|
Post by uilson on Jan 20, 2016 0:30:47 GMT
Dear,
I am a new user of GCTA and I am trying to analyze some cacao (the chocolate tree) data, on Linux (Ubuntu 64 bits/Virtualbox). Since I have many SNPs (11 thousand), first I generated 2 files marcos.tfam and marcos.tped, as described in PLINK manual (item 3.4. p 37) Then: 1. I generated the files marcos.bed, marcos.fam and marcos.bim, using PLINK on MS-DOS/Windows (not in Linux, because I had to compile PLINK for Linux and I am not familiar with that) 2. In linux I generated the GRM matrix using "./gcta64 –bfile marcos –autosome-num 10 –make-grm-gz –out marcos" . Note cacao has n=10 chromosomes. This generated the files marcos.grm.gz and marcos.grm.id
3. When I tried to do a reml analysis using these data with the command "./gcta --reml --grm marcos --pheno marcos.phen", it produced the message Reading IDs of the GRM from [marcos.grm.id] 295 IDs read from [marcos.grm.id]. Error: can not open the file [marcos.grm.bin] to read
What may be causing that message?
When I open the txt file marcos.grm.id, it looks fine. See part of that file below.
Thank you very much
1 1 2 6 3 7 ... 41 SG-2007 41 SG-2013 41 SG-2017 ... 130 SG-2909 131 SG-2916 131 SG-2924
|
|
|
Post by Jian Yang on Jan 20, 2016 0:44:05 GMT
You could try either
./gcta64 –bfile marcos –autosome-num 10 –make-grm-gz –out marcos ./gcta --reml --grm-gz marcos --pheno marcos.phen
or
./gcta64 –bfile marcos –autosome-num 10 –make-grm –out marcos ./gcta --reml --grm marcos --pheno marcos.phen
|
|
|
Post by uilson on Jan 20, 2016 1:02:12 GMT
Thanks. Both options worked.
|
|