Post by rs on May 27, 2014 18:04:06 GMT
Hi,
According to the GCTA website I can get BLUP estimates of SNPs --reml-pred-rand and --blup-snp. IF multiple GRMs are fitted in the initial --reml then the *.blp file will have two columns corresponding to each of the GRMs. Now, I am performing--reml with two two GRMs constructed from two non-overlapping sets of SNPs and want to estimate the BLUPs of the SNPs from this combined fit. So if I use the approach described on the website, I will still get two total genetic effect estimate for each SNP, corresponding to each GRM. However, I used each SNP to construct only one GRM. What might be the best way to go about this?
Thanks in advance
According to the GCTA website I can get BLUP estimates of SNPs --reml-pred-rand and --blup-snp. IF multiple GRMs are fitted in the initial --reml then the *.blp file will have two columns corresponding to each of the GRMs. Now, I am performing--reml with two two GRMs constructed from two non-overlapping sets of SNPs and want to estimate the BLUPs of the SNPs from this combined fit. So if I use the approach described on the website, I will still get two total genetic effect estimate for each SNP, corresponding to each GRM. However, I used each SNP to construct only one GRM. What might be the best way to go about this?
Thanks in advance
