I was attempting to impute the ROSMAP data genotyped with the Affy chip (https://doi.org/10.7303/syn2580853) but it fails for chr11 and chr12 due to call rates of <50% for one or more individuals. There do indeed seem to be multiple participants with high rates of missingness on these chromosomes, which I believe tended to be an issue for data coming from one of the sites. I was wondering how people have dealt with this in the past when imputing? I can use the imputed data provided here, but should those participants be filtered out anyway given high rate of missing genotypes? Thank you in advance for your help, Jeremy