Jintao Zhu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jintao Zhu has a total of 39 co-authors. Here are the top co-authors by number of shared publications:
- Jianfeng Pei
Center for Quantitative Biology, Academy for Advanced Interdisciplinary Studies, Peking University, Beijing 100871, China
7 shared publications - Luhua Lai
Center for Quantitative Biology, Academy for Advanced Interdisciplinary Studies, Peking University, Beijing 100871, China
6 shared publications - Juan Xie
BNLMS, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, China
3 shared publications - Yiyan Liao
School of Life Sciences, Peking University, Beijing 100871, China
2 shared publications - Chenmin Wu
State Key Laboratory of Cognitive Intelligence, University of Science and Technology of China, Hefei, China
1 shared publication - Chenxi Du
Tsinghua University, Beijing, China
1 shared publication
Latest Publications
- Open Access
DiffTernary: a Diffusion Model for Molecular Glue Ternary Complex Docking
Molecular glues (MGs) modulate protein function by inducing or stabilizing protein interactions. However, accurately modeling MG ternary complexes remains challenging. Here, we present DiffTernary, a diffusion-based docking method for the joint prediction of relative protein arrangement and MG binding pose. We introduce Ternary-PDB, a dataset of 76,702 experimentally determined ternary complexes curated from the Protein Data Bank using structural filters and a release cutoff aligned with AlphaFold 3 (AF3). When trained on this dataset, DiffTernary achieved acceptable protein–protein interaction (PPI) predictions (DockQ > 0.23) for 50.0% of MGBench complexes with a single prediction, compared with 51.1% for AF3. In comparisons with AF3, OpenFold3, ESMFold2, Protenix-v1 and Protenix-v2, DiffTernary ranked second with a best-of-25 PPI prediction success rate of 78.4%, behind ESMFold2 (90.9%) and ahead of AF3 (72.7%). A separate DiffTernary model trained on TernaryDB achieved a mean DockQ of 0.306 on the 94 complexes in the TernaryDB MGD test set with a single prediction, compared with 0.228 for DeepTernary. Confidence-based selection among 40 samples yielded acceptable PPI predictions for 41.5% of complexes, compared with 30.9% for DeepTernary. An E2Former variant increased training and sampling throughput 6.90-fold and 4.37-fold, respectively, under the evaluated configurations. Together, DiffTernary and Ternary-PDB provide an efficient docking approach for MG complex prediction and a valuable structural resource for MG discovery and design.
Open AccessDiffTernary: a Diffusion Model for Molecular Glue Ternary Complex Docking
Molecular glues (MGs) modulate protein function by inducing or stabilizing protein interactions. However, accurately modeling MG ternary complexes remains challenging. Here, we present DiffTernary, a diffusion-based docking method for the joint prediction of relative protein arrangement and MG binding pose. We introduce Ternary-PDB, a dataset of 76,702 experimentally determined ternary complexes curated from the Protein Data Bank using structural filters and a release cutoff aligned with AlphaFold 3 (AF3). When trained on this dataset, DiffTernary achieved acceptable protein–protein interaction (PPI) predictions (DockQ > 0.23) for 50.0% of MGBench complexes with a single prediction, compared with 51.1% for AF3. In comparisons with AF3, OpenFold3, ESMFold2, Protenix-v1 and Protenix-v2, DiffTernary ranked second with a best-of-25 PPI prediction success rate of 78.4%, behind ESMFold2 (90.9%) and ahead of AF3 (72.7%). A separate DiffTernary model trained on TernaryDB achieved a mean DockQ of 0.306 on the 94 complexes in the TernaryDB MGD test set with a single prediction, compared with 0.228 for DeepTernary. Confidence-based selection among 40 samples yielded acceptable PPI predictions for 41.5% of complexes, compared with 30.9% for DeepTernary. An E2Former variant increased training and sampling throughput 6.90-fold and 4.37-fold, respectively, under the evaluated configurations. Together, DiffTernary and Ternary-PDB provide an efficient docking approach for MG complex prediction and a valuable structural resource for MG discovery and design.
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 10138-10149
Benchmarking LLM Agents on Real-World Biological Database Curation for Data-Driven Scientific Discovery
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 10138-10149
Open AccessBenchmarking LLM Agents on Real-World Biological Database Curation for Data-Driven Scientific Discovery
- Open Access
- 1 January 2026
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
MetaPLI: retrieval-augmented in-context regression over protein–ligand structures for absolute binding affinity prediction
MetaPLI predicts binding affinity from complex structures and a few activity measurements. New assay results update the retrieved context, guiding subsequent prediction and compound selection in a lead optimization loop without retraining.
- 1 January 2026
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
Open AccessMetaPLI: retrieval-augmented in-context regression over protein–ligand structures for absolute binding affinity prediction
MetaPLI predicts binding affinity from complex structures and a few activity measurements. New assay results update the retrieved context, guiding subsequent prediction and compound selection in a lead optimization loop without retraining.
- Open Access
- 29 October 2025
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 54 (D1) , D1500-D1509
MGTbind: a comprehensive database of molecular glue ternary interactome
Molecular glues (MGs) are an emerging class of small molecules capable of inducing or enhancing protein–protein interactions, leading to reprograming of cellular events. In recent years, MG discovery has gained significant attention in drug discovery and synthetic biology studies. However, MG discovery remains exceptionally challenging as it continues to rely on serendipity and experimental screening efforts. In order to contribute data resources to facilitate the rational design of MGs, we developed the molecular glue and ternary binding (MGTbind) database, providing comprehensive resources about ternary structures and experimental data for the coverage of MG-engaged interactome. MGTbind database contains 3093 manually curated MGs with their chemical structures and physicochemical properties, along with 3924 ternary interactions with bioactivity measurements (e.g. degradation capacities for MG degraders, in vitro biochemical activities and cellular activities). Hierarchical binding affinity is also collected to enable exploring of cooperative binding mechanisms. Ternary complex structural data are systematically integrated, either from the Protein Data Bank or generated by AlphaFold 3. All of them can be intuitively analyzed and visualized along with various predicted metrics. A fully functional and user-friendly web interface that allows global search and smooth browsing is provided and is freely accessible at https://mgtbind.pkumdl.cn.
- 29 October 2025
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 54 (D1) , D1500-D1509
Open AccessMGTbind: a comprehensive database of molecular glue ternary interactome
Molecular glues (MGs) are an emerging class of small molecules capable of inducing or enhancing protein–protein interactions, leading to reprograming of cellular events. In recent years, MG discovery has gained significant attention in drug discovery and synthetic biology studies. However, MG discovery remains exceptionally challenging as it continues to rely on serendipity and experimental screening efforts. In order to contribute data resources to facilitate the rational design of MGs, we developed the molecular glue and ternary binding (MGTbind) database, providing comprehensive resources about ternary structures and experimental data for the coverage of MG-engaged interactome. MGTbind database contains 3093 manually curated MGs with their chemical structures and physicochemical properties, along with 3924 ternary interactions with bioactivity measurements (e.g. degradation capacities for MG degraders, in vitro biochemical activities and cellular activities). Hierarchical binding affinity is also collected to enable exploring of cooperative binding mechanisms. Ternary complex structural data are systematically integrated, either from the Protein Data Bank or generated by AlphaFold 3. All of them can be intuitively analyzed and visualized along with various predicted metrics. A fully functional and user-friendly web interface that allows global search and smooth browsing is provided and is freely accessible at https://mgtbind.pkumdl.cn.
- Open Access
- 21 June 2025
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 4 (8) , 2083-2103
DeepRLI: a multi-objective framework for universal protein–ligand interaction prediction
DeepRLI is a multi-objective deep learning framework for universal protein–ligand interaction prediction, applicable to binding affinity prediction, molecular docking, and virtual screening tasks, offering better guidance for drug discovery.
- 21 June 2025
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 4 (8) , 2083-2103
Open AccessDeepRLI: a multi-objective framework for universal protein–ligand interaction prediction
DeepRLI is a multi-objective deep learning framework for universal protein–ligand interaction prediction, applicable to binding affinity prediction, molecular docking, and virtual screening tasks, offering better guidance for drug discovery.
- Open Access
- 28 May 2025
- Published by Cold Spring Harbor Laboratory in bioRxiv
Comparative assessment of co-folding methods for molecular glue ternary structure prediction
Molecular glues (MGs) represent an emerging therapeutic paradigm capable of inducing or stabilizing protein-protein interactions, with broad applications in creating neomorphic interactomes and targeted protein degradation. However, current discovery efforts remain largely confined to experimental screening, whilein silicorational design of MGs persists as a formidable challenge. A critical step toward rational design lies in accurate ternary complex modeling. Given the scarcity of such complexes in the Protein Data Bank for training specialized predictive models, we tested the ability of recently developed co-folding models, including AlphaFold 3, Boltz-1, Chai-1, Protenix, and RoseTTAFold All-Atom in building the complex models. We systematically curated a dataset, named MG-PDB, with 221 non-covalent MG-engaged ternary complexes. MGBench were further introduced as a comprehensive benchmark set, which comprises 88 ternary structures excluded from co-folding models’ training data through rigorous time-based partitioning. Our benchmark results demonstrated that AlphaFold3 achieved the best overall performance among co-folding methods, in terms of both protein-protein interaction interface prediction (50.6% success rate) and MG-protein interaction recovery (32.9% success rate). However, our homology study showed that most of their successful predictions actually stemmed from memorization. Further analysis revealed three phenomena of current co-folding methods for MG ternary structure prediction. Firstly, these methods struggle to accurately model large interaction interfaces. Secondly, their predictive accuracy is notably reduced for domain-domain complexes compared to domain-motif interactions. Lastly, they face specific challenges in modeling MG degrader complexes with sufficient accuracy. We showcased they relied on the existing interaction patterns, and highlighted the need for further improved in novel E3 ligase systems. These findings reveal fundamental gaps in existing methods to learn atomic-level interaction rules for MG-engaged ternary complex modeling. MG-PDB and MGBench provide critical resources and mechanistic insights to advance computational MG discovery.Table of Contents Graphic:
- 28 May 2025
- Published by Cold Spring Harbor Laboratory in bioRxiv
Open AccessComparative assessment of co-folding methods for molecular glue ternary structure prediction
Molecular glues (MGs) represent an emerging therapeutic paradigm capable of inducing or stabilizing protein-protein interactions, with broad applications in creating neomorphic interactomes and targeted protein degradation. However, current discovery efforts remain largely confined to experimental screening, whilein silicorational design of MGs persists as a formidable challenge. A critical step toward rational design lies in accurate ternary complex modeling. Given the scarcity of such complexes in the Protein Data Bank for training specialized predictive models, we tested the ability of recently developed co-folding models, including AlphaFold 3, Boltz-1, Chai-1, Protenix, and RoseTTAFold All-Atom in building the complex models. We systematically curated a dataset, named MG-PDB, with 221 non-covalent MG-engaged ternary complexes. MGBench were further introduced as a comprehensive benchmark set, which comprises 88 ternary structures excluded from co-folding models’ training data through rigorous time-based partitioning. Our benchmark results demonstrated that AlphaFold3 achieved the best overall performance among co-folding methods, in terms of both protein-protein interaction interface prediction (50.6% success rate) and MG-protein interaction recovery (32.9% success rate). However, our homology study showed that most of their successful predictions actually stemmed from memorization. Further analysis revealed three phenomena of current co-folding methods for MG ternary structure prediction. Firstly, these methods struggle to accurately model large interaction interfaces. Secondly, their predictive accuracy is notably reduced for domain-domain complexes compared to domain-motif interactions. Lastly, they face specific challenges in modeling MG degrader complexes with sufficient accuracy. We showcased they relied on the existing interaction patterns, and highlighted the need for further improved in novel E3 ligase systems. These findings reveal fundamental gaps in existing methods to learn atomic-level interaction rules for MG-engaged ternary complex modeling. MG-PDB and MGBench provide critical resources and mechanistic insights to advance computational MG discovery.Table of Contents Graphic:
- Open Access
- 28 February 2025
- Published by American Chemical Society (ACS) in ChemRxiv
RefineScore: Improving Ligand Docking Accuracy and Interpretability by Predicting MDN Corrective Physical Interactions
Precisely evaluating the protein-ligand interaction is crucial in drug screening and optimization. There are significant advances in the application of machine learning approaches to developing protein-ligand scoring functions in recent years. However, traditional docking softwares and existing deep-learning methods remain unsolved limitations in terms of pose quality and binding affinity prediction accuracy. Furthermore, machine learning-based scoring functions are hard to generalize to unseen cases due to the scarcity of structure-affinity data, and mostly lack physical interpretability. In this study, we propose RefineScore, which learns the underlying physical laws by the neural network parameterized physical energy terms. In addition, mixed density network (MDN) is introduced to predict the distance likelihoods of van der Waals (vdW) and hydrogen bonding (Hbond) pairs and further correct the classical energy equations for the absolute binding free energy calculation. RefineScore provides a versatile atomic interaction evaluation framework within protein-ligand complexes and is highly interpretable for the key interaction analysis and energy contribution decomposition. By integrating these carefully tailored techniques, RefineScore maintains the advantages of machine learning models while providing physical interpretability and capturing complex interaction patterns. Our evaluation results show that RefineScore has superior scoring and ranking power.
- 28 February 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessRefineScore: Improving Ligand Docking Accuracy and Interpretability by Predicting MDN Corrective Physical Interactions
Precisely evaluating the protein-ligand interaction is crucial in drug screening and optimization. There are significant advances in the application of machine learning approaches to developing protein-ligand scoring functions in recent years. However, traditional docking softwares and existing deep-learning methods remain unsolved limitations in terms of pose quality and binding affinity prediction accuracy. Furthermore, machine learning-based scoring functions are hard to generalize to unseen cases due to the scarcity of structure-affinity data, and mostly lack physical interpretability. In this study, we propose RefineScore, which learns the underlying physical laws by the neural network parameterized physical energy terms. In addition, mixed density network (MDN) is introduced to predict the distance likelihoods of van der Waals (vdW) and hydrogen bonding (Hbond) pairs and further correct the classical energy equations for the absolute binding free energy calculation. RefineScore provides a versatile atomic interaction evaluation framework within protein-ligand complexes and is highly interpretable for the key interaction analysis and energy contribution decomposition. By integrating these carefully tailored techniques, RefineScore maintains the advantages of machine learning models while providing physical interpretability and capturing complex interaction patterns. Our evaluation results show that RefineScore has superior scoring and ranking power.
- Open Access
- 9 April 2024
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 15 (21) , 7926-7942
DiffBindFR: an SE(3) equivariant network for flexible protein–ligand docking
DiffBindFR, a diffusion model based flexible full-atom protein–ligand docking tool, demonstrates its superior docking and side-chain refinement accuracy with reliable physical plausibility.
- 9 April 2024
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 15 (21) , 7926-7942
Open AccessDiffBindFR: an SE(3) equivariant network for flexible protein–ligand docking
DiffBindFR, a diffusion model based flexible full-atom protein–ligand docking tool, demonstrates its superior docking and side-chain refinement accuracy with reliable physical plausibility.