Guijun Zhang

ForceFlowAb improves antibody CDR design by combining adaptive routing and physics-based guidance, achieving better interaction energies than existing methods.

This paper introduces DiffEnsemble, a diffusion-based framework for modeling protein conformational ensembles, showing improved performance over existing methods.

ProSiteHunter improves protein binding site prediction by integrating sequence-based features and a fine-tuned language model, achieving better performance than existing methods.

This paper introduces DeepUMQA-Global, a deep learning framework for estimating protein structure model accuracy, which outperforms existing methods in correlation metrics and consensus rankings.

MViewEMA improves protein complex model accuracy estimation with multi-view representation learning, achieving higher efficiency and performance than previous methods.

This paper introduces DEMO-EMReF, a Transformer-based framework for refining cryo-EM density maps, improving resolution and atomic model accuracy.

PathDiffusion models protein folding pathways using evolution-guided diffusion, accurately reconstructing folding events and demonstrating versatility across multiple applications.

This paper introduces DeepSCFold, a method for improving protein complex structure prediction by using sequence-based deep learning to model inter-chain interactions, achieving higher accuracy than existing methods.

SAMAffinity predicts mutation-induced changes in protein-protein binding affinity using a sequence-aware deep learning model, outperforming existing methods on multiple datasets.

DeepAAAssembly improves antibody-antigen complex modeling by integrating dynamic interactions, outperforming AlphaFold3 on a benchmark set.

M-DeepAssembly2 predicts multiple conformations of multi-domain proteins using deep learning, improving accuracy over previous methods.

MultiSAAI predicts antibody-antigen interactions using a multi-scale deep learning framework, achieving high AUROC scores on benchmark datasets.

DeepAssembly2 improves protein complex structure prediction using domain-domain interactions and integrates new features and models to enhance 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.

SPDesign designs protein sequences using structural sequence profiles and ultrafast shape recognition, outperforming existing methods in sequence recovery and structural accuracy.

This paper introduces MultiSFold, a method for predicting multiple protein conformations using a multi-objective evolutionary algorithm, achieving higher accuracy than AlphaFold2 and RoseTTAFold on benchmark tests.

This paper introduces FoldPAthreader, a method predicting protein folding pathways using a novel force field derived from evolutionary data, achieving 70% accuracy on 30 targets.

GraphCPLMQA evaluates protein model quality using deep graph coupled networks with protein language model embeddings, achieving top performance on benchmark datasets.

E2EDA improves multi-domain protein structure prediction by accurately assembling domains using deep learning, achieving higher accuracy and faster performance than existing methods.

Pathfinder predicts protein folding pathways by analyzing conformational sampling data, showing accurate predictions for some proteins and potential pathways for others.

This paper introduces GraphGPSM, a graph neural network-based scoring model for protein structure prediction, which improves accuracy over existing methods on benchmark datasets.

This paper introduces DeepAssembly, a method for predicting multi-domain protein and complex structures using inter-domain interactions, achieving higher accuracy than AlphaFold2 on multi-domain targets.

DeepUMQA3 evaluates protein complex structures using a deep neural network, achieving high accuracy in interface residue assessment and ranking first in CASP15 blind tests.

This paper introduces DeepIDDP, a deep learning method for inter-domain distance prediction, which improves domain assembly accuracy when integrated with SADA.

DeepUMQA2 improves protein model quality assessment using enhanced deep neural networks, achieving better performance on CASP13, CASP14, and CAMEO 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.

This paper introduces SADA, a deep learning-assisted method for assembling multi-domain protein structures, achieving higher accuracy than existing methods.

This paper introduces DEMO-EM, a method for assembling multi-domain protein structures from cryo-EM maps, achieving high accuracy and outperforming existing methods.

This paper introduces VFlib, a variable-length fragment library for de novo protein structure prediction, which improves global precision over existing methods.

DeepUMQA improves protein model quality assessment by integrating Ultrafast Shape Recognition features with deep learning, outperforming existing methods on benchmark datasets.