#39 Problems with Mira-V5rc1

[Arthur Weyna]

Hi Mr Chevreux,

This ticket is actually about the mira assembler V5rc1 which is currently on your github page, but I could not find the milestone for it while creating this ticket. Please tell me if this belongs somewhere else. I am trying to use mirabait in the course of my work, in order to extract specific sequences from paired fastq files.

I am encountering a strange (to me at least) problem with the available static binary: it works wonders when used on my laptop, running on ubuntu 18, but the same exact binary fails when used on another distant machine with the exact same command and test files.

./mirabait -b plop.fasta -p plop2.fasta plop3.fasta (plop files are minimum test files containing one small sequence in fasta format)
~~~

Baiting sequences with at least 1 exact kmer matches.
Unknown exception caught, aborting the process.

Please contact: bach@chevreux.org

Baiting process finished.

Total number of sequences read: 0
Pairs baited: 0 (-nan%)
Pairs missed: 0 (-nan%)
Unpaired baited: 0 (-nan%)
Unpaired missed: 0 (-nan%)

~~~

The kernel on the machine is :
Linux popphyl-sge 3.2.0-86-generic #124-Ubuntu SMP Wed Jun 17 21:40:14 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
(I am sure you will need more information on this system, please just ask)

Is it possible that despite the static status of the binary, some program could be missing?

I also tried to compile the source code from scratch. I am a beginner at best when it comes to compiling C code so it would not surprise me at all if I did anything wrong. I followed the documentation you made available and compiled expat, boost, zlib and boost from source before I compiled mira (see linked mira_install file). The configure file does not complain about the system (see linked configure_log file) but static and dynamic compilations fail anyway (see linked make_log file).

I am creating this ticket as a last resort hoping you could help me finding and resolving the problem.
I am not aware of any other software that could allow me to extract sequences from paired fastq files and any implementation of what I need that I could come up with would be soooo slow.

I hope I am not being to much of an annoyance, and I stay at your service for any supplementary info you need.

Best regards,

Arthur Weyna

[Bastien Chevreux]

I suspect the runtime error on the distant machine somehow really being by some sort of incopatibility between the compiled binary and available system libraries. But I have no idea what causes it.

Regarding compilation error: that was a bug fixed in git, but I did not make a release before. There's now rc2 as release on GitHub

Best,
Bastien

[Arthur Weyna]

The new release works as expected!
Thank you a million times for taking the time to help (and being so quick at it too).

Best regards,

Arthur