arno
New Member
Posts: 4
|
Post by arno on Nov 5, 2013 10:32:00 GMT
Hi everybody,
Yesterday I started using GCTA to estimate heritability of a neurological disease. I'm using GCTA 1.02 on a MAC. Somehow the program doesn't recognize the command: gcta64 --grm-bin test --grm-cutoff 0.025 --make-grm-bin --out test_rm025. The following error occurs: Error: invalid option "--grm-bin".
Is the file test.grm.N.bin necessary for this command? As a result of the following command: gcta64 --bfile test --autosome --maf 0.01 --make-grm-bin- --out test, I only got two binary files: test.grm.bin and test.grm.id.
Arno
|
|
|
Post by chrchang on Nov 6, 2013 10:50:49 GMT
The .grm.bin format was introduced with GCTA 1.1, so GCTA 1.02 does not support it. Use the .grm.gz format instead.
|
|
arno
New Member
Posts: 4
|
Post by arno on Nov 7, 2013 10:42:23 GMT
Thank you very much for replying!
|
|
arno
New Member
Posts: 4
|
Post by arno on Nov 7, 2013 10:51:37 GMT
When I try to estimate the genetic variance explained by all autosomal SNPs (case-control) I get the following:
Options: --reml --grm test1 --pheno test.phen --mpheno 4 --prevalence 0.03 --qcovar test.qcovar --covar test.covar --out test
Reading IDs of the genetic relationship matrix (GRM) from [test1.grm.id]. 3171 IDs read from [test1.grm.id]. Reading the GRM from [test1.grm.gz]. Pairwise genetic relationships between 3171 individuals are included from [test1.grm.gz]. Reading phenotypes from [test.phen]. There are 4 traits specified in the file [test.phen]. The 4th trait is included for analysis. Nonmissing phenotypes of 3721 individuals are included from [test.phen]. Reading quantative covariates from [test.qcovar]. 2 quantative covarites of 3171 individuals read from [test.qcovar]. Reading discrete covariates from [test.covar].
Error: each row should have the same number of columns. {\colortbl;\red255\green255\blue255;}:\paperw11900\paperh16840\margl1440\margr1440\vieww10800\viewh8400\viewkind0
The test.covar file exists of 3 columns in each row: familial id, individual id and gender. Perhaps someone knows what the problem is? BTW, it works without the extra covariate gender (test.covar).
Arno
|
|
|
Post by Jian Yang on Nov 10, 2013 8:44:49 GMT
It should be simply format issue of the test.covar file.
|
|