Allele-specific expression (ASE) analysis
I’ve heard this term for years, but never once I bothered to know what it actually is. I guess I assumed that it’s just a fancy name for a linear regression of the allele counts on the gene expression levels. Recently, while I was glancing through the GTEx long-read RNA sequencing preprint, it struck me that I didn’t know what ASE analysis means. When I read about it, I was fascinated by how simple yet powerful the concept behind ASE analysis is. It’s a molecular version of within-family GWAS that is gathering attention recently in the behavioral genetics field.
ASE analysis is possible only for heterozygous variants. When a variant is heterozygous, then one can simply compare the RNA transcript levels between two alleles and precisely quantify the effect of the variant on the RNA abundance. This is powerful because here the genetic background remains constant and so no confounding due to differences in the genetic backgrounds (e.g. different ancestries).
If you’re wondering why the same cannot be done for GWAS phenotypes like I did: remember, the RNA transcripts can be traced back to their DNA strand of origin (based on which allele it carries). So, we can measure the RNA transcript levels allele-specifically and compare the transcript levels between the two alleles. And meta-analyse the effect size across all heterozygous individuals, I guess. This is not possible for GWAS phenotypes.
