I have generated 30 peers using PEER package on my dataset of 146 samples.
Including all of the peers along with other covariates to run cis-eqtl analysis on my cohorts generated NA in p-values, however if I include only 15 first peers every thing goes well.
Interesting point is if I only include peers #16 -30 everything is ok with it.
Could anyone know what could be the cause pf the issue?