SEQanswers

SEQanswers (http://seqanswers.com/forums/index.php)
-   Bioinformatics (http://seqanswers.com/forums/forumdisplay.php?f=18)
-   -   SOAP denovo seg faults during map (http://seqanswers.com/forums/showthread.php?t=13366)

ians 08-10-2011 12:52 PM

SOAP denovo seg faults during map
 
Hi all,
I'm getting a seg fault while running map during a SOAP denovo run. I'm running on OSX with SOAPdenovo v 1.05.
I have checked to be sure input data has reads larger than 30 bp and that no empty lines exist. Does anyone have any other debugging ideas?

Here is my cfg:

max_rd_len=70
[LIB]
rank=1
avg_ins=200
reverse_seq=0
asm_flags=3
q1=s_1_1.fastq
q2=s_1_2.fastq


Here's the map execution:

map -s soap.cfg -g test
K = 23
contig len cutoff: 25

there're 3424 contigs in file: test, max seq len 60625, min seq len 24, max name len 10
time spent on parse contigs file 0s
8 thread created
time spent on hash reads: 1s
4517379 nodes allocated, 5614229 kmer in reads, 5614229 kmer processed
time spent on De bruijn graph construction: 1s

time spent on mapping long reads: 0s

In file: soap.cfg, max seq len 70, max name len 256

8 thread created
5450 edges in graph
basicContigInfo: 2 vs 3
basicContigInfo: 3 vs 5
basicContigInfo: 5 vs 7
basicContigInfo: 7 vs 9
......
basicContigInfo: 4331 vs 5457
basicContigInfo: 4333 vs 5459
basicContigInfo: 4334 vs 5461
Segmentation fault

ians 08-11-2011 07:51 AM

I've been troubleshooting now by trying to divide and conquer. I've been halving the data set to see if i can get one to not seg fault. However, this strategy has really got me scratching my head. I can get the top 1st 1/4 and 2nd 1/4 to get past map. However, the top 1/2 fails. Also, when I concatenate the 1st and 2nd 1/4, it also fails (cmp shows this concat with the top 1/2 are identical).

Does anyone have ideas??


All times are GMT -8. The time now is 06:21 AM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2021, vBulletin Solutions, Inc.