Stan Ziqing Li

The paper introduces VitaGRN, a multimodal biophysical framework that predicts perturbation responses and uncovers non-canonical regulatory mechanisms, demonstrating robust generalization across cell lines and factors.

Rider improves RNA virus discovery from metatranscriptomes by combining structure-informed screening with structural validation, outperforming existing tools and recovering divergent sequences.

The paper introduces SaprotHub, a platform that simplifies training and sharing of protein language models, enabling non-experts to build and collaborate on customized models.

This paper introduces modality-chain reasoning for multimodal protein modeling and design, showing improved performance in structure prediction, inverse folding, and fitness prediction compared to existing models.

SearchNovo improves peptide sequencing by combining database search and de novo methods, outperforming existing models on multiple datasets.

This paper introduces FoldToken4, a model that generates consistent and hierarchical multiscale fold languages for protein structures, reducing redundant training across scales.

This paper introduces laruGL-PPI, a label-aware hierarchical subgraph learning method for protein-protein interaction prediction that improves performance and efficiency by addressing topological shortcuts and computational costs.

FoldToken3 compresses protein structures into 256 tokens or less while maintaining reconstruction quality comparable to FoldToken2.

FoldToken2 improves protein structure representation by creating a compact, invariant language that outperforms previous methods in reconstruction accuracy.

This paper introduces M-SADA, an evolutionary algorithm for modeling multiple conformational states of multidomain proteins, showing improved performance over AlphaFold2 on benchmark datasets.

GeoAB improves antibody design by generating realistic CDR structures and enabling accurate affinity maturation through mutation effect prediction.

PPFlow designs target-aware peptides using torsional flow matching, achieving state-of-the-art performance in peptide drug generation and optimization.

This paper introduces CoupleNet, a model that integrates protein sequence and structure to improve representation learning, achieving better performance in fold classification and function prediction.

This paper introduces a visualization method for scRNA-seq data that preserves data structure and corrects batch effects, showing its effectiveness across multiple datasets.

This paper introduces PAthreader, a method for identifying remote homologous structures using deep learning, which improves template accuracy and enhances protein folding pathway exploration.

The paper proposes two strategies to improve generalization in fine-tuning pre-trained molecular graph models by augmenting data and regularizing model complexity.

This paper develops a deep learning method using DIA tensor for phenotype prediction from mass spectrometry data, achieving high accuracy in cancer classification.