Entering edit mode
10.3 years ago
Pavel Senin
★
1.9k
I use bwa mem to align paired reads to few hundreds of microbial contigs; then I sort the alignment, and trying to get a coverage using bedtools genomecov -ibam alignments.paired.sorted.bam -bg >ranges.txt, which fails with an exception:
*** glibc detected *** bedtools: double free or corruption (out): 0x0000000001c5f270 ***
======= Backtrace: =========
/lib64/libc.so.6[0x3d7b2750c6]
bedtools[0x45ab43]
bedtools[0x45b146]
bedtools[0x45c163]
bedtools[0x45e2ed]
bedtools[0x434c4b]
/lib64/libc.so.6(__libc_start_main+0xfd)[0x3d7b21ecdd]
if I run the same using not paired alignment, everything is ok. So I am really not sure where is my mistake... maybe bedtools doesn't digest the paired alignment?
-- edit: works with the latest versions of these tools. Here are the ones that failed:
$ bwa
Program: bwa (alignment via Burrows-Wheeler transformation)
Version: 0.7.0-r313
Contact: Heng Li <lh3@sanger.ac.uk>
$ bedtools -version
bedtools v2.16.1
Can you post a snippet of the alignments that produces the crash? BTW, I could imagine bwa mem's chimeric alignments producing a crash.
i really can not see which one does that. the alignment file is too large to figure out which particular alignment fails, bedtools doesn't indicate that too... sorry.
There's always recompiling and then running it in
gdb, though that's not the most enjoyable experience :(downloading and running latest versions of tools fixed the problem. must be bugs.
Cool, I put a "[resolved]" in your post title so people don't see the lack of an answer and assume someone should help out.