HH-suite: Segmentation fault (core dumped)
Hi all,
Everytime I try to run HHblits from HH-suite (https://github.com/soedinglab/hh-suite/wiki#hh-suite-databases), it runs to a certain point then fails with this message: Segmentation fault (core dumped). I noticed this usually occurs right after the HMM-HMM Viterbi alignment starts. I thought it might be a memory issue so I increased the memory and # of cpus, however I still get the same message. Any ideas?
hhblits -cpu 4 -M first -i myMSA.fasta -o result.hhr -d ./databases/uniclust30_2018_08
12427 Segmentation fault (core dumped)
Thank you
0 answers
No answers yet.
Log in to answer this question.
More posts like this
-
Full length sequences from hhblits
written by ejl •
-
Segmentation faults when running some PoPoolation2 scripts
written by will.shim •
-
Error when running hhblits: Cannot allocate memory/Could not read index file...
written by nkuhuangyan •
-
PSI-BLAST core dumped
written by sepehr.salsabili •
-
STAR alignment - segmentation fault error
written by mgmohsen •
-
Erorr running HHblits
written by max_19
-
fastq_to_fasta segmentation fault (core dumped)
written by Mitra •
-
STAR aligner - Segmentation fault (core dumped) multiple fastq paired-end samples
written by deep.pandya •
-
Metavelvet segmentation fault
written by makisthegreat •
-
Trowel Segmentation fault (core dumped)
written by isaac.akogwu •
Is it a single fasta?
How was HHsuite installed?
The fasta is made up of 35 multiple sequence alignments that I have trimmed and concatenated. (Muscle x 35 then trimAl then FASconCAT) I am using a pre installed version on a compute cluster.
Do you have to request memory allocations on your cluster or is it an open pool?
That’s quite a lot of data, so it sounds like you may just be running out of RAM.
Thanks for your help. I have to request memory allocations. I just ran it on a much smaller subset of my fasta file though, and the error persists.
The issue may not be number of sequences, but rather the length of the sequence. How long is the alignment?
I’m not sure it makes a huge amount of sense to build HMMs from concatenated alignments.
I see. Each alignment is ~11,000 characters long. maybe there is a better way to go about this. I have 35 orthologous proteins (across 30 similar species) and i'd like to build an HMM profile with them and then compare them to HMM profiles in HH suite databases (profile vs profile), so I can determine how conserved they are with other species.
Since one of the input formats of hhblits is .fasta, I figured I could concatenate the 35 MSAs, but perhaps this is not the best strategy?
You don’t need to concatenate them. Just use all 35 MSAs individually. Since it’s looking for domains etc via HMM matching, artificially stitching them all together would likely only confuse matters anyway.
It would probably be better just to filter the results files after the fact
I will try that instead! thanks for your help!