Best practices in eQTL analysis

Question

[i.sudbery]

We are in the process of trying to get to grips with some eQTL analysis. I have a intuition that we should be correcting for unknown population structure, but I don't seem to be able to find guides for how to do it. I'm sure this also means there are other consdierations we are missing.

Does anyone know of a good best practices guide/tutorial/review for eQTL: not just how to operate the software (which the vignettes are generally good), but what statical considerations and pit-falls to look out for?

[Asaf]

One pitfall I can mention is reads mapping to pseudogenes better than genes - if in one genome the gene is different than the reference and the reads are more similar to a pseudogene you will end up with reads mapped to a different location.

Answer 1

[rodd]

I think a good starting point is the GTEx documentation, here: https://www.gtexportal.org/home/documentationPage#staticTextAnalysisMethods

They also have published all their scripts (although frankly, I find them a bit complicated, as I switched to bioinformatics in the last 2-3 years or so, so I'm not at that level yet).

[i.sudbery]

Thanks, this is what we ended up going with.