Machine learning using genotype and gene-expression data identifies alterations of genes involved in infection susceptibility, antigen presentation and cytokine signalling as key contributors to juvenile idiopathic arthritis (JIA) risk prediction
Authors' conclusion
By applying a machine learning approach to rank SNP-gene-tissue contributions to JIA risk, our findings offer insights into the genetic mechanisms underlying JIA pathogenesis.
Study context
- Population
- JIA-associated single-nucleotide polymorphism (SNPs) from previous genome-wide association studies (GWAS) with spatial eQTL effects mapped across 49 human tissues
- Outcome
- Contribution of each SNP to JIA risk (model weight magnitudes) in a model predicting JIA disease status
Methodological detailstwo-sample Mendelian randomisation, spatial eQTLs as instrumental variables, Lasso-regularised logistic regression, 50 cross-validation runs
- Two-sample Mendelian randomisation
- Spatial eQTLs as instrumental variables
- Lasso-regularised logistic regression
- 50 cross-validation runs
Main finding
As reported in the abstract
Top-ranked SNP-gene-tissue contributions to JIA risk
- Top-ranked SNPs included rs7775055, tagging the human leukocyte antigen (HLA) class II haplotype DRB1*0801-DQA1*0401-DQB1*0402
- Top-ranked SNPs included rs6679677, a non-coding variant in 100% linkage with a coding variant in PTPN22
- Instrumental variables (IVs) for genes implicated in infection-related immune processes (eg, MSH5, MICA and LINC01149) made significant contributions to JIA risk
- Spatial eQTL rs10849448 upregulated the cytokine signalling gene LTBR across all 49 tissues
Original publication
Machine learning using genotype and gene-expression data identifies alterations of genes involved in infection susceptibility, antigen presentation and cytokine signalling as key contributors to JIA risk prediction. RMD Open. Published 9 July 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →