Bwa mem is exiting with segmentation fault for large contigs (> 0.4Mbp)

[ramya1]

Hello all,

I am using bwa mem to align contigs to a reference genome and I observed that it is exiting with an error message "Segmentation Fault". For example:

[M::main_mem] read 1 sequences (405327 bp)...
Segmentation fault

Upon closer examination, I found that whenever the size of input sequence >=0.4 Mbp, I see this message. In bwa mem page (http://bio-bwa.sourceforge.net/bwa.shtml), it is mentioned that it can handle contigs upto 1Mbp. I am unable to figure out the source of this upper limit at 0.4Mbp. Is it specific to my computer or is there any parameter by which I can assign more memory to bwa?
I am currently thinking of breaking contigs greater than 0.4Mbp into overlapping smaller contigs of size 399999. Is there any other way to go about it.

Thanks all,
Ramya

[dpryan]

What happens to memory usage during this? Another cause might be just that bwa is trying to use more memory than you have.

[lindenb]

cross posted: https://www.biostars.org/p/98534/

[ramya1]

@dpryan I checked the memory usage during the few seconds before it exits with the error.
The stats before running the process were:
Cpu(s): 0.7%us, 0.1%sy, 0.0%ni, 99.1%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 16446968k total, 4512156k used, 11934812k free, 138536k buffers
Swap: 31249404k total, 1169612k used, 30079792k free, 1910296k cached

The stats while running the process were:
Cpu(s): 10.9%us, 0.3%sy, 0.0%ni, 88.7%id, 0.1%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 16446968k total, 4606148k used, 11840820k free, 138592k buffers
Swap: 31249404k total, 1169588k used, 30079816k free, 1910316k cached

As you can see, there is not much of a difference in terms of memory usage.

@lindenb I am sorry but was I not supposed to cross post. I just tried to get help from any quarter I could. If the two forums are equivalent (wrt the people I can reach out to), in future I shall restrict myself to one.

I ran bwa-sw and it ran without giving me any errors, so this solves my problem for the time being. I am aligning contigs which are expected to be very similar to the reference sequence wise, differing mostly in chromosomal rearrangements. Thus, I expect both, bwa-sw and bwa-mem to do equally well, accuracy wise. Still I will run both for the same dataset to compare the two.

Thanks,
Ramya

[ramya1]

Dear all,

For completion purposes, I wish to report the solution to my problem. As it turns out, I was using bwa-0.7.0 version which is known to have bugs. Thus, on the sugestion of lh3-Alt (https://www.biostars.org/p/98534/#98886), I used the latest version i.e. bwa-0.7.8 and I did not get any segmentation faults.

Thanks all for the help and time :-)
Ramya