search

DaehwanKimLab / hisat2

Graph-based alignment (Hierarchical Graph FM index)
GNU General Public License v3.0
464 stars 113 forks source link

Having troubles building index with variation #254

Open jonassibbesen opened 4 years ago

jonassibbesen commented 4 years ago

Hi,

I am trying to build an index with variation from the 1000 genomes project. I first tried to build an index using all variation, but I got the following error:

exceeded integer bounds, remove adjacent SNPs, use haplotypes, or switch to a large index (--large-index).

I am already using haplotypes so I decided to filter low frequent alleles. I tried removing all alleles with a frequency below 0.01 and 0.05 resulting in ~14M and ~8.5M variants, respectively. However, I still got the same error. Would it help if I increased --inter-gap and --intra-gap in hisat2_extract_snps_haplotypes_VCF.py? I was thinking that longer haplotypes might help to not make the number of nodes explode.

Next I decided to try to switch to a large index as suggested using --large-index on the filtered variants. However, that resulted in a crash with the following error:

Command terminated by signal 6

I am only adding variants and not splice-junctions. Hope you are able to help me solve this.

Thanks,

Jonas

a00101 commented 3 years ago

I got same problem exactly.

a00101 commented 3 years ago

@parkchanhee I've been waiting for Developers. please help us.

a00101 commented 3 years ago

@jonassibbesen did you solve this problem? I can't wait any more.

jonassibbesen commented 3 years ago

@a00101 No, not yet.

shanwai1234 commented 3 years ago

@jonassibbesen @a00101 I also see this issue. I remember it works for me previously. But I don't know why it happens now.