Suyu Dong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Suyu Dong has a total of 72 co-authors. Here are the top co-authors by number of shared publications:
- Kuanquan Wang
Harbin Institute of Technology
14 shared publications - Gongning Luo
School of Computer Science and Technology, Harbin Institute of Technology, Harbin, 150001, China
13 shared publications - Wei Wang
Harbin Institute of Technology
9 shared publications - Xiaokun Li
Heilongjiang University
7 shared publications - Guohua Wang
College of Computer and Control Engineering, Northeast Forestry University, Harbin, China
5 shared publications - Xin Gao
King Abdullah University of Science and Technology, Thuwal, Saudi Arabia
5 shared publications
Latest Publications
- Open Access
- 22 June 2026
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 22 (6) , e1014396
TCRBinder: Unified pre-trained language model with paired-chain synergy for predicting T-cell receptor binding specificity
Deciphering how human T cells recognise peptide-HLA (pHLA) complexes underpins next-generation vaccines and personalised immunotherapies, yet extreme sequence diversity and paired-chains interdependence still hamper reliable in silico prediction of T-cell receptor (TCR) specificity. To overcome these hurdles, we built TCRBinder, a paired-chain-aware deep model with a multi-branch encoder that routes each molecular component through dedicated transformer-based modules to capture contextual signals in both HLA pseudo-sequences and antigenic peptides while simultaneously processing the TCR α and β chains. This design captures the synergistic interaction between paired chains to emulate peptide-HLA-TCR (PHT) interactions and expose residue-level contact motifs. Across PHT and peptide-TCR (pTCR) benchmarks, the model delivered state-of-the-art performance (AUC-ROC = 0.911, AUPR = 0.791 for the PHT task) and remained superior on multiple independent datasets. We tracked the dynamics of clonal expansion and, in a large SARS-CoV-2 repertoire containing completely unseen peptides, improved the AUC-ROC by up to 16.3% over the leading alternatives. Moreover, TCRBinder provided mechanistic insights by pinpointing contact hotspots and quantifying residue contributions to binding probability. These capabilities position TCRBinder as a versatile tool for rational antigen discovery, immunotherapy stratification, and neoantigen vaccine design.
- 22 June 2026
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 22 (6) , e1014396
Open AccessTCRBinder: Unified pre-trained language model with paired-chain synergy for predicting T-cell receptor binding specificity
Deciphering how human T cells recognise peptide-HLA (pHLA) complexes underpins next-generation vaccines and personalised immunotherapies, yet extreme sequence diversity and paired-chains interdependence still hamper reliable in silico prediction of T-cell receptor (TCR) specificity. To overcome these hurdles, we built TCRBinder, a paired-chain-aware deep model with a multi-branch encoder that routes each molecular component through dedicated transformer-based modules to capture contextual signals in both HLA pseudo-sequences and antigenic peptides while simultaneously processing the TCR α and β chains. This design captures the synergistic interaction between paired chains to emulate peptide-HLA-TCR (PHT) interactions and expose residue-level contact motifs. Across PHT and peptide-TCR (pTCR) benchmarks, the model delivered state-of-the-art performance (AUC-ROC = 0.911, AUPR = 0.791 for the PHT task) and remained superior on multiple independent datasets. We tracked the dynamics of clonal expansion and, in a large SARS-CoV-2 repertoire containing completely unseen peptides, improved the AUC-ROC by up to 16.3% over the leading alternatives. Moreover, TCRBinder provided mechanistic insights by pinpointing contact hotspots and quantifying residue contributions to binding probability. These capabilities position TCRBinder as a versatile tool for rational antigen discovery, immunotherapy stratification, and neoantigen vaccine design.
- Top Cited Papers
- 27 March 2026
- Published by Springer Nature in Nature Computational Science
- Vol. 6 (4) , 372-387
HorusEye: a self-supervised foundation model for generalizable X-ray tomography restoration
X-ray tomography is widely used across scientific and clinical domains, yet image degradation remains a major obstacle to reliable analysis, particularly under low-dose or data-scarce conditions. Existing restoration methods are typically designed for specific modalities and predefined degradation, limiting their generalizability. Here we show that image restoration can instead be formulated as learning realistic, nonparametric acquisition degradation processes directly from data. We introduce HorusEye, a self-supervised foundation model for X-ray tomography restoration that leverages interslice contrastive pretraining to jointly learn structural priors and degradation without paired supervision or predefined assumptions. Trained on over 100 million images, HorusEye generalizes across diverse modalities, restoration tasks and previously unseen imaging modalities, consistently outperforming task-specific approaches. Extensive evaluations demonstrate improved photon efficiency and recovery of high-frequency information. Clinical studies further demonstrate enhanced detectability of low-contrast anatomy and lesions, as well as improved performance on downstream tasks, highlighting HorusEye as a general postprocessing tool for X-ray tomography.
- 27 March 2026
- Published by Springer Nature in Nature Computational Science
- Vol. 6 (4) , 372-387
Top Cited PapersHorusEye: a self-supervised foundation model for generalizable X-ray tomography restoration
X-ray tomography is widely used across scientific and clinical domains, yet image degradation remains a major obstacle to reliable analysis, particularly under low-dose or data-scarce conditions. Existing restoration methods are typically designed for specific modalities and predefined degradation, limiting their generalizability. Here we show that image restoration can instead be formulated as learning realistic, nonparametric acquisition degradation processes directly from data. We introduce HorusEye, a self-supervised foundation model for X-ray tomography restoration that leverages interslice contrastive pretraining to jointly learn structural priors and degradation without paired supervision or predefined assumptions. Trained on over 100 million images, HorusEye generalizes across diverse modalities, restoration tasks and previously unseen imaging modalities, consistently outperforming task-specific approaches. Extensive evaluations demonstrate improved photon efficiency and recovery of high-frequency information. Clinical studies further demonstrate enhanced detectability of low-contrast anatomy and lesions, as well as improved performance on downstream tasks, highlighting HorusEye as a general postprocessing tool for X-ray tomography.
- 14 March 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 341
PLATO: ProbabiListic hierArchical mulTi-head mOdel for plug-and-play ambiguous medical image segmentation
- 14 March 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 341
PLATO: ProbabiListic hierArchical mulTi-head mOdel for plug-and-play ambiguous medical image segmentation
- 30 January 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 337
SCULPT: Semantic-aware causal prompt tuning for out-of-distribution detection of whole slide images
- 30 January 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 337
SCULPT: Semantic-aware causal prompt tuning for out-of-distribution detection of whole slide images
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (8) , 1-14
TKRL: Targeted Knowledge Rectification Learning Against Teacher-Originated Defects in Domain Continual Segmentation
Knowledge distillation can mitigate catastrophic forgetting in domain continual segmentation by transferring knowledge from the older model to the newer model. However, existing distillation-based methods primarily emphasize knowledge retention while overlooking inherent defects in the older teacher models. As a result, these teacher-originated defects, such as knowledge gaps or biases, are propagated and exacerbate forgetting. To address this challenge, we propose a Targeted Knowledge Rectification Learning framework (TKRL) to probe and correct teacher-originated defects. TKRL consists of two modules: (1) Probe-augmented Class Distillation, which generates gradient-driven "probes" to uncover underrepresented features in the older model, thereby bridging knowledge gaps by distilling hidden information into the new model; (2) Variance-guided Masked Autoencoder, which selectively masks and reconstructs critical high-uncertainty patches across multi-level semantic regions, thereby correcting biases inherited from the older model. Our experimental results show that TKRL effectively rectifies knowledge gaps and biases, thereby mitigating catastrophic forgetting and enhancing performance in domain continual segmentation. The implementation code is publicly available at: https://github.com/PerceptionComputingLab/TKRL_DCMIS.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (8) , 1-14
Open AccessTKRL: Targeted Knowledge Rectification Learning Against Teacher-Originated Defects in Domain Continual Segmentation
Knowledge distillation can mitigate catastrophic forgetting in domain continual segmentation by transferring knowledge from the older model to the newer model. However, existing distillation-based methods primarily emphasize knowledge retention while overlooking inherent defects in the older teacher models. As a result, these teacher-originated defects, such as knowledge gaps or biases, are propagated and exacerbate forgetting. To address this challenge, we propose a Targeted Knowledge Rectification Learning framework (TKRL) to probe and correct teacher-originated defects. TKRL consists of two modules: (1) Probe-augmented Class Distillation, which generates gradient-driven "probes" to uncover underrepresented features in the older model, thereby bridging knowledge gaps by distilling hidden information into the new model; (2) Variance-guided Masked Autoencoder, which selectively masks and reconstructs critical high-uncertainty patches across multi-level semantic regions, thereby correcting biases inherited from the older model. Our experimental results show that TKRL effectively rectifies knowledge gaps and biases, thereby mitigating catastrophic forgetting and enhancing performance in domain continual segmentation. The implementation code is publicly available at: https://github.com/PerceptionComputingLab/TKRL_DCMIS.
- 26 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (6) , 4920-4931
A Novel Multi-Perspective Framework for Molecule Pretraining: From Atom to Motif Views
Predicting molecular properties is vital for drug discovery, but experimental measurement is costly and limited by scarce labeled data. Self-supervised molecular pretraining can leverage large unlabeled datasets, reducing dependence on extensive annotations. However, most methods struggle to preserve domain-specific chemical knowledge, especially clinically relevant substructures such as motifs. Random masking and generic graph augmentations often degrade critical chemical information and harm interpretability. Many approaches also work at a single scale-either atom or motif-missing opportunities for cross-scale integration. We propose A2M-Mol, a multi-perspective molecular pretraining framework that combines atom-level and motif-level views through four parallel graph constructions. This design enables cross-view alignment and multiscale fusion, explicitly encoding chemical knowledge. A2M-Mol employs a suite of self-supervised tasks, including cross-view correspondence, atomic reconstruction, global topology modeling, and property constraint enforcement, all coordinated via tailored contrastive learning. Extensive experiments across benchmarks and backbone architectures show consistent improvements over state-of-the-art methods. Ablation studies confirm strong synergies among the tasks. A2M-Mol maintains robust predictive accuracy across data scales, demonstrating effectiveness for real-world molecular property prediction and potential to accelerate drug discovery.
- 26 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (6) , 4920-4931
A Novel Multi-Perspective Framework for Molecule Pretraining: From Atom to Motif Views
Predicting molecular properties is vital for drug discovery, but experimental measurement is costly and limited by scarce labeled data. Self-supervised molecular pretraining can leverage large unlabeled datasets, reducing dependence on extensive annotations. However, most methods struggle to preserve domain-specific chemical knowledge, especially clinically relevant substructures such as motifs. Random masking and generic graph augmentations often degrade critical chemical information and harm interpretability. Many approaches also work at a single scale-either atom or motif-missing opportunities for cross-scale integration. We propose A2M-Mol, a multi-perspective molecular pretraining framework that combines atom-level and motif-level views through four parallel graph constructions. This design enables cross-view alignment and multiscale fusion, explicitly encoding chemical knowledge. A2M-Mol employs a suite of self-supervised tasks, including cross-view correspondence, atomic reconstruction, global topology modeling, and property constraint enforcement, all coordinated via tailored contrastive learning. Extensive experiments across benchmarks and backbone architectures show consistent improvements over state-of-the-art methods. Ablation studies confirm strong synergies among the tasks. A2M-Mol maintains robust predictive accuracy across data scales, demonstrating effectiveness for real-world molecular property prediction and potential to accelerate drug discovery.
- Open Access
- 1 November 2025
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 26 (6)
TCRdesign: an antigen-specific generative language model for de novo design of T-cell receptors
T-cell receptors (TCR), which are heterodimers of $\alpha $ and $\beta $ chains that recognize foreign antigens, are of great significance to current immunotherapy. Although artificial intelligence (AI) has explosively accelerated de novo protein design, the challenge of therapeutic TCR design has been overlooked by most researchers. Existing TCR engineering relies heavily on isolating antigen-specific TCRs from tumor tissues, which requires a large amount of labor resources and wet experimental verification. To mitigate this issue, we present TCRdesign, a pretrained generative protein language model (PLM) for the de novo design of artificial TCR $\beta $-chain complementarity-determining region 3 sequences conditioned on antigen-binding specificity (BS). In parallel, we develop a high-accuracy binding predictor (TCRBinder) that couples paired $\alpha $/$\beta $ chain information with antigen sequences to assess BS. Our in silico comparisons demonstrate that (i) TCRdesign surpasses state-of-the-art baselines in generating antigen-specific TCR sequences. The model leverages paired-chain coherence to refine amino-acid level interaction patterns. (ii) TCRdesign-generated TCR sequences exhibit better antigen binding capability to diverse oncogenic hotspots compared with natural counterparts. (iii) TCRdesign inherits the intrinsic properties of large PLMs, enabling effectively identify the determinant residues in TCR-antigen binding, which enhances its interpretability. These results highlight the significant capability of TCRdesign in understanding and generating TCR sequences with an antigen-specific interaction pattern, charting a versatile path toward AI-driven T-cell engineering for precision immunotherapy.- 1 November 2025
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 26 (6)
Open AccessTCRdesign: an antigen-specific generative language model for de novo design of T-cell receptors
T-cell receptors (TCR), which are heterodimers of $\alpha $ and $\beta $ chains that recognize foreign antigens, are of great significance to current immunotherapy. Although artificial intelligence (AI) has explosively accelerated de novo protein design, the challenge of therapeutic TCR design has been overlooked by most researchers. Existing TCR engineering relies heavily on isolating antigen-specific TCRs from tumor tissues, which requires a large amount of labor resources and wet experimental verification. To mitigate this issue, we present TCRdesign, a pretrained generative protein language model (PLM) for the de novo design of artificial TCR $\beta $-chain complementarity-determining region 3 sequences conditioned on antigen-binding specificity (BS). In parallel, we develop a high-accuracy binding predictor (TCRBinder) that couples paired $\alpha $/$\beta $ chain information with antigen sequences to assess BS. Our in silico comparisons demonstrate that (i) TCRdesign surpasses state-of-the-art baselines in generating antigen-specific TCR sequences. The model leverages paired-chain coherence to refine amino-acid level interaction patterns. (ii) TCRdesign-generated TCR sequences exhibit better antigen binding capability to diverse oncogenic hotspots compared with natural counterparts. (iii) TCRdesign inherits the intrinsic properties of large PLMs, enabling effectively identify the determinant residues in TCR-antigen binding, which enhances its interpretability. These results highlight the significant capability of TCRdesign in understanding and generating TCR sequences with an antigen-specific interaction pattern, charting a versatile path toward AI-driven T-cell engineering for precision immunotherapy. - 27 October 2025
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 172, 104948
TransDiffECG: Semantically controllable ECG synthesis via transformer-based diffusion modeling
- 27 October 2025
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 172, 104948
TransDiffECG: Semantically controllable ECG synthesis via transformer-based diffusion modeling
- Open Access
- 12 September 2025
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 21 (9) , e1013050
THLANet: A deep learning framework for predicting TCR-pHLA binding in immunotherapy applications
Adaptive immunity is a targeted immune response that enables the body to identify and eliminate foreign pathogens, playing a critical role in the anti-tumor immune response. Tumor cell expression of antigens forms the foundation for inducing this adaptive response. However, the human leukocyte antigens (HLA)-restricted recognition of antigens by T-cell receptors (TCR) limits their ability to detect all neoantigens, with only a small subset capable of activating T-cells. Accurately predicting neoantigen binding to TCR is, therefore, crucial for assessing their immunogenic potential in clinical settings. We present THLANet, a deep learning model designed to predict the binding specificity of TCR to neoantigens presented by class I HLAs. THLANet employs evolutionary scale modeling-2 (ESM-2), replacing the traditional embedding methods to enhance sequence feature representation. Using scTCR-seq data, we obtained the TCR immune repertoire and constructed a TCR-pHLA binding database to validate THLANet’s clinical potential. The model’s performance was further evaluated using clinical cancer data across various cancer types. Additionally, by analyzing divided complementarity-determining region (CDR3) sequences and simulating alanine scanning of antigen sequences, we provided new insights into the 3D binding interactions of TCRs and antigens. Predicting TCR-neoantigen pairing remains a significant challenge in immunology, THLANet provides accurate predictions using only the TCR sequence (CDR3β), antigen sequence, and class I HLA, offering novel insights into TCR-antigen interactions.
- 12 September 2025
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 21 (9) , e1013050
Open AccessTHLANet: A deep learning framework for predicting TCR-pHLA binding in immunotherapy applications
Adaptive immunity is a targeted immune response that enables the body to identify and eliminate foreign pathogens, playing a critical role in the anti-tumor immune response. Tumor cell expression of antigens forms the foundation for inducing this adaptive response. However, the human leukocyte antigens (HLA)-restricted recognition of antigens by T-cell receptors (TCR) limits their ability to detect all neoantigens, with only a small subset capable of activating T-cells. Accurately predicting neoantigen binding to TCR is, therefore, crucial for assessing their immunogenic potential in clinical settings. We present THLANet, a deep learning model designed to predict the binding specificity of TCR to neoantigens presented by class I HLAs. THLANet employs evolutionary scale modeling-2 (ESM-2), replacing the traditional embedding methods to enhance sequence feature representation. Using scTCR-seq data, we obtained the TCR immune repertoire and constructed a TCR-pHLA binding database to validate THLANet’s clinical potential. The model’s performance was further evaluated using clinical cancer data across various cancer types. Additionally, by analyzing divided complementarity-determining region (CDR3) sequences and simulating alanine scanning of antigen sequences, we provided new insights into the 3D binding interactions of TCRs and antigens. Predicting TCR-neoantigen pairing remains a significant challenge in immunology, THLANet provides accurate predictions using only the TCR sequence (CDR3β), antigen sequence, and class I HLA, offering novel insights into TCR-antigen interactions.
- 20 August 2025
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 196 (Pt C) , 110879
Advancing electrocardiogram synthesis: Analyzing key metrics for enhanced evaluation
- 20 August 2025
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 196 (Pt C) , 110879
Advancing electrocardiogram synthesis: Analyzing key metrics for enhanced evaluation