|
Post by kristi2901 on Dec 4, 2017 17:59:04 GMT
Hi!
I would really appriciate the help to understand why is my code not working. I have ~3000 WGS individuals and approximately ~30000 pruned SNPs (obtained by -clump procedure implemented in PLINK using the same WGS dataset). I wanted to estimate joint effects without model selection. I tried to to this separately for each chromosome gcta64 --bfile chr${i}.wgs --chr ${i} --extract snplist_full_gcta.txt --cojo-file metabo_full_gcta.txt --cojo-joint --out meta_full_chr${i}
and I recieve a following error for couple of chromosomes: Error: too many SNPs. The number of SNPs in a joint analysis should not be larger than the sample size.
I am puzzled by this, as to my understanding, WGS is only used to estimate LD between SNPs and therefore the fact that there are less individuals than SNPs should not be a problem? Or have I missed something?
Best, Kristi
|
|