Transcriptome-wide association study identifies key genes in Graves' ophthalmopathy

Graves' Ophthalmopathy (GO) is classified as an autoimmune condition that involves inflammation and structural changes within the orbit, frequently linked to thyroid abnormalities. The objective of this research is to clarify the genetic foundations associated with GO.

We employed summary statistics derived from a genome-wide association study (GWAS) focusing on Graves' Ophthalmopathy (GO), which included 753 affected individuals and 499,595 controls. To explore the genetic associations with the disease, we conducted an extensive transcriptome-wide association study (TWAS) that merged GWAS findings related to GO with expression quantitative trait loci (eQTL) data obtained from thyroid tissues featured in the Genotype-Tissue Expression Version 8 (GTEx V8) dataset. The validation of significant genes was carried out using three complementary methodologies: FUSION, FOCUS, and Multi-marker Analysis of GenoMic Annotation (MAGMA). To gain further insights into the genetic framework linked to GO, we utilized conditional and joint analysis in conjunction with fine-mapping approaches. Colocalization analysis was performed to identify potential overlapping single nucleotide polymorphisms (SNPs) between the significant genes and GO. In order to determine causality, we executed a Mendelian randomization analysis. The GeneMANIA tool was applied to elucidate gene interactions and their functional roles. Subsequently, we predicted transcription factor target genes, with a particular emphasis on potential ZSCAN9 targets. Finally, functional enrichment analysis revealed significant pathways associated with ZSCAN9 and its related targets.

By employing a combination of three methodologies for TWAS-namely FUSION, FOCUS, and MAGMA-we have identified two candidate genes that demonstrate a significant correlation with GO. Mendelian randomization analysis yielded a suggestive but non-significant association between genetically predicted ZSCAN9 expression and GO, which warrants further investigation. Moreover, colocalization analysis indicates that the ZSCAN9 and HLA-DMA genes may share genetic signals with eQTL, suggesting that these genes could play a role in a shared pathway that contributes to the pathogenesis of GO.

Our findings highlight the genetic architecture of GO, identifying ZSCAN9 and HLA - DMA as potential causal genes, and underscore the importance of integrating multi-omics data to unravel complex disease mechanisms. Subsequent investigations ought to concentrate on corroborating these results and identifying potential therapeutic targets for GO.