Functional annotation of noncoding causal variants in autoimmune diseases

Area of Science:

• Genetics
• Immunology
• Computational Biology

Background:

• Noncoding variants constitute the majority of Genome-wide association studies (GWAS) findings but are challenging to interpret.
• Assigning causal nucleotide changes from GWAS loci to specific diseases remains a significant hurdle.
• Understanding the mechanisms and physiological roles of noncoding elements is crucial for disease genetics.

Purpose of the Study:

• To develop and apply a computational approach for identifying causal noncoding variants associated with autoimmune diseases.
• To investigate the impact of specific genetic variants on the expression of key genes involved in autoimmune conditions.
• To refine the interpretation of GWAS results by integrating epigenomic data.

Main Methods:

• Utilized high-density genotyping and epigenomic data.
• Employed a random forest model for the discovery of noncoding causal variants.
• Developed a filtering pipeline to triage putative causal variants for atopic dermatitis and inflammatory bowel diseases.

Main Results:

• Identified three single nucleotide polymorphisms (rs1800630, rs1799964, rs4796793) upstream of TNF and STAT3 genes.
• Demonstrated that these variants influence the expression levels of TNF and STAT3.
• These genes are frequently implicated in various autoimmune diseases.

Conclusions:

• The combination of genotyping and epigenomic data with machine learning effectively identifies causal noncoding variants.
• Specific noncoding variants near TNF and STAT3 play a role in autoimmune disease pathogenesis by modulating gene expression.
• This approach offers a powerful strategy for interpreting complex genetic associations in human diseases.