Why do we have eQTLs?

There is a lot of gene regulatory elements, and variations occur by random change, so it makes sense there are eQTLs? But why are there so many at high frequency? Why are they not selected out more quickly? How robust are we to variations in regulation?

This is a small project and simple, but worth doing.

Make a few graphs where the y-axis is the non-synonymous/synonymous variant ratio for a gene and the x-axis a measure of variation: 1) total expression variation 2) heritability due to all cis-eQTLs, 3) variation due to the top eQTL

I want to see how robust are we to gene expression variation. Are high conserved genes also regulatory conserved?