Qiong Luo

ProtBLIP2-SST generates open-ended protein function captions by integrating sequence, structure, and text using a two-stage framework.

This paper evaluates pLM-guided data augmentation for protein sequence prediction, showing that it improves performance across tasks with variation levels between 2-30%.

This study benchmarks nine protein structure alignment tools across homology detection, phylogeny reconstruction, and function inference, showing structure-based methods often outperform sequence-based ones.

DualNetGO selects features from PPI networks and protein attributes to predict protein functions, improving Fmax scores across Gene Ontology categories in human and mouse datasets.