Agreed. Sorry again.
I am having problem in indexing .fa file using bwa - any help would be greatly appreciated.
I am trying to index Homo_sapiens.GRCh37 ...fa file using the command
bwa index -p myGenome -a bwtsw /directory/myGenome.fa
but it keeps giving me the following error
[bwaindex] Pack FASTA... 56.76 sec [bwaindex] Reverse the packed sequence... Segmentation fault
Can someone please help me with possible suggestion to fix this. Thank you.
1 answer
I want to use this occasion to give some advice on how to write a good bug report not only in bioinformatics. To do that, try to take the point of view of the person to debug the software, they need a reproducible example, otherwise they might respond, "sorry works for me". There are various versions of software and OS. a good report (for a lokal tool) contains at least 4 things:
- The version of the software and how it was built (moist software has a --version switch)
- The OS and its version (give the output of
uname -a) - The exact command or what you did exactly, in particular this case required to post the input data.
- The exact error message
In addition, information about hardware and memory would be required, because memory is so obivously an issue here.
Example:
$ bwa
Program: bwa (alignment via Burrows-Wheeler transformation)
Version: 0.5.7 (r1310)
(downloaded from sourceforge)
$ uname -a
Darwin MyMac.local 10.8.0 Darwin Kernel Version 10.8.0: Tue Jun 7 16:33:36 PDT 2011; root:xnu-1504.15.3~1/RELEASE_I386 i386
bwa index -p myGenome -a bwtsw myGenome.fa (make the file myGenome.fa downloadable)
Some nasty error message ... core dumped
Oh, and my machine has only 1 GB of memory!
You only had points 3 and 4 partially.
don't worry, at least we got a template now. And, if out of memory, the program should say so and not segfault.
Log in to answer this question.
can you please give a little more info about your OS and the version of bwa you are using? It is amazing how few people know about how to make a proper bug-report these days.
also, why don't you write to the mailling list? http://sourceforge.net/mail/?group_id=276243
Thanks so much for your time in looking into post and replying. Actually I figured it out - I didn't know that to index human genome using bwa, need atleast 2gb of memory. So it was more hardware related issue. Thanks again.