Jun Wen    文俊

Post-doc Research Fellow

Department of Biomedical Informatics (DBMI)

Harvard Medical School

Address: Boston, MA, USA, 02138

Email: jun_wen@hms.harvard.edu/jungel2star@gmail.com

[Curriculum Vitae][Google Scholar]


About Me

I am currently a Postdoctoral Research Fellow working with Professor Tianxi Cai at the Translational Data Science Center for a Learning Health System (CELEHS), Harvard Medical School. I also serve as a data scientist at the Veterans Affairs Boston Healthcare System, collaborating with Dr. Kelly Cho. I received my Ph.D. in Computer Science from Zhejiang University in 2020. In January 2026, I will join the Department of Computational Biology at Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) as an Assistant Professor.


Research Interest

My research focuses on developing AI-driven frameworks to advance precision medicine by integrating multimodal biomedical data. These include electronic health records (EHRs), genetic variants, protein interactions, drug-target relationships, and knowledge graphs. I develop methods that leverage techniques such as semi-supervised learning, contrastive learning, knowledge distillation, and hypergraph neural networks to:

  • Characterize phenotype events, disease risks, and temporal medical concept dependencies (e.g., LATTE, SeDDLeR, DOME)
  • Predict the phenotypic consequences of genetic variants (e.g., PheMART)
  • Discover drug-disease-gene relationships for repurposing and safety (e.g., INTERLACE, HERMES, CaSBRE)
I aim to make these models interpretable, label-efficient, and applicable to real-world clinical settings, particularly for rare diseases and pharmacogenomics.

Selected Publications

SAMGEP: A Semi-supervised Adaptive Markov Gaussian Embedding Process for Phenotype Event Prediction
Y. Ahuja*, J. Wen*, ..., T. Cai
Scientific Reports, 12(1), 17737, 2022

LATTE: Label-efficient Incident Phenotyping from Longitudinal EHR
J. Wen, J. Hou, C. Bonzel, Y. Zhao, ..., T. Cai
Patterns (Cell), 5(1), Cover Article, 2024

Label-efficient Phenotyping for Long COVID Using EHR
C. Hong*, J. Wen*, H. Zhang*, ..., T. Cai
npj Digital Medicine, 8(1), 405, 2025

Deep Learning from EHR to Identify RCC Recurrence
J. Hou*, J. Wen*, R. Bhattacharya, ..., T. Cai
Annals of Oncology, 35, S1027, 2024

SeDDLeR: Semi-supervised Double Deep Learning for Temporal Risk Prediction
I. E. Nogues*, J. Wen*, Y. Zhao, ..., T. Cai
Journal of Biomedical Informatics, 157, 104685, 2024

DOME: Directional Medical Embeddings from EHR
J. Wen*, H. Xue*, E. Rush, ..., T. Cai
Journal of Biomedical Informatics, 104768, 2025

Generating Analysis-Ready Data for Real-World Evidence from EHR
J. Hou, R. Zhao, J. Gronsbell, ..., J. Wen, ..., T. Cai
Journal of Medical Internet Research, 25, e45662, 2023

Weakly Semi-supervised Phenotyping Using EHR
I. E. Nogues, J. Wen, Y. Lin, ..., C. Hong
Journal of Biomedical Informatics, 134, 104175, 2022

Phenotypic Prediction of Missense Variants via Deep Contrastive Learning
J. Wen, S. Zeng, ..., T. Cai
Under revision, 2024

Heterogeneous Entity Representation for Medicinal Synergy Prediction
J. Wu*, J. Wen*, M. Yan*, ..., C. Chen
Bioinformatics, 41(1), btae750, 2025

CaSBRE: Causality-inspired Semi-supervised Biomedical Relation Extraction
S. Zeng*, J. Wen*, J. Du, ..., H. Wang
Under revision, 2025

Multimodal Representation Learning for Predicting Molecule–Disease Relations
J. Wen, X. Zhang, E. Rush, ..., T. Cai
Bioinformatics, 39(2), btad085, 2023

HOVER: Hyperbolic Video-Text Retrieval
J. Wen*, Y. Chen*, ..., R. Zimmermann
Under revision, 2024

Contrast-Reconstruction Representation Learning for Self-supervised Skeleton-Based Action Recognition
P. Wang, J. Wen, C. Si, ..., L. Wang
IEEE Transactions on Image Processing, 31, 6224–6238, 2022

Bayesian Uncertainty Matching for Unsupervised Domain Adaptation
J. Wen, N. Zheng, J. Yuan, Z. Gong, C. Chen
IJCAI, 2019

Exploiting Local Feature Patterns for Unsupervised Domain Adaptation
J. Wen, R. Liu, N. Zheng, Q. Zheng, Z. Gong, J. Yuan
AAAI, 2019
[PDF]

Towards More General Loss and Setting in Unsupervised Domain Adaptation
C. Shui, R. Pu, G. Xu, J. Wen, ..., B. Wang
IEEE Transactions on Knowledge and Data Engineering, 35(10), 10140–10150, 2023

Unsupervised Representation Learning with Long-Term Dynamics for Skeleton-Based Action Recognition
J. Wen, N. Zheng, R. Liu, L. Long, J. Dai, Z. Gong
AAAI, 2018
[PDF]

A Two-layer Neural Circuit Controls Fast Forward Locomotion in Drosophila
Q. Zhao, X. Li, J. Wen, Y. He, ..., Z. Gong
Current Biology, 34(15), 3439–3453, 2024