Hisat2-align failing about hisat2 HOT 6 CLOSED

Hi Crag,

Would you like to be more specific about the genome you used? It might be possible index building is incomplete. You may want to check if you have all 8 HISAT2 index files by running "ls -l genomeindex*"?

Best,
Daehwan

from hisat2.

Hi Daehwan,

The genome is the Chinese Hamster (criGri1, hosted at ucsc), a genome made up of ~50,000 scaffolds.

All 8 index files are present in the directory specified, however they are not shown when I use” ls –l criGri1*”

The ls –l command only returns the bowtie indices used by tophat which are in the same directory. Is there an issue with having these in the same location?

Thanks,
Craig

From: Daehwan Kim [mailto:[email protected]]
Sent: 28 September 2015 21:23
To: infphilo/hisat2
Cc: Craig Monger
Subject: [Disarmed] Re: [hisat2] Hisat2-align failing (#1)

Hi Crag,

Would you like to be more specific about the genome you used? It might be possible index building is incomplete. You may want to check if you have all 8 HISAT2 index files by running "ls -l genomeindex*"?

Best,
Daehwan

—
Reply to this email directly or view it on GitHubhttps://github.com/infphilo/issues/1#issuecomment-143863755.

from hisat2.

I build hisat from source instead of using the binary and reindexed the genome in a directory on its own and it is now working correctly. Not sure what the problem was before though!

from hisat2.

Hi Craig,

HISAT2 index files have a different extension, "ht2", instead of Bowtie2 indexes, "bt2." It doesn't seem likely your own binary is a solution to this issue. Perhaps, HISAT2 didn't know where to look for its index files if you didn't define a shell variable such as HISAT2_INDEXES.

Thanks,
Daehwan

from hisat2.

I think I have a related issue. My hisat2 align just fails. It runs for whatever time I provide but the sam file is size 0. No errors are produced. I have been error testing by removing unnecessary parts of the command and I have this basic command.

/programs/hisat2-2.0.4/hisat2 \
-p 8 \
-x "hisat2index-genome/hs2drg" \
-1 $1 \
-2 $2 \
-S "out.sam"

The only thing that can be wrong here is the index files. The index was created like so:

 /programs/hisat2-2.0.4/hisat2-build \
-p 8 \
--ss "hisat2-splice-sites.txt" \
--exon "hisat2-exons.txt" \
"dna.fa" \
"hisat2index-ensembl-genome/hs2drg"`

The -ss and --exon files were prepared using the provided python scripts from a gtf file. The gtf file has been used for a lot of other things and it seems to be ok. The index folder has 7 files:

hs2drg.1.ht2 hs2drg.2.ht2 hs2drg.3.ht2 hs2drg.4.ht2 hs2drg.7.ht2 hs2drg.8.ht2 hs2drg.rf

Is it supposed to have 8? The indexing run is given 1 hour and it completes in 15 mins. So it's not a time-out issue.

I tried build without the -ss and --exon file and it works fine. 8 index files are produced in 15 mins. The align works too.

from hisat2.

In my case, hisat2 showed such error when I used repeatmasked reference genome.

from hisat2.