paperKB
coga / coga-kb
Help
Sign in

Chunk #39 — Methods — Statistical methods. — Partitioning heritability with S-LDSC.

Source
Improving the trans-ancestry portability of polygenic risk scores by prioritizing variants in predicted cell-type-specific regulatory elements.
Embedded
yes

Text

the average per-SNP heritability of a category of SNPs, where a single SNP may claim membership to one or more categories. τ* has units of heritability and is comparable between traits, annotation and populations, because it is normalized for the total heritability (indicative of the power of the GWAS), the dispersion of the annotation values (annotation size) and the number of common SNPs (population specific) considered in the model, respectively. τ, the precursor of τ*, is the coefficient estimated in the S-LDSC regression. τ and τ* are conditionally dependent on the provided baseline-LD annotations. Therefore, the τ* estimate for an IMPACT annotation is considered a measure of cell-type-specific or annotation-specific SNP heritability, as the remaining annotations in the baseline-LD model are not cell-type-specific. Significance of τ* is computed using a z-test of how different the τ* estimate is from 0; the significance of strictly positive τ* estimates are reported in our study. A negative τ* would indicate a depletion of heritability, suggesting that lower values of the annotation are more enriched for trait-associated genetic variation.