AI tool predicts autoimmune disease risk using genomic data

A research team from Pennsylvania State University has developed an AI model to gain insight into how gene expression impacts autoimmune disease risk, according to a study published recently in Nature Communications.

The tool, EXpression PREdiction with Summary Statistics Only (EXPRESSO), is designed to incorporate single-cell expression quantitative trait loci, 3D genomic data and epigenetics to model how autoimmune disease-associated genes are expressed and regulated. This information can then be used to flag additional genes of risk and improve therapies.

“We all carry some DNA mutations, and we need to figure out how any one of these mutations may influence gene expression linked to disease so we can predict disease risk early. This is especially important for autoimmune disease,” said co-senior author of the study Dajiang Liu, PhD, distinguished professor, vice chair for research and director of artificial intelligence and biomedical informatics at the Penn State College of Medicine, in a news release. “If an AI algorithm can more accurately predict disease risk, it means we can carry out interventions earlier.”

The EXPRESSO approach combat a major limitation inherent in genome-wide association studies (GWAS): the inability to pinpoint specific genes that influence disease risk. The researchers indicated that the GWAS approach is useful for homing in on regions within the genome associated with different diseases or traits, but additional granularity is necessary for diseases with complex trait risk genes.

The research team indicated that gene expression is often specific to certain cell types, meaning that approaches like GWAS, which cannot distinguish between cell types, cannot surface valuable causal relationships between gene expression and genetic variants.

EXPRESSO’s ability to analyze genomic data at the cellular level could help bridge this gap.

The researchers applied the tool to 14 GWAS datasets for a host of autoimmune diseases, including ulcerative colitis, lupus, rheumatoid arthritis and Crohn’s disease.

When compared to existing methodologies, EXPRESSO successfully identified over 25% more novel gene and trait associations.

“With this new method, we were able to identify many more risk genes for autoimmune disease that actually have cell-type specific effects, meaning that they only have effects in a particular cell type and not others,” said Bibo Jiang, PhD, assistant professor at the Penn State College of Medicine and senior author of the study.

These insights could be used to improve existing therapeutics for these diseases and identify new potential treatments.

“Most treatments are designed to mitigate symptoms, not cure the disease. It’s a dilemma knowing that autoimmune disease needs long-term treatment, but the existing treatments often have such bad side effects that they can’t be used for long. Yet, genomics and AI offer a promising route to develop novel therapeutics,” said co-senior author of the study Laura Carrel, PhD, professor of biochemistry and molecular biology at the Penn State College of Medicine.

Using EXPRESSO, the research team found that FDA-approved drug compounds like metformin and vitamin K could be repurposed to help reverse gene expression in cell types associated with diseases like type 1 diabetes and ulcerative colitis.

Moving forward, the researchers will work to validate the tool in lab settings and clinical trials.