Second low variance error after filtering --mac 100 with Plink

[jdarias93]

Hello, I'm using Regenie v3.2.9 I run the low variance plink fix after getting the error running Step 1:

plink2 \ --bfile subjects_sex \ --chr 1-23 \ --mac 100 \ --write-snplist \ --out subjects

Here is my step 1: regenie \ --step 1 \ --bed ${DIRECTORY}/Genotype/${BEDPREFIX} \ --covarFile ${DIRECTORY}/Covariates_and_Phenotypes/${COVARIATE} \ --phenoFile ${DIRECTORY}/Covariates_and_Phenotypes/${PHENOTYPE} \ --extract ${DIRECTORY}/Genotype/${BEDPREFIX}.snplist \ --force-step1 \ --spa --bsize 1000 --bt \ --out ./${OUTPUTNOM}

Then when I run the updated plink bim/bam/fam file through Step1, I get a SECOND low variance error. FYI, there are >100,000 variants, which forces me to use --force-step1. I'm not sure if that's the issue, however. Please let me know if any more info is needed.

[joellembatchou]

Hi,

Are you also passing the IDs of samples with non-missing phenotypes and covariates (e.g. this can be obtained using --write-samples)?

Cheers, Joelle