tracy
New Member
Posts: 2
|
Post by tracy on Mar 24, 2019 16:44:39 GMT
Hello,
I am studying a non-model mouse species, and I am trying to produce a genomewide relatedness matrix using gcta64. I have a reference genome, but it is not yet annotated; therefore, the "chromosomes" are really scaffolds, and they number into the thousands. The error I receive when trying to produce the GRM is as follows:
Error: Line 1 of [file.bim] contains illegal chr number, please check
I tried to amend this by adding in an autosome-num option:
gcta64 --bfile rads --autosome --maf 0.01 --make-grm --out test --thread-num 10 --autosome-num 23873
But so far my jobs (I'm submitting this command via a supercomputer) have all timed out. I submitted my newest request for 24 hours, but I fear that the job will time out because the options I have specified are incorrect.
Any suggestions with this would be greatly appreciated!
Thanks,
Tracy
|
|
|
Post by diesel on May 9, 2019 15:39:14 GMT
Hi Tracy I am having the same issue and it is annoying.
Did you manage to sort this out?
I need something like the --allow-extra-chr of plink.
Also using the flag --chr followed by chromosome name does not work.
D.
|
|
ying
New Member
Posts: 2
|
Post by ying on May 10, 2019 16:31:13 GMT
I have the same issue. I don't have a reference genome and I use denovo to do variant calling. The "chromosomes" are actually contigs, I think.
I tried to set the --autosome-nun 2100000. It gives me Error: invalid number specified after the option --autosome-num.
Can anyone please give me some suggestions? Thanks in advance!
Ying
|
|
zack
New Member
Posts: 1
|
Post by zack on Sept 26, 2019 19:59:08 GMT
I am also having this problem. Did any of you figure this out?
|
|
ester
New Member
Posts: 1
|
Post by ester on Nov 30, 2020 13:44:18 GMT
Hello! I have solved using --autosome-num 29 (I work with bovine).
|
|
|
Post by mbonar on Aug 8, 2022 20:35:24 GMT
Was this ever resolved? I am also having this problem
|
|