Wenyi Yang
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
Wenyi Yang has a total of 45 co-authors. Here are the top co-authors by number of shared publications:
- Qinghua Jiang
Center for Bioinformatics, School of Life Science and Technology, Harbin Institute of Technology, Harbin 150080, China
23 shared publications - Pingping Wang
School of Interdisciplinary Medicine and Engineering, Harbin Medical University, Harbin, China
22 shared publications - Zhaochun Xu
School of Interdisciplinary Medicine and Engineering, Harbin Medical University , 157 Baojian Road, Nangang District, Harbin 150081, Heilongjiang Province
22 shared publications - Wenyang Zhou
Harbin Medical University 2 School of Interdisciplinary Medicine and Engineering Harbin, China
21 shared publications - Xiyun Jin
School of Interdisciplinary Medicine and Engineering, Harbin Medical University, Harbin, China
21 shared publications - Guangfu Xue
Harbin Institute of Technology 1 Center for Bioinformatics, School of Life Science and Technology Harbin, China
20 shared publications
Latest Publications
- 30 October 2026
- Published by American Association for Cancer Research (AACR) in Cancer Research
- p. OF1-OF12
ScanTCR Enables Early Cancer Detection by Repertoire-Scale Representation of Cancer-Associated T-cell Receptors in Peripheral Blood
T cells initiate anticancer immune responses at an early stage of tumorigenesis, thus enabling dynamic monitoring of peripheral blood T-cell receptor (TCR) repertoire changes as a promising strategy for early cancer detection. However, the current TCR-based cancer detection methods focus on limited high-frequency TCRs instead of repertoire-scale representation, overlooking numerous crucial cancer-associated TCRs (caTCR). Herein, we developed ScanTCR, a computational framework for early cancer detection that introduces a reference-guided repertoire-scale encoding method capable of incorporating all TCRs. ScanTCR exhibited remarkable precision in distinguishing patients with cancer from noncancer individuals using peripheral blood TCRs. In a pan-cancer analysis encompassing seven cancer types, ScanTCR demonstrated robust performance in detecting both cancer types included in the training data and unseen cancer types, surpassing state-of-the-art TCR-based methods. Furthermore, ScanTCR identified caTCRs in which corresponding T cells exhibit enhanced cellular immune capacity. Collectively, this work proposes a TCR-based approach enabling repertoire-scale representation of immune receptors for noninvasive early cancer detection. Significance: Overcoming limitations of current methods that favor high-frequency cancer-associated T-cell receptors, repertoire-scale profiling preserves all sequences for accurate cancer detection, supporting the potential of this approach for early-stage cancer diagnostics.
- 30 October 2026
- Published by American Association for Cancer Research (AACR) in Cancer Research
- p. OF1-OF12
ScanTCR Enables Early Cancer Detection by Repertoire-Scale Representation of Cancer-Associated T-cell Receptors in Peripheral Blood
T cells initiate anticancer immune responses at an early stage of tumorigenesis, thus enabling dynamic monitoring of peripheral blood T-cell receptor (TCR) repertoire changes as a promising strategy for early cancer detection. However, the current TCR-based cancer detection methods focus on limited high-frequency TCRs instead of repertoire-scale representation, overlooking numerous crucial cancer-associated TCRs (caTCR). Herein, we developed ScanTCR, a computational framework for early cancer detection that introduces a reference-guided repertoire-scale encoding method capable of incorporating all TCRs. ScanTCR exhibited remarkable precision in distinguishing patients with cancer from noncancer individuals using peripheral blood TCRs. In a pan-cancer analysis encompassing seven cancer types, ScanTCR demonstrated robust performance in detecting both cancer types included in the training data and unseen cancer types, surpassing state-of-the-art TCR-based methods. Furthermore, ScanTCR identified caTCRs in which corresponding T cells exhibit enhanced cellular immune capacity. Collectively, this work proposes a TCR-based approach enabling repertoire-scale representation of immune receptors for noninvasive early cancer detection. Significance: Overcoming limitations of current methods that favor high-frequency cancer-associated T-cell receptors, repertoire-scale profiling preserves all sequences for accurate cancer detection, supporting the potential of this approach for early-stage cancer diagnostics.
- Open Access
- 24 October 2026
- Published by Oxford University Press (OUP) in Genomics, Proteomics & Bioinformatics
RNASTOP: A Deep Learning Framework for mRNA Chemical Stability Prediction and Optimization
Messenger RNA (mRNA) vaccines offer promising therapeutic potential for combating various diseases, yet their inherent chemical instability hampers long-term efficacy. Although several methods have been developed to predict mRNA degradation, they often suffer from limited accuracy and lack the capability for rational codon optimization. Here, we propose RNASTOP, a novel framework that integrates deep learning with heuristic search to simultaneously predict and optimize mRNA chemical stability. RNASTOP achieves a 13% improvement in accuracy over the top-performing model on the Stanford OpenVaccine competition dataset and demonstrates robust generalization in predicting full-length mRNA degradation. When applied to mRNA codon optimization, RNASTOP reduces the minimum free energy of the varicella-zoster virus vaccine sequence by 75.73% while maintaining high translation efficiency. Overall, RNASTOP serves as a powerful tool for predicting and optimizing mRNA chemical stability, accelerating the development of mRNA therapeutics. The source code of RNASTOP is available at https://github.com/xlab-BioAI/RNASTOP.
- 24 October 2026
- Published by Oxford University Press (OUP) in Genomics, Proteomics & Bioinformatics
Open AccessRNASTOP: A Deep Learning Framework for mRNA Chemical Stability Prediction and Optimization
Messenger RNA (mRNA) vaccines offer promising therapeutic potential for combating various diseases, yet their inherent chemical instability hampers long-term efficacy. Although several methods have been developed to predict mRNA degradation, they often suffer from limited accuracy and lack the capability for rational codon optimization. Here, we propose RNASTOP, a novel framework that integrates deep learning with heuristic search to simultaneously predict and optimize mRNA chemical stability. RNASTOP achieves a 13% improvement in accuracy over the top-performing model on the Stanford OpenVaccine competition dataset and demonstrates robust generalization in predicting full-length mRNA degradation. When applied to mRNA codon optimization, RNASTOP reduces the minimum free energy of the varicella-zoster virus vaccine sequence by 75.73% while maintaining high translation efficiency. Overall, RNASTOP serves as a powerful tool for predicting and optimizing mRNA chemical stability, accelerating the development of mRNA therapeutics. The source code of RNASTOP is available at https://github.com/xlab-BioAI/RNASTOP.
- Open Access
RNASTOP: A Deep Learning Framework for mRNA Chemical Stability Prediction and Optimization
Messenger RNA (mRNA) vaccines offer promising therapeutics for combating various diseases, yet their inherent chemical instability hampers their long-term efficacy. Although several methods have been developed to predict mRNA degradation, they exhibit limited accuracy and lack the capability for rational sequence optimization. Here, we propose RNASTOP, a novel framework integrating deep learning with heuristic search to simultaneously predict and optimize mRNA chemical stability. RNASTOP achieves a 13% accuracy improvement over the top-performing model on the Stanford OpenVaccine competition dataset and demonstrates robust generalization in predicting full-length mRNA degradation. Applied to mRNA codon optimization, RNASTOP reduces the minimum free energy of the Varicella-Zoster Virus vaccine sequence by 75.73% while maintaining high translation efficiency. Overall, RNASTOP serves as a powerful tool for predicting and optimizing mRNA chemical stability, poised to expedite the development of mRNA therapeutics. The source code of RNASTOP can be accessed at https://github.com/xlab-BioAI/RNASTOP .
Open AccessRNASTOP: A Deep Learning Framework for mRNA Chemical Stability Prediction and Optimization
Messenger RNA (mRNA) vaccines offer promising therapeutics for combating various diseases, yet their inherent chemical instability hampers their long-term efficacy. Although several methods have been developed to predict mRNA degradation, they exhibit limited accuracy and lack the capability for rational sequence optimization. Here, we propose RNASTOP, a novel framework integrating deep learning with heuristic search to simultaneously predict and optimize mRNA chemical stability. RNASTOP achieves a 13% accuracy improvement over the top-performing model on the Stanford OpenVaccine competition dataset and demonstrates robust generalization in predicting full-length mRNA degradation. Applied to mRNA codon optimization, RNASTOP reduces the minimum free energy of the Varicella-Zoster Virus vaccine sequence by 75.73% while maintaining high translation efficiency. Overall, RNASTOP serves as a powerful tool for predicting and optimizing mRNA chemical stability, poised to expedite the development of mRNA therapeutics. The source code of RNASTOP can be accessed at https://github.com/xlab-BioAI/RNASTOP .
- Open Access
- 7 January 2026
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 27 (1)
Systematic evaluation of computational methods for cell segmentation
Cell segmentation plays a crucial role in elucidating cell structure and function, understanding disease mechanisms, and aiding pathological diagnosis. Current surveys primarily categorize methods by their technical evolution stages, which may not fully capture the paradigm shift brought by deep learning. Moreover, their evaluation scope is largely confined to image-only approaches, overlooking the significant potential of multimodal data in enhancing cell/nucleus segmentation performance. Therefore, we propose a dual-dimensional classification framework for deep learning methods. It categorizes such methods into two types: task-oriented (e.g. semantic or instance segmentation) and data-oriented (e.g. single or multimodal inputs). Based on this, we systematically classify and summarize methods across various segmentation tasks and imaging modalities. We also develop a benchmark test that covers both single-modal and multimodal methods. This test uses five diverse datasets, among which four are from conventional microscopy and one integrates sequencing with image data. Furthermore, it assesses seven algorithms based on three dimensions: effectiveness, robustness, and efficiency. Key findings indicate that deep learning models generally outperform traditional algorithms, with their advantage becoming more pronounced when image data is integrated with sequencing information.
- 7 January 2026
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 27 (1)
Open AccessSystematic evaluation of computational methods for cell segmentation
Cell segmentation plays a crucial role in elucidating cell structure and function, understanding disease mechanisms, and aiding pathological diagnosis. Current surveys primarily categorize methods by their technical evolution stages, which may not fully capture the paradigm shift brought by deep learning. Moreover, their evaluation scope is largely confined to image-only approaches, overlooking the significant potential of multimodal data in enhancing cell/nucleus segmentation performance. Therefore, we propose a dual-dimensional classification framework for deep learning methods. It categorizes such methods into two types: task-oriented (e.g. semantic or instance segmentation) and data-oriented (e.g. single or multimodal inputs). Based on this, we systematically classify and summarize methods across various segmentation tasks and imaging modalities. We also develop a benchmark test that covers both single-modal and multimodal methods. This test uses five diverse datasets, among which four are from conventional microscopy and one integrates sequencing with image data. Furthermore, it assesses seven algorithms based on three dimensions: effectiveness, robustness, and efficiency. Key findings indicate that deep learning models generally outperform traditional algorithms, with their advantage becoming more pronounced when image data is integrated with sequencing information.
- Open Access
- 4 June 2024
- Published by American Association for Cancer Research (AACR)
Supplementary Table S3 from The Deep Learning Framework iCanTCR Enables Early Cancer Detection Using the T-cell Receptor Repertoire in Peripheral Blood
Sample information of non-cancer cohorts in Emerson's study
- 4 June 2024
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table S3 from The Deep Learning Framework iCanTCR Enables Early Cancer Detection Using the T-cell Receptor Repertoire in Peripheral Blood
Sample information of non-cancer cohorts in Emerson's study
- Open Access
- 4 June 2024
- Published by American Association for Cancer Research (AACR)
Supplementary Table S6 from The Deep Learning Framework iCanTCR Enables Early Cancer Detection Using the T-cell Receptor Repertoire in Peripheral Blood
Results of iCanTCR performance to detect three early-stage cancer types using PBMC TCR repertoire
- 4 June 2024
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table S6 from The Deep Learning Framework iCanTCR Enables Early Cancer Detection Using the T-cell Receptor Repertoire in Peripheral Blood
Results of iCanTCR performance to detect three early-stage cancer types using PBMC TCR repertoire
- Open Access
- 4 June 2024
- Published by American Association for Cancer Research (AACR)
Supplementary Table S2 from The Deep Learning Framework iCanTCR Enables Early Cancer Detection Using the T-cell Receptor Repertoire in Peripheral Blood
Information of benchmark testing set
- 4 June 2024
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table S2 from The Deep Learning Framework iCanTCR Enables Early Cancer Detection Using the T-cell Receptor Repertoire in Peripheral Blood
Information of benchmark testing set
- Open Access
- 4 June 2024
- Published by American Association for Cancer Research (AACR)
Supplementary Table S2 from The Deep Learning Framework iCanTCR Enables Early Cancer Detection Using the T-cell Receptor Repertoire in Peripheral Blood
Information of benchmark testing set
- 4 June 2024
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table S2 from The Deep Learning Framework iCanTCR Enables Early Cancer Detection Using the T-cell Receptor Repertoire in Peripheral Blood
Information of benchmark testing set
- Open Access
- 4 June 2024
- Published by American Association for Cancer Research (AACR)
Supplementary Table S6 from The Deep Learning Framework iCanTCR Enables Early Cancer Detection Using the T-cell Receptor Repertoire in Peripheral Blood
Results of iCanTCR performance to detect three early-stage cancer types using PBMC TCR repertoire
- 4 June 2024
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table S6 from The Deep Learning Framework iCanTCR Enables Early Cancer Detection Using the T-cell Receptor Repertoire in Peripheral Blood
Results of iCanTCR performance to detect three early-stage cancer types using PBMC TCR repertoire