|
|
Post by Robin on Oct 31, 2014 13:10:47 GMT
When running a conditional analysis I'm getting some of the p values coming out as NA. Also some of the variants in the input file don't seem to be in the output files. What's the reason for this?
Thanks
|
|
|
|
Post by Jian Yang on Nov 4, 2014 3:00:44 GMT
Let C = the set of SNPs fixed in the model
Let S = a SNP to be tested
Re 1) If the genotype of C is a linear function of S (i.e. co-linearity issue), then a "NA" is reported.
Re 2) SNPs exist in the summary file but not in the PLINK genotype data are not reported.
|
|
|
|
Post by Robin on Nov 4, 2014 10:51:52 GMT
Hi Jian, thanks for the reply.
Re 2) I've had a look in the PLINK genotype data I used and the summary file and both files seem to contain the SNPs which are not being reported in the output file. Do you know of any other reason for them to not be reported?
If it makes a difference I'm using gcta version 1.24
|
|
|
|
Post by Jian Yang on Nov 7, 2014 15:24:36 GMT
SNPs with alleles that are not matched in the two sets of data will also be excluded.
|
|
