WGS vs WES

[boldizsara]

I've read Novogene’s post: https://www.seqanswers.com/forum/seq...-right-for-you
I'm curious about what could cause such significant differences in the allele frequency (AF) of a variant depending on the sequencing method used. For example, in this variant 11-108250683-CT-C from gnomAD, the allele frequency is 0.1047 when determined by whole exome sequencing (WES) but drops to 0.002328 when using whole genome sequencing (WGS). What factors might account for this discrepancy? (https://gnomad.broadinstitute.org/va...108250683-CT-C)
Since this variant is located within a homopolymer region, I suspect that this might affect the sequencing results. Considering this, would whole-genome sequencing (WGS) or whole-exome sequencing (WES) provide more reliable results in homopolymer regions?

[fchatonnet]

Hi!

Considering that the considered variant is located in an intron, I would rather trust the WGS data, because WES might be based on less reads.
A basic parameter to check is how many reads are supporting each variant, in average, for every genome / individual.
More generally, in the case of homopolymer stretches, I have the feeling that differences wouldn't be so great between WES and WGS (provided the variant is in a coding region), but rather between different sequencing strategies. Short reads approaches (such as Illumina) will always be more prone to errors in homopolymer regions, while sequencing quality may be improved by using long reads sequencing (Oxford Nanopore, PacBio).