How Multi-Omics and a Smarter Bayesian Model Reveal Hidden Risk Genes
Multiple sclerosis (MS) is one of the most enigmatic immune-mediated diseases, affecting over two million people worldwide. Its unpredictable symptoms, diverse subtypes, and a combination of genetic and environmental influences have made it notoriously difficult to decode. In the study by Liu and colleagues, the authors take a major step forward by integrating multiple layers of biological data to pinpoint genes most likely involved in MS risk. Their work combines genomics, epigenomics, gene regulation, chromatin interactions, and cell-type information through a refined Bayesian model.
Refining the iRIGS Bayesian Framework
At the heart of the study lies a powerful updated version of the Integrative Risk Gene Selector (iRIGS), a Bayesian approach that weighs diverse molecular evidence to prioritize potential disease-causing genes. Rather than assuming that the closest gene to a GWAS variant must be causal—a method known to miss long-range regulatory effects—the refined iRIGS algorithm evaluates enhancer interactions, chromatin looping (Hi-C), DNA methylation, tissue-specific expression, and functional networks. This integrative method analyzes over 3,600 candidate genes across 200 MS-associated genomic regions, ultimately identifying 163 high-confidence MS risk genes (MS-PRGenes).
Validating Risk Genes Through Mendelian Randomization
To validate these predictions, the authors employed two-sample Mendelian randomization (2SMR), which tests whether changes in gene expression causally influence MS risk. Out of the 163 prioritized genes, 35 showed significant causal associations, particularly in tissues like whole blood, spleen, and brain cerebellum. Notably, the gene IQGAP1 emerged as a standout candidate, consistently upregulated in MS across 18 of 19 tissues—suggesting a robust and tissue-wide impact on disease mechanisms.
These Genes Aren’t Just Close to the Variants—They’re Biologically Important
What sets MS-PRGenes apart is their biological relevance. The authors found that these genes are more intolerant to loss-of-function mutations (high pLI scores), exhibit slower evolutionary rates, and overlap more frequently with known disease-associated genes than the widely used MAGMA gene set. These features indicate that the MS-PRGenes are not coincidental associations; they are genes under strong evolutionary constraint and likely vital to human biology and immune regulation.
Zooming into Cells: Microglia and Macrophages Take Center Stage
To understand where these genes act, the researchers turned to single-nucleus RNA sequencing (snRNA-seq) data from MS lesions and control brain tissues. They discovered that MS-PRGenes are enriched in activated microglia and macrophages, two cell types central to neuroinflammation and myelin damage in MS. Interestingly, this inflammatory enrichment was absent in MAGMA-identified genes, highlighting the power of the multi-omics Bayesian approach to detect disease-context-specific molecular signals.
Illuminating Biological Pathways: The Immune System Speaks Loudly
Pathway enrichment analysis revealed that MS-PRGenes are heavily involved in immune signaling pathways, especially the toll-like receptor (TLR) pathway, T-cell and B-cell receptor signaling, and chemokine signaling. These pathways are central in orchestrating autoimmune responses. The authors also observed enrichment of cancer-related pathways, echoing previous findings that immune dysregulation can overlap mechanistically with oncogenic processes.
A Step Toward Drug Repositioning for MS
One of the most exciting implications of this study is drug repurposing. By comparing MS-PRGenes with molecular signatures of known drugs (via the Connectivity Map), the authors highlighted compounds such as fisetin, mitoxantrone, monorden, MS-275, and pioglitazone as potentially beneficial in modifying MS-related gene expression patterns. Some of these—like mitoxantrone—are already used in MS, while others, such as fisetin and pioglitazone, have shown promising neuroprotective or anti-inflammatory effects in preclinical models.
Looking Forward: A Benchmark for MS Genetics
This study represents a major leap in MS genomics by combining multi-omics data into a unified framework capable of identifying biologically meaningful risk genes. The 163 MS-PRGenes serve as a high-confidence reference set for future research, offering clearer paths toward understanding disease mechanisms, refining therapeutic targets, and exploring new drug candidates. The authors emphasize that integrating disease-specific epigenomic and chromatin data will continue to improve these predictions—but their current results already provide a powerful foundation for advancing MS research.
Disclaimer: This blog post is based on the provided research article and is intended for informational purposes only. It is not intended to provide medical advice. Please consult with a healthcare professional for any health concerns.
References:
Liu, A., Manuel, A.M., Dai, Y. et al. Prioritization of risk genes in multiple sclerosis by a refined Bayesian framework followed by tissue-specificity and cell type feature assessment. BMC Genomics 23 (Suppl 4), 362 (2022). https://doi.org/10.1186/s12864-022-08580-y
