IMMUNOLOGY2026™ Conference Recordings For Attendees-GPS integration of electronic health record and GWAS data to predict autoimmune disease progression risk

GPS integration of electronic health record and GWAS data to predict autoimmune disease progression risk

Video Transcription

Video Summary

The speaker discussed using machine learning, genomics, and electronic health record data to improve diagnosis and prediction in autoimmune disease, especially lupus. He emphasized the importance of the preclinical stage, when patients show early immune activity but have not yet met full diagnostic criteria. Current tools like antibody tests and polygenic risk scores have limitations in sensitivity, specificity, or ability to predict progression. To address this, his team developed a “Genetic Progression Score” (GPS) that combines case-control genetic studies with longitudinal biobank data to better predict who will progress from preclinical autoimmunity to full disease. The method outperformed alternatives, revealed meaningful disease associations and pathways, and supported drug repurposing findings, including known and potential treatments. He concluded that integrating AI, omics, and EHR data can reveal biology, improve risk prediction, and guide new therapies.

Meta Tag

Date April 16, 2026 8:00 AM - 8:23 AM

Room 205

Session Artificial Intelligence Approaches in Immune-Mediated Disease, Sponsored by DAIT, AMIB, NIAID, NIH

Speaker Dajiang Liu

Track Computational and Systems Immunology (COMP)

Year 2026

Keywords

machine learning

genomics

electronic health records

autoimmune disease

lupus

genetic progression score

April 16, 2026 8:00 AM - 8:23 AM

205

Artificial Intelligence Approaches in Immune-Mediated Disease, Sponsored by DAIT, AMIB, NIAID, NIH

Dajiang Liu

Computational and Systems Immunology (COMP)

2026