Question: Metavelvet stops at Scaffolding
gravatar for biotacg
4.5 years ago by
biotacg0 wrote:

Why does Metavelvet stops at "Scaffolding" without error message or at "Starting pebble resolution..." with a Segmentation fault?

I'm using MetaVelvet to analyse ~3million PE Illumina MiSeq reads with an insert length of ~400. The reads were checked for quality and trimmed before assembly. Metavelvet stops at the scaffolding stage.

The commands used to run Metavelvet were:

velveth output 77 -fastq -shortPaired -separate inputR1.fastq inputR2.fastq
velvetg output -exp_cov auto -ins_length 400
meta-velvetg output -ins_length 400 | tee logfile

The script stops either at:

[VelvetGraph]  === Scaffolding ===

or at

Starting pebble resolution...
Segmentation fault (core dumped)

Do you know why it stops?

I also tried an -ins_length 300, but without success.

This question has been asked before , but not useful answer has been given so far, so with this post I hope to reopen the discussion and to find a solution to the problem.

I'm running on a 16 CPUs 32GB RAM server and I've installed Metavelvet as described in the following Dockerfile:

FROM fedora:23

RUN dnf -y update && dnf -y install wget tar make zlib-devel gcc-c++

RUN wget
RUN tar zxf velvet_1.2.10.tgz
# install velvet allow for long sequences and multiple threads
RUN cd velvet_1.2.10 && make MAXKMERLENGTH=127 LONGSEQUENCES=1 OPENMP=1
RUN cp velvet_1.2.10/velvet* /usr/bin

RUN wget
RUN tar zxf MetaVelvet-1.2.02.tgz 
RUN cd MetaVelvet-1.2.02 && make MAXKMERLENGTH=127
RUN cp MetaVelvet-1.2.02/meta-velvetg /usr/bin

I also tried installing velvet without the LONGSEQUENCES=1 OPENMP=1 options i.e:make MAXKMERLENGTH=127, but this didn't change anything.

How can this problem be solved?

ADD COMMENTlink modified 22 months ago by RamRS28k • written 4.5 years ago by biotacg0
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 679 users visited in the last hour