Page 112 - Read Online

Ratnapriya. J Transl Genet Genom 2022;6:240-256 https://dx.doi.org/10.20517/jtgg.2021.54 Page 246

Figure 2. Gene expression regulation by a genetic variant. (A) cis-eQTL affects the expression of transcripts from nearby genes
[generally within ± 1 MB of transcription start site (TSS)]. (B) trans-eQTL affects the expression of the distant gene ( > 1 MB) either on
the same chromosome or elsewhere in the genome. eQTL: Expression quantitative trait loci.

Several lines of evidence suggest that a trait-associated variant is more likely to be associated with cis-
eQTLs [78,82,83] . Thus, most analyses of integrating GWAS findings to gene expression regulation have focused
on cis-eQTLs to identify the causal variant and target gene (gene whose expression is affected as a result of
risk variant) [Figure 3]. In the early days, if a GWAS variant was also significant eQTL, it was considered to
be causal. However, these analyses had a high number of false-positive results as genetic architecture
underlying GWAS and eQTL signals were not accounted for. Newer methods integrate eQTL with the
GWAS data to identify genes regulated by GWAS loci using colocalization methods , which apply
[2]
probabilistic analysis in the Bayesian framework for SNP-level colocalization analysis that can provide
molecular insights into complex diseases .
[84]

eQTL STUDIES IN AMD
[80]
Ratnapriya et al. reported the first reference eQTL map of over 500 post-mortem human donor retinae
(from controls and AMD cases at early, intermediate or advanced stages) and reported 14,565 genetic
variants (eVariants) that control the expression of 10,474 genes (eGenes). The integration of AMD-GWAS
data with eQTLs showed that nine lead SNPs at the GWAS loci were significant eQTLs in the retina for 19
SNP-gene associations, and an integrative analysis highlighted the most plausible target genes at six AMD
loci (B3GLCT, BLOC1S1, SH2B3, PLA2G12A, PILRB and POLDIP2/TMEM199) . This data provided a
[80]
basic framework for functional genomic dissection of AMD risk loci and other related ocular traits such as
[80]
glaucoma and diabetic retinopathy . A second study analyzed the gene expression in healthy retina (n =
311) that included the control retina data from the previous study , and reported 403,151 significant eQTL
[80]
[85]
variants (eVariants) that regulate 3007 genes (eGenes) . The integrated results from 16 published GWAS
investigating 12 distinct ocular traits including AMD reported 65 unique GWAS variants to be regulating
gene expression in retinal tissue . Another study comprised of cis-eQTL analysis in human RPE/choroid
[85]
and retina in bulk tissue samples from the macula and non-macular regions in ~120 donors (98 control and