Maq: Segmentation Fault

mrxcm3 · 11-08-2010, 04:16 AM

Hi

I am trying to map my SOLiD reads in Maq. I have converted my reads from .csfasta and .qual to .fastq using 'solid2fastq.pl'. I then generated a binary file using 'maq fastq2bfq'. I turned my reference genome (GRCh37) from .fa to .csfa using 'maq fasta2csfa' and made this binary using 'fasta2bfa'.

When I use 'maq map' I get a segmentation fault. There is no other information. Here is my command;

maq map -c aln.cs.map ref.csbfa reads.bfq 2> aln.log

Does anyone have any ideas what could be causing this? I am currently using v0.7.1, but have tried the same command on earlier versions and get the same. I am wondering is the 'maq fasta2csfa' accepts reference in '.fa'?

TenchBB · 03-21-2011, 12:48 PM

I have recently experienced a similar problem. I am also attempting to use MAQ to map my SOLiD reads.

I had recieved the .fastq files from a collaborator so I first used "maq fastq2bfq" and "maq fasta2csfa" and then ran the following the command:

maq map -c out.cs.map ref.csbfa reads_1.bfq reads_2.bfq 2

I got back the following error:

-- maq-0.7.1
[ma_load_reads] loading reads...
[ma_load_reads] set length of the first read as 1.
[ma_load_reads] set length of the second read as 1.
maq: read.cc:61: longreads_t* ma_load_reads(void*, int, void*, int): Assertion `strncmp(name, lr->name[j], tl-1) == 0' failed.

I then tried to go back a step and use the .csfasta and _QV.qual files to create the .fastq files using solid2fastq.pl. This ran with no error, howvever, the output files are empty!

I am stuck. Can anyone offer any advice??

Sincere thanks,

Tench

nilshomer · 03-21-2011, 03:43 PM

@TenchBB: Please do not double-post in two different forums (or here for that matter). Thank-you!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Maq Segmentation fault	Seq84	Bioinformatics	3	07-13-2012 05:05 AM
soap segmentation fault	scami	Bioinformatics	6	04-17-2012 07:08 AM
cufflinks segmentation fault help please	seq_newbie	Bioinformatics	1	06-29-2011 12:52 PM
SOAP:segmentation fault	Mansequencer	Bioinformatics	1	11-19-2010 11:47 AM
MAQ - Segmentation fault	dukevn	Bioinformatics	1	04-30-2009 01:07 PM

11-08-2010, 04:16 AM	#1
mrxcm3 Junior Member Location: Nottingham Join Date: Oct 2010 Posts: 9	Maq: Segmentation Fault Hi I am trying to map my SOLiD reads in Maq. I have converted my reads from .csfasta and .qual to .fastq using 'solid2fastq.pl'. I then generated a binary file using 'maq fastq2bfq'. I turned my reference genome (GRCh37) from .fa to .csfa using 'maq fasta2csfa' and made this binary using 'fasta2bfa'. When I use 'maq map' I get a segmentation fault. There is no other information. Here is my command; maq map -c aln.cs.map ref.csbfa reads.bfq 2> aln.log Does anyone have any ideas what could be causing this? I am currently using v0.7.1, but have tried the same command on earlier versions and get the same. I am wondering is the 'maq fasta2csfa' accepts reference in '.fa'?

03-21-2011, 12:48 PM	#2
TenchBB Junior Member Location: Washington DC Join Date: Mar 2011 Posts: 1	I have recently experienced a similar problem. I am also attempting to use MAQ to map my SOLiD reads. I had recieved the .fastq files from a collaborator so I first used "maq fastq2bfq" and "maq fasta2csfa" and then ran the following the command: maq map -c out.cs.map ref.csbfa reads_1.bfq reads_2.bfq 2 I got back the following error: -- maq-0.7.1 [ma_load_reads] loading reads... [ma_load_reads] set length of the first read as 1. [ma_load_reads] set length of the second read as 1. maq: read.cc:61: longreads_t* ma_load_reads(void, int, void, int): Assertion `strncmp(name, lr->name[j], tl-1) == 0' failed. I then tried to go back a step and use the .csfasta and _QV.qual files to create the .fastq files using solid2fastq.pl. This ran with no error, howvever, the output files are empty! I am stuck. Can anyone offer any advice?? Sincere thanks, Tench

03-21-2011, 03:43 PM	#3
nilshomer Nils Homer Location: Boston, MA, USA Join Date: Nov 2008 Posts: 1,285	@TenchBB: Please do not double-post in two different forums (or here for that matter). Thank-you!