Qiyang Hong
PhD Candidate in Biomedical Engineering (Medical AI)
Institute of Basic Medical Sciences, Peking Union Medical College & Tsinghua University School of Medicine
Interests
Foundation Models
Large Language Models
Multimodal Learning
Temporal Modeling
Representation Learning
Deep Phenotyping
Clinical AI
Model Interpretability
Clinical Bioinformatics
Professional Summary
PhD candidate at the Institute of Basic Medical Sciences, Peking Union Medical College & Tsinghua University School of Medicine (Beijing, China). I build foundation models, large language model evaluation frameworks, and clinical bioinformatics pipelines for precision medicine. My current work spans UK Biobank-scale deep phenotyping, longitudinal multi-omics for COPD, medical LLMs, and clinical genomics workflow development.
Education
PhD in Biomedical Engineering (Medical AI)
Peking Union Medical College & Tsinghua University School of MedicineAdvisor: Prof. Erping Long (Researcher; National Overseas Young Talent). Core expertise: foundation models, LLMs, multimodal and temporal modeling; deep phenotyping, clinical AI; model interpretability; high-performance computing (4xA6000, 2xH100).MSc in Biochemistry and Molecular Biology
Xiamen UniversityAdvisor: Prof. Ning-Shao Xia (Academician, Chinese Academy of Engineering). Specialization: transcriptomics, viral immunology, structural biology, computational molecular analysis.BEng in Bioengineering
Jimei University
First-Author & Co-First-Author Publications
Journal
A foundational model encodes deep phenotyping data and enables diverse downstream applications.
npj Digital Medicine, 2026.
Submitted
Cough, sputum, and CD164: novel risk factors/biomarkers for COPD and lung function decline.
Respiratory Research, 2026.
In revision
Educator-LLM collaboration strategies for discharge education in hematopoietic stem cell transplantation: a randomized controlled trial of real-time prompting.
NEJM AI, 2026.
Journal
Model confrontation and collaboration: a debate intelligence framework for enhancing medical reasoning in large language models.
Cell Reports Medicine, 2025.
Journal
Multi-omics Mendelian Randomization Identifies SERPING1 as a COPD Modulator.
Signal Transduction and Targeted Therapy, 2025.
Under review
Proteo-metabolomic insights into the progression of COPD and lung function decline.
Respiratory Research, 2026.
Journal
Cervical HPV infection and related diseases among 149,559 women in Fujian: an epidemiological study from 2018 to 2023.
Frontiers in Microbiology, 2024.
Co-Authored Publications
In press
GWAS meta-analyses with a new Chinese population expand the genetic architecture and improve ancestry-specific genetic risk prediction of chronic obstructive pulmonary disease.
Genomics Proteomics and Bioinformatics, 2026.
Under review
A mediation-informed causal pathway powers a large language model for individualized health advisory.
Nature Aging, 2026.
Journal
Deep multi-omics profiling reveals three distinct molecular subtypes of chronic obstructive pulmonary disease in a unique biomass-exposed Chinese population.
Med, 2025.
Journal
Genetic determinants of gene expression noise and its role in complex trait variation.
Cell Reports, 2025.
Journal
A stepwise docking molecular dynamics approach for simulating antibody recognition with substantial conformational changes.
Computational and Structural Biotechnology Journal, 2022.
Journal
Identification of strategic residues at the interface of antigen-antibody interactions by in silico mutagenesis.
Interdisciplinary Sciences: Computational Life Sciences, 2018.
Journal
Atomic structures of Coxsackievirus A6 and its complex with a neutralizing antibody.
Nature Communications, 2017.
Journal
A shared N-terminal hydrophobic tail for the formation of nanoparticulates.
Nanomedicine, 2016.
Patents
A method, device, medium and product for user phenotype identification based on hospital clinical data (一种基于医院临床数据的用户表型识别方法、设备、介质及产品)
Invention patent · China (CNIPA) · Application No. 202610660571.0 · Filed, under substantive examination · Inventor 2 of 2
Research & Projects
My research turns foundation models and rigorous LLM evaluation into tools that work on real clinical and genomic data — from UK Biobank-scale deep phenotyping and longitudinal multi-omics in COPD to production bioinformatics pipelines that move from raw sequence to a clinical report.
Selected projects:
- Temporal foundation models for COPD progression (National Science and Technology Major Project, Youth Scientist Program, 2025-2028).
- Biomedical foundation model on UK Biobank deep phenotyping (>500,000 participants), now online in npj Digital Medicine (2026), for disease prediction, multimorbidity analysis, and patient stratification across 289 conditions.
- Automated clinical WES analysis pipeline from FASTQ to SNP/Indel/CNV detection, annotation, ACMG classification, and reporting.
- Neoantigen prediction and immunogenomics pipeline integrating WES and RNA-seq with NetMHC/MHCflurry.
- WES clinical interpretation and visualization platform (Django) for variants, coverage, CNVs, and Sanger traces.
- Consumer genomics analysis framework plus inherited disease and metabolic panel pipelines (thalassemia, SMA, mitochondrial disorders, mtDNA).
- Custom WES probe design optimization for pathogenic ClinVar regions.
- High-throughput transcriptomics analysis with differential expression and KEGG/GO enrichment.
Experience
Molecular Group Leader, Prenatal Diagnosis Center
Affiliated Hospital of Putian UniversityLed molecular diagnostics workflows for prenatal diagnosis, including genetic testing pipelines, clinical reporting, and quality control.Senior Bioinformatics Engineer (Part-time), Data & Information Department
Jinfeng Biotechnology Co., Ltd.Supported data platform development and bioinformatics workflows for genomic products and clinical data operations.Bioinformatics Supervisor, Bioinformatics Department
Berry Genomics (Beijing Berry and Kang Biotechnology Co., Ltd.)Managed pipeline development, clinical interpretation, and quality control for genomic diagnostics, including WES interpretation, CNV detection, ACMG-based classification, and reporting workflows.
Skills
AI / Machine Learning
Foundation Models & LLMs
Multimodal & Temporal Modeling
Representation Learning
Model Interpretability
Model Training & HPC
PyTorch · Distributed Training
GPU Clusters (Linux HPC, A6000/H100)
Large-scale Data & Performance Optimization
Clinical Bioinformatics & Genomics
WES/WGS & RNA-seq Pipelines
Variant Calling & ACMG Interpretation
Immunogenomics / Neoantigen Prediction
Clinical Diagnostics Workflows
Programming & Engineering
Python (NumPy / pandas / scikit-learn)
Reproducible Pipelines & Automation
Django Web Applications
Data Visualization (matplotlib / networkx)
Certifications & Licenses
- Health Professional Qualification (Clinical Laboratory / Medical Testing), National Health Authority of China (2012).
- Clinical PCR Laboratory Technician Certification, Fujian Provincial Clinical Laboratory Center (Sept 2020).
- Bioinformatics Engineer Certification, ICT Support / ICTTT (Jan 2015).
Peer Review Service
Nature Medicine
Co-reviewer (with Prof. Erping Long) ∙
December 2025
Manuscript peer review for clinical and translational medicine research (Springer Nature).
Frontiers in Aging Neuroscience
Co-reviewer (with Prof. Erping Long) ∙
January 2024
Manuscript peer review for aging and neuroscience research.
Nature Biomedical Engineering
Co-reviewer (with Prof. Erping Long) ∙
August 2023
Manuscript peer review for biomedical engineering and AI research (Springer Nature).