Back

A proteomic polygenic score to identify IL-18 driven inflammatory bowel disease

Turchin, M. C.; Raghupathy, N.; Carty, C. L.; Morris, M.; Maranville, J. C.; Holzinger, E. R.

2026-05-21 genetic and genomic medicine
10.64898/2026.05.18.26353508 medRxiv
Show abstract

High levels of IL-18 have been causally implicated in IBD risk and may represent a unique mechanism driving IBD yet to be therapeutically targeted. To identify individuals predisposed to increased levels of IL-18, we implemented a polygenic approach to predict IL-18 plasma protein levels. Using a dataset with over 50,000 individuals with both genetic and plasma protein levels from Olink, we developed a 27 SNP polygenic score that predicts IL-18 levels and IBD risk. Further, we identified a threshold to classify patients as 'IL-18 High' using a data-driven approach that optimized prediction of both IL-18 and IBD risk. We show that ~30% of the overall IBD patient population is 'IL-18 High', meaning a genetic predisposition towards higher protein levels. The IL-18 PGS and corresponding threshold have the potential to identify IBD patients with IL-18-driven IBD that may respond more effectively to a therapy targeting this mechanism.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.