Integrating eQTL data with GWAS summary statistics in pathway-based analysis with application to schizophrenia.

Area of Science:

• Genetics
• Bioinformatics
• Statistical Genomics

Background:

• Complex traits are influenced by numerous genetic variants affecting gene expression.
• Transcriptome-wide association studies (TWAS) leverage gene expression data to enhance genome-wide association studies (GWASs).
• Polygenic inheritance means multiple genes within functional pathways often contribute to complex traits.

Purpose of the Study:

• To extend TWAS from gene-based to pathway-based analysis for identifying gene pathways associated with complex traits.
• To integrate public pathway collections, expression quantitative trait locus (eQTL) data, and GWAS summary statistics.
• To develop a computationally efficient method for pathway association testing.

Main Methods:

• Developed a pathway-based association test by weighting single nucleotide polymorphisms (SNPs) based on their cis-effects on gene expression.
• Adaptively aggregated association signals across genes within pathways.
• Applied the method to KEGG and Gene Ontology (GO) pathways using two schizophrenia (SCZ) GWAS datasets.

Main Results:

• Successfully identified significant pathways associated with schizophrenia in both GWAS datasets.
• Demonstrated reproducibility of findings between the smaller (SCZ1) and larger (SCZ2) datasets.
• Discovered 15 novel pathways associated with SCZ, including the GABA receptor complex (GO:1902710), not detectable by gene-based TWAS or SNP-based analysis.

Conclusions:

• The proposed pathway-based analysis effectively integrates gene expression and functional annotations for GWAS.
• This approach enhances the discovery of biologically relevant pathways underlying complex traits.
• The identified novel pathways offer new insights into the genetic architecture and biological mechanisms of schizophrenia.