Permutation conducted across family members

[Dongmeng-wang]

Hi there, our analysis involves samples from the same family. I was wondering how to conduct permutation taking family structure into account.

One method I can think of is to regress out the relatedness between samples by including family ID as a random effect in the linear mixed effect model (LMM), and then put the residuals got from the LMM model to fastQTL tools and treat them as independent samples and run permutation as usual.

Does this strategy make sense to you, or do you have any other ideas about dealing with relatedness between samples when running permutation with fastQTL? Thank you so much for your help.

All the best, Diana