Haohuai He
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Haohuai He has a total of 58 co-authors. Here are the top co-authors by number of shared publications:
- Calvin Yu-Chian Chen
AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China
7 shared publications - Guanxing Chen
Department of Computer Science, City University of Hong Kong (Dongguan), Dongguan, China
4 shared publications Guanxing Chen
State Key Laboratory of Chemical Oncogenomics, Key Laboratory of Chemical Genomics, School of Chemical Biology and Biotechnology, Peking University Shenzhen Graduate School, Shenzhen, China
3 shared publications- Bing He
AI for Life Sciences Lab, Tencent, Shenzhen, China
2 shared publications - Jianhua Yao
AI for Life Sciences Lab, Tencent, Shenzhen, China
2 shared publications - Linlin You
School of Intelligent Systems Engineering, Shenzhen Campus of Sun Yat-sen University, Shenzhen, China
2 shared publications
Latest Publications
- Open Access
- 23 October 2026
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 42 (11)
AbLWR: a context-aware listwise ranking framework for antibody-antigen binding affinity prediction via positive-unlabeled learning
Motivation: Accurate prediction of antibody-antigen binding affinity is fundamental to therapeutic design and the rapid response to evolving pathogens. However, current computational approaches predominantly formulate this as a point-wise regression task. This paradigm struggles to discriminate subtle affinity differences among highly homologous variants and lacks the global context necessary to effectively prioritize top-tier candidates. Furthermore, the field is severely constrained by label sparsity. Although sequence data is abundant, experimentally verified affinities are scarce, leaving a massive pool of unverified antibody-antigen structural interactions unexploited by standard supervised paradigms. Results: We propose AbLWR (Antibody-antigen binding affinity Listwise Ranking), a novel framework that reformulates the conventional affinity regression task as a listwise ranking problem to directly optimize candidate prioritization. To mitigate label sparsity, AbLWR incorporates a Positive-Unlabeled (PU) learning mechanism, leveraging a dual-level contrastive objective and meta-optimized label refinement to distill robust structural representations from vast unlabeled data. To accommodate antigenic variation, we employ a homologous antigen sampling strategy coupled with a Multi-Head Self-Attention (MHSA) module, which explicitly models inter-sample dependencies within training lists to capture subtle affinity nuances. Extensive evaluations demonstrate that AbLWR significantly outperforms state-of-the-art sequence-, structure-, and complex-based baselines. It improves the Full Rank Accuracy (FRA) by approximately 8% compared to the strongest baseline in randomized cross-validation, and exhibits strong robustness in out-of-distribution scenarios involving distinct antigen and antibody clusters. Case studies on Influenza Viruses and Human Interleukin-33 further validate its practical utility. Specifically, AbLWR successfully tracks affinity drops caused by antigenic drift and immediately captures optimal binders in simulated wet-lab screening. Availability: Source code and datasets are freely available at https://github.com/Josie-xufan/AbLWR.
- 23 October 2026
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 42 (11)
Open AccessAbLWR: a context-aware listwise ranking framework for antibody-antigen binding affinity prediction via positive-unlabeled learning
Motivation: Accurate prediction of antibody-antigen binding affinity is fundamental to therapeutic design and the rapid response to evolving pathogens. However, current computational approaches predominantly formulate this as a point-wise regression task. This paradigm struggles to discriminate subtle affinity differences among highly homologous variants and lacks the global context necessary to effectively prioritize top-tier candidates. Furthermore, the field is severely constrained by label sparsity. Although sequence data is abundant, experimentally verified affinities are scarce, leaving a massive pool of unverified antibody-antigen structural interactions unexploited by standard supervised paradigms. Results: We propose AbLWR (Antibody-antigen binding affinity Listwise Ranking), a novel framework that reformulates the conventional affinity regression task as a listwise ranking problem to directly optimize candidate prioritization. To mitigate label sparsity, AbLWR incorporates a Positive-Unlabeled (PU) learning mechanism, leveraging a dual-level contrastive objective and meta-optimized label refinement to distill robust structural representations from vast unlabeled data. To accommodate antigenic variation, we employ a homologous antigen sampling strategy coupled with a Multi-Head Self-Attention (MHSA) module, which explicitly models inter-sample dependencies within training lists to capture subtle affinity nuances. Extensive evaluations demonstrate that AbLWR significantly outperforms state-of-the-art sequence-, structure-, and complex-based baselines. It improves the Full Rank Accuracy (FRA) by approximately 8% compared to the strongest baseline in randomized cross-validation, and exhibits strong robustness in out-of-distribution scenarios involving distinct antigen and antibody clusters. Case studies on Influenza Viruses and Human Interleukin-33 further validate its practical utility. Specifically, AbLWR successfully tracks affinity drops caused by antigenic drift and immediately captures optimal binders in simulated wet-lab screening. Availability: Source code and datasets are freely available at https://github.com/Josie-xufan/AbLWR.
- Open Access
- 28 August 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (35)
scProtoTransformer: Scalable reference mapping across molecules, cells, and donors
The rapid accumulation of single-cell data has made it possible to comprehensively characterize biological systems at molecular, cellular, and donor levels. However, scalable reference mapping across different resolutions remains a major challenge in current research. Here, we propose scProtoTransformer, a prototype-based Transformer architecture designed to achieve scalable reference mapping across molecular, cell, and donor levels. scProtoTransformer introduces a knowledge-guided prototype tokenizer that projects gene expression into biologically interpretable pathway prototypes, effectively reducing numerical batch effects while preserving biological semantic patterns. Furthermore, by leveraging knowledge distilled from the foundation model and a dynamic supervised fine-tuning strategy, scProtoTransformer achieves robust biological representations with reduced pretraining requirements. Benchmark experiments across molecular, cell, and donor-level reference mapping demonstrate that scProtoTransformer delivers competitive or even superior performance compared with state-of-the-art approaches while providing interpretability through biological prototypes. Together, these results establish scProtoTransformer as a unified framework for scalable reference mapping, laying the foundation for systematic understanding from genes to individuals.
- 28 August 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (35)
Open AccessscProtoTransformer: Scalable reference mapping across molecules, cells, and donors
The rapid accumulation of single-cell data has made it possible to comprehensively characterize biological systems at molecular, cellular, and donor levels. However, scalable reference mapping across different resolutions remains a major challenge in current research. Here, we propose scProtoTransformer, a prototype-based Transformer architecture designed to achieve scalable reference mapping across molecular, cell, and donor levels. scProtoTransformer introduces a knowledge-guided prototype tokenizer that projects gene expression into biologically interpretable pathway prototypes, effectively reducing numerical batch effects while preserving biological semantic patterns. Furthermore, by leveraging knowledge distilled from the foundation model and a dynamic supervised fine-tuning strategy, scProtoTransformer achieves robust biological representations with reduced pretraining requirements. Benchmark experiments across molecular, cell, and donor-level reference mapping demonstrate that scProtoTransformer delivers competitive or even superior performance compared with state-of-the-art approaches while providing interpretability through biological prototypes. Together, these results establish scProtoTransformer as a unified framework for scalable reference mapping, laying the foundation for systematic understanding from genes to individuals.
- Open Access
- 28 January 2025
- Published by Springer Nature in npj Digital Medicine
- Vol. 8 (1) , 1-14
Dual modality feature fused neural network integrating binding site information for drug target affinity prediction
Accurately predicting binding affinities between drugs and targets is crucial for drug discovery but remains challenging due to the complexity of modeling interactions between small drug and large targets. This study proposes DMFF-DTA, a dual-modality neural network model integrates sequence and graph structure information from drugs and proteins for drug-target affinity prediction. The model introduces a binding site-focused graph construction approach to extract binding information, enabling more balanced and efficient modeling of drug-target interactions. Comprehensive experiments demonstrate DMFF-DTA outperforms state-of-the-art methods with significant improvements. The model exhibits excellent generalization capabilities on completely unseen drugs and targets, achieving an improvement of over 8% compared to existing methods. Model interpretability analysis validates the biological relevance of the model. A case study in pancreatic cancer drug repurposing demonstrates its practical utility. This work provides an interpretable, robust approach to integrate multi-view drug and protein features for advancing computational drug discovery.
- 28 January 2025
- Published by Springer Nature in npj Digital Medicine
- Vol. 8 (1) , 1-14
Open AccessDual modality feature fused neural network integrating binding site information for drug target affinity prediction
Accurately predicting binding affinities between drugs and targets is crucial for drug discovery but remains challenging due to the complexity of modeling interactions between small drug and large targets. This study proposes DMFF-DTA, a dual-modality neural network model integrates sequence and graph structure information from drugs and proteins for drug-target affinity prediction. The model introduces a binding site-focused graph construction approach to extract binding information, enabling more balanced and efficient modeling of drug-target interactions. Comprehensive experiments demonstrate DMFF-DTA outperforms state-of-the-art methods with significant improvements. The model exhibits excellent generalization capabilities on completely unseen drugs and targets, achieving an improvement of over 8% compared to existing methods. Model interpretability analysis validates the biological relevance of the model. A case study in pancreatic cancer drug repurposing demonstrates its practical utility. This work provides an interpretable, robust approach to integrate multi-view drug and protein features for advancing computational drug discovery.
- Open Access
- 5 November 2024
- Published by Springer Nature in Communications Biology
- Vol. 7 (1) , 1-10
Knowledge-based inductive bias and domain adaptation for cell type annotation
Measurement techniques often result in domain gaps among batches of cellular data from a specific modality. The effectiveness of cross-batch annotation methods is influenced by inductive bias, which refers to a set of assumptions that describe the behavior of model predictions. Different annotation methods possess distinct inductive biases, leading to varying degrees of generalizability and interpretability. Given that certain cell types exhibit unique functional patterns, we hypothesize that the inductive biases of cell annotation methods should align with these biological patterns to produce meaningful predictions. In this study, we propose KIDA, Knowledge-based Inductive bias and Domain Adaptation. The knowledge-based inductive bias constrains the prediction rules learned from the reference dataset, composed of multiple batches, to functional patterns relevant to biology, thereby enhancing the generalization of the model to unseen batches. Since the query dataset also contains gaps from multiple batches, KIDA’s domain adaptation employs pseudo labels for self-knowledge distillation, effectively narrowing the distribution gap between model predictions and the query dataset. Benchmark experiments demonstrate that KIDA is capable of achieving accurate cross-batch cell type annotation.
- 5 November 2024
- Published by Springer Nature in Communications Biology
- Vol. 7 (1) , 1-10
Open AccessKnowledge-based inductive bias and domain adaptation for cell type annotation
Measurement techniques often result in domain gaps among batches of cellular data from a specific modality. The effectiveness of cross-batch annotation methods is influenced by inductive bias, which refers to a set of assumptions that describe the behavior of model predictions. Different annotation methods possess distinct inductive biases, leading to varying degrees of generalizability and interpretability. Given that certain cell types exhibit unique functional patterns, we hypothesize that the inductive biases of cell annotation methods should align with these biological patterns to produce meaningful predictions. In this study, we propose KIDA, Knowledge-based Inductive bias and Domain Adaptation. The knowledge-based inductive bias constrains the prediction rules learned from the reference dataset, composed of multiple batches, to functional patterns relevant to biology, thereby enhancing the generalization of the model to unseen batches. Since the query dataset also contains gaps from multiple batches, KIDA’s domain adaptation employs pseudo labels for self-knowledge distillation, effectively narrowing the distribution gap between model predictions and the query dataset. Benchmark experiments demonstrate that KIDA is capable of achieving accurate cross-batch cell type annotation.
- Open Access
- 10 August 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-19
De novo generation of SARS-CoV-2 antibody CDRH3 with a pre-trained generative large language model
Artificial Intelligence (AI) techniques have made great advances in assisting antibody design. However, antibody design still heavily relies on isolating antigen-specific antibodies from serum, which is a resource-intensive and time-consuming process. To address this issue, we propose a Pre-trained Antibody generative large Language Model (PALM-H3) for the de novo generation of artificial antibodies heavy chain complementarity-determining region 3 (CDRH3) with desired antigen-binding specificity, reducing the reliance on natural antibodies. We also build a high-precision model antigen-antibody binder (A2binder) that pairs antigen epitope sequences with antibody sequences to predict binding specificity and affinity. PALM-H3-generated antibodies exhibit binding ability to SARS-CoV-2 antigens, including the emerging XBB variant, as confirmed through in-silico analysis and in-vitro assays. The in-vitro assays validate that PALM-H3-generated antibodies achieve high binding affinity and potent neutralization capability against spike proteins of SARS-CoV-2 wild-type, Alpha, Delta, and the emerging XBB variant. Meanwhile, A2binder demonstrates exceptional predictive performance on binding specificity for various epitopes and variants. Furthermore, by incorporating the attention mechanism inherent in the Roformer architecture into the PALM-H3 model, we improve its interpretability, providing crucial insights into the fundamental principles of antibody design.
- 10 August 2024
- Published by Springer Nature in Nature Communications
- Vol. 15 (1) , 1-19
Open AccessDe novo generation of SARS-CoV-2 antibody CDRH3 with a pre-trained generative large language model
Artificial Intelligence (AI) techniques have made great advances in assisting antibody design. However, antibody design still heavily relies on isolating antigen-specific antibodies from serum, which is a resource-intensive and time-consuming process. To address this issue, we propose a Pre-trained Antibody generative large Language Model (PALM-H3) for the de novo generation of artificial antibodies heavy chain complementarity-determining region 3 (CDRH3) with desired antigen-binding specificity, reducing the reliance on natural antibodies. We also build a high-precision model antigen-antibody binder (A2binder) that pairs antigen epitope sequences with antibody sequences to predict binding specificity and affinity. PALM-H3-generated antibodies exhibit binding ability to SARS-CoV-2 antigens, including the emerging XBB variant, as confirmed through in-silico analysis and in-vitro assays. The in-vitro assays validate that PALM-H3-generated antibodies achieve high binding affinity and potent neutralization capability against spike proteins of SARS-CoV-2 wild-type, Alpha, Delta, and the emerging XBB variant. Meanwhile, A2binder demonstrates exceptional predictive performance on binding specificity for various epitopes and variants. Furthermore, by incorporating the attention mechanism inherent in the Roformer architecture into the PALM-H3 model, we improve its interpretability, providing crucial insights into the fundamental principles of antibody design.
- Open Access
- 1 December 2023
- Published by Wiley in Advanced Intelligent Systems
- Vol. 6 (3)
MFAE: Multilevel Feature Aggregation Enhanced Drug‐Target Affinity Prediction for Drug Repurposing Against Colorectal Cancer
Colorectal cancer (CRC), a leading cause of cancer‐related deaths globally, demands innovative therapeutic strategies to improve patient outcomes. Drug repurposing, identifying new uses for existing drugs, provides a cost‐effective solution. To this end, this study constructs the first drug‐target affinity dataset specifically for two novel therapeutic targets for CRC, P2X4 and mTOR, and designs a new deep learning‐based multilevel feature aggregation enhanced (MFAE) model. The model implements hierarchical feature extraction and multilevel feature aggregation and enhancement to simulate complex drug‐target interactions. Evaluations using fivefold cross‐validation on the collected CRC dataset showcase MFAE's superior predictive accuracy. Fine‐tuning of the model on external experimental data further enhances its performance, with a concordance index of 0.930, a determination coefficient of 0.782, and a mean squared error of 0.191. Ablation studies further highlight the key role of the group‐wise feature enhancement mechanism and ensemble learning strategy in enhancing the model's performance. Virtual screening of the Food and Drug Administration‐approved drugs identifies Ponatinib and Talazoparib as potential repurposing candidates. Despite limitations in experimental validation, this study establishes an innovative computational framework designed for CRC drug discovery. Overall, this research offers a valuable perspective on leveraging computational approaches for precision oncology.
- 1 December 2023
- Published by Wiley in Advanced Intelligent Systems
- Vol. 6 (3)
Open AccessMFAE: Multilevel Feature Aggregation Enhanced Drug‐Target Affinity Prediction for Drug Repurposing Against Colorectal Cancer
Colorectal cancer (CRC), a leading cause of cancer‐related deaths globally, demands innovative therapeutic strategies to improve patient outcomes. Drug repurposing, identifying new uses for existing drugs, provides a cost‐effective solution. To this end, this study constructs the first drug‐target affinity dataset specifically for two novel therapeutic targets for CRC, P2X4 and mTOR, and designs a new deep learning‐based multilevel feature aggregation enhanced (MFAE) model. The model implements hierarchical feature extraction and multilevel feature aggregation and enhancement to simulate complex drug‐target interactions. Evaluations using fivefold cross‐validation on the collected CRC dataset showcase MFAE's superior predictive accuracy. Fine‐tuning of the model on external experimental data further enhances its performance, with a concordance index of 0.930, a determination coefficient of 0.782, and a mean squared error of 0.191. Ablation studies further highlight the key role of the group‐wise feature enhancement mechanism and ensemble learning strategy in enhancing the model's performance. Virtual screening of the Food and Drug Administration‐approved drugs identifies Ponatinib and Talazoparib as potential repurposing candidates. Despite limitations in experimental validation, this study establishes an innovative computational framework designed for CRC drug discovery. Overall, this research offers a valuable perspective on leveraging computational approaches for precision oncology.
- Open Access
- 11 August 2023
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 9 (32) , eabo5128
DeepAIR: A deep learning framework for effective integration of sequence and 3D structure to enable adaptive immune receptor analysis
Structural docking between the adaptive immune receptors (AIRs), including T cell receptors (TCRs) and B cell receptors (BCRs), and their cognate antigens are one of the most fundamental processes in adaptive immunity. However, current methods for predicting AIR-antigen binding largely rely on sequence-derived features of AIRs, omitting the structure features that are essential for binding affinity. In this study, we present a deep learning framework, termed DeepAIR, for the accurate prediction of AIR-antigen binding by integrating both sequence and structure features of AIRs. DeepAIR achieves a Pearson’s correlation of 0.813 in predicting the binding affinity of TCR, and a median area under the receiver-operating characteristic curve (AUC) of 0.904 and 0.942 in predicting the binding reactivity of TCR and BCR, respectively. Meanwhile, using TCR and BCR repertoire, DeepAIR correctly identifies every patient with nasopharyngeal carcinoma and inflammatory bowel disease in test data. Thus, DeepAIR improves the AIR-antigen binding prediction that facilitates the study of adaptive immunity.
- 11 August 2023
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 9 (32) , eabo5128
Open AccessDeepAIR: A deep learning framework for effective integration of sequence and 3D structure to enable adaptive immune receptor analysis
Structural docking between the adaptive immune receptors (AIRs), including T cell receptors (TCRs) and B cell receptors (BCRs), and their cognate antigens are one of the most fundamental processes in adaptive immunity. However, current methods for predicting AIR-antigen binding largely rely on sequence-derived features of AIRs, omitting the structure features that are essential for binding affinity. In this study, we present a deep learning framework, termed DeepAIR, for the accurate prediction of AIR-antigen binding by integrating both sequence and structure features of AIRs. DeepAIR achieves a Pearson’s correlation of 0.813 in predicting the binding affinity of TCR, and a median area under the receiver-operating characteristic curve (AUC) of 0.904 and 0.942 in predicting the binding reactivity of TCR and BCR, respectively. Meanwhile, using TCR and BCR repertoire, DeepAIR correctly identifies every patient with nasopharyngeal carcinoma and inflammatory bowel disease in test data. Thus, DeepAIR improves the AIR-antigen binding prediction that facilitates the study of adaptive immunity.
- Top Cited PapersOpen Access
- 8 August 2023
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 14 (39) , 10684-10701
TCMBank: bridges between the largest herbal medicines, chemical ingredients, target proteins, and associated diseases with intelligence text mining
We developed TCMBank which contains 9192 herbs, 61 966 unduplicated ingredients, 15 179 targets, 32 529 diseases, and their pairwise relationships. We developed an ensemble learning-based drug discovery protocol for identifying potential lead.
- 8 August 2023
- Published by Royal Society of Chemistry (RSC) in Chemical Science
- Vol. 14 (39) , 10684-10701
Top Cited PapersOpen AccessTCMBank: bridges between the largest herbal medicines, chemical ingredients, target proteins, and associated diseases with intelligence text mining
We developed TCMBank which contains 9192 herbs, 61 966 unduplicated ingredients, 15 179 targets, 32 529 diseases, and their pairwise relationships. We developed an ensemble learning-based drug discovery protocol for identifying potential lead.
- 1 November 2014
- Published by Elsevier in European Journal of Cancer
- Vol. 50, 95
287 The discovery and optimization of small molecule antagonists of the WDR5–MLL interaction
- 1 November 2014
- Published by Elsevier in European Journal of Cancer
- Vol. 50, 95
287 The discovery and optimization of small molecule antagonists of the WDR5–MLL interaction