About the Lab
Join a method-driven biomedical AI and genome informatics group.
Professor Degui Zhi, PhD, FACMI, FAIMBE, will join Yale University as Professor in the Department of Biomedical Informatics and Data Science and Director of Bioinformatics.
The lab focuses on computational genomics and pangenomics, clinical EHR deep learning including Med-BERT and CovRNN-style foundation models, and AI-powered imaging genetics and imaging GWAS.
The group has stable NIH support, including R01 and U01 awards, and provides a research environment for independent methodological development and biomedical collaboration.
01
Computational Genomics and Pangenomics
The lab has a sustained research program in large-scale IBD detection and PBWT-based algorithms.
RaPID was introduced as a biobank-scale IBD detection tool with linear-time behavior and orders-of-magnitude speedups.
Subsequent work includes RAFFI for relatedness inference, FiMAP for IBD mapping, ROH analysis, local ancestry inference, and new GBWT/RLBWT-based pangenome indexing methods.
Current projects include efficient pangenome graph construction and query, cross-population haplotype analysis, and next-generation IBD algorithms for million-sample cohorts.
02
EHR Deep Learning and Clinical AI
The lab is an early contributor to structured EHR foundation models.
Med-BERT, published in npj Digital Medicine, introduced pretraining and fine-tuning for diagnosis-code sequences at large scale and established a technical paradigm for clinical AI foundation models.
Current projects include continued pretraining and multi-task fine-tuning for clinical foundation models, deployment-oriented validation and benchmarking, multimodal clinical representation learning, CovRNN and PK-RNN-style trajectory modeling, and biobank-linked genetic discovery.
03
Imaging Genetics and Neuroimaging GWAS
The lab develops UDIP, an unsupervised deep imaging phenotype framework for imaging genetics.
UDIP-Brain used UK Biobank brain MRI to discover genetic loci beyond conventional image-derived phenotypes, and UDIP-FA has been accepted by Nature Communications in 2026.
Current projects include multimodal imaging representation learning, imaging-derived phenotype GWAS, Alzheimer's disease neuroimaging biomarkers, retinal imaging genetics, and multi-omics integration.
Qualifications
- PhD, completed or expected, in bioinformatics, computer science, statistics, computational biology, biomedical informatics, or a related quantitative field.
- Strong programming skills in Python and/or C++.
- Experience with large-scale genomic, clinical, imaging, or biomedical data is preferred.
- Peer-reviewed publication record appropriate for career stage.
- Ability to lead independent research projects while collaborating across disciplines.
- Excellent candidates from adjacent quantitative fields are encouraged to apply.
What We Provide
- Competitive postdoctoral salary aligned with NIH NRSA standards and Yale benefits.
- Access to Yale clinical data resources, large genomic cohorts including ADSP and UK Biobank, and high-performance GPU computing.
- Stable multi-year NIH-supported projects, including R01 and U01 awards.
- Individualized mentoring for academic faculty careers, industry research positions, and national laboratory paths.
- Collaborative environment across Yale School of Medicine and external partner institutions.
How to Apply
Please include the subject line Postdoctoral Application - Your Name.
Applications will be reviewed on a rolling basis until the positions are filled.
- Cover letter, 1-2 pages, describing research interests and fit with the lab.
- Curriculum vitae, including a complete publication list.
- Names and contact information for three references.
Contact: degui.zhi@yale.edu · Yale BIDS, 101 College St, New Haven, Connecticut