Minjie Mou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Minjie Mou has a total of 151 co-authors. Here are the top co-authors by number of shared publications:
- Feng Zhu
Zhejiang University
25 shared publications - Huaicheng Sun
College of Pharmaceutical Sciences, Zhejiang University, Hangzhou 310058, China
6 shared publications - Ziqi Pan
Zhejiang University
6 shared publications - Hanyu Zhang
Innovation Institute for Artificial Intelligence in Medicine of Zhejiang University, Alibaba-Zhejiang University Joint Research Center of Future Digital Healthcare, Hangzhou 330110, China
5 shared publications Xiuna Sun
Department of Otolaryngology Head and Neck Surgery, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine , Hangzhou 310058
5 shared publications- Yintao Zhang
College of Pharmaceutical Sciences, The Second Affiliated Hospital, Zhejiang University School of Medicine, State Key Laboratory of Advanced Drug Delivery and Release Systems, Zhejiang University , Hangzhou 310058
5 shared publications
Latest Publications
- Open Access
- 24 October 2026
- Published by Oxford University Press (OUP) in Nucleic Acids Research
TTPdb: a comprehensive database of targeted therapeutic peptides
Targeted therapeutic peptides (TTPs) represent a promising therapeutic modality that bridges the gap between small-molecule drugs and macromolecular antibody-based therapeutics by combining high target specificity, superior tissue permeability, and controllable synthesis. Recent advances in chemical modification and computational design have accelerated the clinical translation of TTPs across diverse therapeutic areas, generating multidimensional data that encompass pharmaceutical profiles, structural binding insights, and structure–activity relationships. However, these data remain fragmented across disparate sources, and no dedicated platform exists to systematically integrate them for mechanistic exploration or rational peptide engineering. Here, we present TTPdb, a specialized database that consolidates comprehensive information on TTPs. TTPdb compiles 5366 TTPs (106 approved, 229 clinical, and 5031 preclinical or investigative), which span >600 disease types and are annotated with 3D structures, binding interactions, and physicochemical properties. TTPdb further catalogs peptide structural modifications and pharmacological mechanisms while establishing associations among peptides, targets, and diseases. TTPdb incorporates a multidimensional search module that supports universal keyword queries, structured conditional filtering, and FASTA-based sequence homology searches, with peptides, targets, and diseases as the primary search entities. As a valuable resource for rational peptide design and optimization in AI-driven biomedical research, TTPdb is freely accessible at https://idrblab.org/TTPdb
- 24 October 2026
- Published by Oxford University Press (OUP) in Nucleic Acids Research
Open AccessTTPdb: a comprehensive database of targeted therapeutic peptides
Targeted therapeutic peptides (TTPs) represent a promising therapeutic modality that bridges the gap between small-molecule drugs and macromolecular antibody-based therapeutics by combining high target specificity, superior tissue permeability, and controllable synthesis. Recent advances in chemical modification and computational design have accelerated the clinical translation of TTPs across diverse therapeutic areas, generating multidimensional data that encompass pharmaceutical profiles, structural binding insights, and structure–activity relationships. However, these data remain fragmented across disparate sources, and no dedicated platform exists to systematically integrate them for mechanistic exploration or rational peptide engineering. Here, we present TTPdb, a specialized database that consolidates comprehensive information on TTPs. TTPdb compiles 5366 TTPs (106 approved, 229 clinical, and 5031 preclinical or investigative), which span >600 disease types and are annotated with 3D structures, binding interactions, and physicochemical properties. TTPdb further catalogs peptide structural modifications and pharmacological mechanisms while establishing associations among peptides, targets, and diseases. TTPdb incorporates a multidimensional search module that supports universal keyword queries, structured conditional filtering, and FASTA-based sequence homology searches, with peptides, targets, and diseases as the primary search entities. As a valuable resource for rational peptide design and optimization in AI-driven biomedical research, TTPdb is freely accessible at https://idrblab.org/TTPdb
- Open Access
- 23 October 2026
- Published by Oxford University Press (OUP) in Nucleic Acids Research
scSpatial: an atlas for single-cell spatial omics
Single-cell spatial omics (SCSO) has reshaped spatial profiling by enabling simultaneous molecular characterization and spatial mapping at single-cell resolution, providing insights into cellular identities and tissue architecture within native biological contexts. By integrating spatial coordinates with molecular profiles and cell identities, SCSO datasets have been widely used to construct spatially resolved cell atlases, characterize disease-associated tissue remodeling, and map disease-relevant cell populations. However, these SCSO datasets remain fragmented across individual publications, and comprehensive SCSO resources are still lacking. Here, scSpatial was therefore developed as an atlas for SCSO with expert-curated cell annotations and high-resolution spatial mappings. The current release of scSpatial curates a large-scale collection of SCSO data, comprising over 1800 tissue sections from 630 projects generated using 16 spatial technologies; provides >20 million spatially resolved cells annotated into 173 standardized Cell Ontology types; and offers SCSO-tailored analyses and interactive visualizations, including spatial gene expression atlases, cell type distribution maps, spatially variable genes, gene expression correlations, and pathway-level spatial activity profiles. scSpatial is freely accessible to all users without requiring login at https://idrblab.org/scspatial/.
- 23 October 2026
- Published by Oxford University Press (OUP) in Nucleic Acids Research
Open AccessscSpatial: an atlas for single-cell spatial omics
Single-cell spatial omics (SCSO) has reshaped spatial profiling by enabling simultaneous molecular characterization and spatial mapping at single-cell resolution, providing insights into cellular identities and tissue architecture within native biological contexts. By integrating spatial coordinates with molecular profiles and cell identities, SCSO datasets have been widely used to construct spatially resolved cell atlases, characterize disease-associated tissue remodeling, and map disease-relevant cell populations. However, these SCSO datasets remain fragmented across individual publications, and comprehensive SCSO resources are still lacking. Here, scSpatial was therefore developed as an atlas for SCSO with expert-curated cell annotations and high-resolution spatial mappings. The current release of scSpatial curates a large-scale collection of SCSO data, comprising over 1800 tissue sections from 630 projects generated using 16 spatial technologies; provides >20 million spatially resolved cells annotated into 173 standardized Cell Ontology types; and offers SCSO-tailored analyses and interactive visualizations, including spatial gene expression atlases, cell type distribution maps, spatially variable genes, gene expression correlations, and pathway-level spatial activity profiles. scSpatial is freely accessible to all users without requiring login at https://idrblab.org/scspatial/.
- Open Access
- 30 September 2026
- Published by Oxford University Press (OUP) in Nucleic Acids Research
SingPro 2.0: a proteomics-centric knowledge base for single-cell multimodal profiling
Protein-inclusive single-cell multimodal profiling has emerged as a transformative strategy for resolving cellular heterogeneity by jointly measuring proteins and complementary molecular features in the same cells. The resulting protein-inclusive datasets are increasingly used to characterize cellular states, construct multimodal cell atlases, and decode disease mechanisms. However, such data are rarely included in existing databases. In this study, SingPro was updated from a single-cell proteomics database into a proteomics-centric knowledge base for single-cell multimodal profiling. Distinct from existing resources, the updated SingPro (a) systematically curates data and experimental details from 814 protein-inclusive multimodal studies encompassing ~81.5 million cells across 29 tissues, 44 diseases, and multiple species; (b) covers 5 omics modalities through 12 experimental technologies that simultaneously profile proteins with messenger RNA, chromatin accessibility, or other measurements in the same cell; and (c) provides cross-modality visualization for joint analysis. SingPro is freely accessible without requiring login or registration at: https://idrblab.org/singpro/.
- 30 September 2026
- Published by Oxford University Press (OUP) in Nucleic Acids Research
Open AccessSingPro 2.0: a proteomics-centric knowledge base for single-cell multimodal profiling
Protein-inclusive single-cell multimodal profiling has emerged as a transformative strategy for resolving cellular heterogeneity by jointly measuring proteins and complementary molecular features in the same cells. The resulting protein-inclusive datasets are increasingly used to characterize cellular states, construct multimodal cell atlases, and decode disease mechanisms. However, such data are rarely included in existing databases. In this study, SingPro was updated from a single-cell proteomics database into a proteomics-centric knowledge base for single-cell multimodal profiling. Distinct from existing resources, the updated SingPro (a) systematically curates data and experimental details from 814 protein-inclusive multimodal studies encompassing ~81.5 million cells across 29 tissues, 44 diseases, and multiple species; (b) covers 5 omics modalities through 12 experimental technologies that simultaneously profile proteins with messenger RNA, chromatin accessibility, or other measurements in the same cell; and (c) provides cross-modality visualization for joint analysis. SingPro is freely accessible without requiring login or registration at: https://idrblab.org/singpro/.
- 12 August 2026
- Published by American Chemical Society (ACS) in Journal of Medicinal Chemistry
- Vol. 69 (16) , 20037-20049
Accurate Identification of Covalently Ligandable Cysteines Using CCSite
Targeted covalent inhibition is an important strategy in modern drug discovery, with cysteine being the most common residue targeted for covalent ligands. Accurate identification of covalently ligandable cysteines is therefore essential, especially for traditionally “undruggable” targets. However, structure-based methods depend on available and reliable protein structures, while sequence-based methods remain scarce and require further improvement. Here, we present CCSite, a protein language model-based framework for discovering covalently ligandable cysteines from protein sequences. It uniquely integrates low-rank adaptation of ESM Cambrian (LoRA-ESMC) with a cysteine-centered encoder-decoder module to capture local microenvironment features and long-range contextual information. Benchmarking and independent evaluation showed that CCSite achieved competitive performance without requiring 3D structures. Moreover, a real-world application revealed that CCSite was capable of prospectively identifying experimentally validated covalent cysteines, and a large-scale screen of human pathogenic X-to-Cys mutations further identified over 2,000 neo-cysteines as covalently ligandable candidates for future covalent drug development.
- 12 August 2026
- Published by American Chemical Society (ACS) in Journal of Medicinal Chemistry
- Vol. 69 (16) , 20037-20049
Accurate Identification of Covalently Ligandable Cysteines Using CCSite
Targeted covalent inhibition is an important strategy in modern drug discovery, with cysteine being the most common residue targeted for covalent ligands. Accurate identification of covalently ligandable cysteines is therefore essential, especially for traditionally “undruggable” targets. However, structure-based methods depend on available and reliable protein structures, while sequence-based methods remain scarce and require further improvement. Here, we present CCSite, a protein language model-based framework for discovering covalently ligandable cysteines from protein sequences. It uniquely integrates low-rank adaptation of ESM Cambrian (LoRA-ESMC) with a cysteine-centered encoder-decoder module to capture local microenvironment features and long-range contextual information. Benchmarking and independent evaluation showed that CCSite achieved competitive performance without requiring 3D structures. Moreover, a real-world application revealed that CCSite was capable of prospectively identifying experimentally validated covalent cysteines, and a large-scale screen of human pathogenic X-to-Cys mutations further identified over 2,000 neo-cysteines as covalently ligandable candidates for future covalent drug development.
- Open Access
- 5 January 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Measuring and locating the changes in protein structure using MELO
Understanding the impact that subtle variations (missense mutation, environmental change, ion chelation, ligand binding, etc.) have on protein structure helps to reveal their biological effects, but remain extremely challenging due to the difficulty in measuring and locating the changes in protein structure. Herein, a method entitled MELO is therefore constructed, which enable a systematic measurement based on residues’ geometric characteristics & relative distance and a high-throughput location of structural change based on secondary structure variation & protein segment shift. Our method performs best in capturing the structure changes of various degrees of magnitude (some increases were >30%) and is capable of precisely locating the regions of alterations for critical case studies. Moreover, it identifies over 10,000 structural changes induced by subtle variation that existing methods fail to detect. An online server allows users to upload their structures for comparison, and all those structural changes identified in this study have also been made available for download.
- 5 January 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessMeasuring and locating the changes in protein structure using MELO
Understanding the impact that subtle variations (missense mutation, environmental change, ion chelation, ligand binding, etc.) have on protein structure helps to reveal their biological effects, but remain extremely challenging due to the difficulty in measuring and locating the changes in protein structure. Herein, a method entitled MELO is therefore constructed, which enable a systematic measurement based on residues’ geometric characteristics & relative distance and a high-throughput location of structural change based on secondary structure variation & protein segment shift. Our method performs best in capturing the structure changes of various degrees of magnitude (some increases were >30%) and is capable of precisely locating the regions of alterations for critical case studies. Moreover, it identifies over 10,000 structural changes induced by subtle variation that existing methods fail to detect. An online server allows users to upload their structures for comparison, and all those structural changes identified in this study have also been made available for download.
- Open Access
- 27 December 2025
- Published by Wiley in Advanced Science
- Vol. 13 (10) , e16530
Accurate Identification of Protein Binding Sites for All Drug Modalities Using ALLSites
Proteins interact with diverse molecular modalities, yet the incomplete identification of their binding sites has left the proteome‐wide druggability largely underexplored. Although various computational methods have been developed for the prediction of protein binding sites, existing approaches are limited by their specificity to a single drug modality, dependence on high‐quality structural data, or insufficient predictive accuracy. Here, a unified sequence‐based framework, ALLSites, is constructed to identify proteome‐wide binding sites across all drug modalities. Leveraging ESM‐2 embeddings, ALLSites integrates a gated convolutional network with a transformer architecture to capture both global and local sequence features, effectively modeling residue interactions directly from sequence. This design bridges the gap between sequence‐based and structure‐based approaches, enabling ALLSites to achieve superior predictive performance across diverse drug modalities, including proteins, peptides, small molecules, carbohydrates, DNA, and RNA. It achieves state‐of‐the‐art performance among sequence‐based methods and matches the accuracy of the best structure‐based tools. By enabling accurate and structure‐free binding site prediction across all drug modalities, ALLSites is expected to expand the druggable proteome and provide a powerful resource for drug discovery.
- 27 December 2025
- Published by Wiley in Advanced Science
- Vol. 13 (10) , e16530
Open AccessAccurate Identification of Protein Binding Sites for All Drug Modalities Using ALLSites
Proteins interact with diverse molecular modalities, yet the incomplete identification of their binding sites has left the proteome‐wide druggability largely underexplored. Although various computational methods have been developed for the prediction of protein binding sites, existing approaches are limited by their specificity to a single drug modality, dependence on high‐quality structural data, or insufficient predictive accuracy. Here, a unified sequence‐based framework, ALLSites, is constructed to identify proteome‐wide binding sites across all drug modalities. Leveraging ESM‐2 embeddings, ALLSites integrates a gated convolutional network with a transformer architecture to capture both global and local sequence features, effectively modeling residue interactions directly from sequence. This design bridges the gap between sequence‐based and structure‐based approaches, enabling ALLSites to achieve superior predictive performance across diverse drug modalities, including proteins, peptides, small molecules, carbohydrates, DNA, and RNA. It achieves state‐of‐the‐art performance among sequence‐based methods and matches the accuracy of the best structure‐based tools. By enabling accurate and structure‐free binding site prediction across all drug modalities, ALLSites is expected to expand the druggable proteome and provide a powerful resource for drug discovery.
- Open Access
- 16 October 2025
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 54 (D1) , D1608-D1615
RadioPharm: the database of radiopharmaceuticals
Radiopharmaceuticals (RPs) have emerged as a promising group of therapeutic/diagnostic agents with tremendous pharmaceutical advantages and profitable market value. With the rapid advances in this direction, there is an increasing demand for a knowledge base providing RP-relevant data. However, no such database has been available yet. Herein, a database RadioPharm was therefore developed to depict the pharmaceutical data for 3141 RPs and the radioactive characteristics for the nuclide in each RP. Particularly, our RadioPharm is unique in (i) providing the largest number of RPs (including 68 approved, 575 in clinical trials, and 2498 investigative RPs) among the existing databases, (ii) offering the comprehensive radioactive characteristics for the radionuclide in each RP, and (iii) implementing a multidirectional approach for data retrieval (from diverse perspectives of RP, radionuclide, target, and indication). Because of the important role of all those accumulated data in radiopharmaceutical research, RadioPharm is expected to attract broad interest from research community, which is now fully accessible by all users at https://idrblab.org/radiopharm/.
- 16 October 2025
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 54 (D1) , D1608-D1615
Open AccessRadioPharm: the database of radiopharmaceuticals
Radiopharmaceuticals (RPs) have emerged as a promising group of therapeutic/diagnostic agents with tremendous pharmaceutical advantages and profitable market value. With the rapid advances in this direction, there is an increasing demand for a knowledge base providing RP-relevant data. However, no such database has been available yet. Herein, a database RadioPharm was therefore developed to depict the pharmaceutical data for 3141 RPs and the radioactive characteristics for the nuclide in each RP. Particularly, our RadioPharm is unique in (i) providing the largest number of RPs (including 68 approved, 575 in clinical trials, and 2498 investigative RPs) among the existing databases, (ii) offering the comprehensive radioactive characteristics for the radionuclide in each RP, and (iii) implementing a multidirectional approach for data retrieval (from diverse perspectives of RP, radionuclide, target, and indication). Because of the important role of all those accumulated data in radiopharmaceutical research, RadioPharm is expected to attract broad interest from research community, which is now fully accessible by all users at https://idrblab.org/radiopharm/.
- Open Access
- 12 May 2025
- Published by Elsevier in Journal of Pharmaceutical Analysis
- Vol. 15 (8) , 101338
Discovery of selective HDAC6 inhibitors driven by artificial intelligence and molecular dynamics simulation approaches
- 12 May 2025
- Published by Elsevier in Journal of Pharmaceutical Analysis
- Vol. 15 (8) , 101338
Open AccessDiscovery of selective HDAC6 inhibitors driven by artificial intelligence and molecular dynamics simulation approaches
- Open Access
- 9 April 2025
- Published by Elsevier in Journal of Pharmaceutical Analysis
- Vol. 15 (6) , 101298
druglikeFilter 1.0: An AI powered filter for collectively measuring the drug-likeness of compounds
- 9 April 2025
- Published by Elsevier in Journal of Pharmaceutical Analysis
- Vol. 15 (6) , 101298
Open AccessdruglikeFilter 1.0: An AI powered filter for collectively measuring the drug-likeness of compounds
- Open Access
- 1 January 2025
- Published by American Association for the Advancement of Science (AAAS) in Research
- Vol. 8, 0615
Deep Learning for Predicting Biomolecular Binding Sites of Proteins
The rapid evolution of deep learning has markedly enhanced protein–biomolecule binding site prediction, offering insights essential for drug discovery, mutation analysis, and molecular biology. Advancements in both sequence-based and structure-based methods demonstrate their distinct strengths and limitations. Sequence-based approaches offer efficiency and adaptability, while structure-based techniques provide spatial precision but require high-quality structural data. Emerging trends in hybrid models that combine multimodal data, such as integrating sequence and structural information, along with innovations in geometric deep learning, present promising directions for improving prediction accuracy. This perspective summarizes challenges such as computational demands and dynamic modeling and proposes strategies for future research. The ultimate goal is the development of computationally efficient and flexible models capable of capturing the complexity of real-world biomolecular interactions, thereby broadening the scope and applicability of binding site predictions across a wide range of biomedical contexts.
- 1 January 2025
- Published by American Association for the Advancement of Science (AAAS) in Research
- Vol. 8, 0615
Open AccessDeep Learning for Predicting Biomolecular Binding Sites of Proteins
The rapid evolution of deep learning has markedly enhanced protein–biomolecule binding site prediction, offering insights essential for drug discovery, mutation analysis, and molecular biology. Advancements in both sequence-based and structure-based methods demonstrate their distinct strengths and limitations. Sequence-based approaches offer efficiency and adaptability, while structure-based techniques provide spatial precision but require high-quality structural data. Emerging trends in hybrid models that combine multimodal data, such as integrating sequence and structural information, along with innovations in geometric deep learning, present promising directions for improving prediction accuracy. This perspective summarizes challenges such as computational demands and dynamic modeling and proposes strategies for future research. The ultimate goal is the development of computationally efficient and flexible models capable of capturing the complexity of real-world biomolecular interactions, thereby broadening the scope and applicability of binding site predictions across a wide range of biomedical contexts.