ngsLD unable to read genotype input file
Closed this issue · 4 comments
Hello,
I'm not sure this was the correct place to post this but I'm having trouble getting ngsLD to run properly and it appears to have a problem with the format of my genotype file. I have a genotype file with samples as columns and rows as sites. Genotypes are denoted by -1, 0, 1, and 2. It's a tab-delimited text file with no headers (I tried with headers as well).
When I run
ngsLD --n_threads 15 --geno snp.ngsld.geno.gz --posH pos.ngsld.txt --n_ind 134 --n_sites 14412 --out mylu.ld
I get the error message:
ERROR: [read_geno] cannot read GZip GENO file. Check GENO file and number of sites!
It also has a bunch of rows of Header found! skipping line....
I have n_sites as 14412 and wc -l snp.ngsld.geno 14412
When I unzip and run it I get the error message
==> BINARY input file (always lkl)
=====
ERROR: [main] invalid/corrupt genotype input file!
Is this something you could help me with?
Thanks,
-Bobby
Can you send me a small example file so I can reproduce the error? Maybe just a couple of SNPs and samples.
Sure, will do. I've attached the position and genotype files containing 10 sites and 6 samples.
sample.pos.txt
sample.snp.txt
Were those files created in windows?
If so, can you try running dos2unix on them and then compressing the geno file with gzip?
Hey, sorry for the late reply. It worked! Those files were altered in windows before I tried running ngsld.
Thank you!