrid features into a cascade deep forest model, achieving higher performance than existing methods.

NetGO improves large-scale protein function prediction by integrating sequence and network information, outperforming existing methods on CAFA benchmark data.

GOLabeler improves sequence-based protein function prediction by integrating multiple classifiers in a learning-to-rank framework, showing better performance than existing methods.