Segmentation fault (core dumped) from Integrate fusion

[Hoon Kim]

Dear Zhang,

Running Integrate fusion to my data keeps causing "Segmentation fault (core dumped)" at the step of "Printing results" as shown below.
Because of this error, I have 'empty' breakpoints.tsv' and bk_sv.vcf; however, 'bk_fusion.bedpe' is not empty, having results.
Of course, I have 'summary.tsv', 'reads.txt' and 'exons.tsv' all of which are valid (Not empty).
I wonder if you know why I have such a segmentation fault error.
Is it ok to assume that this job was successful by considering that results have been written in 'bk_fusion.bedpe' ?

Thank you in advance,

Hoon

---------------------------------------------------------------Log Message--------------------------------------
Processing DNA Tumor...

Warning: for RG 130712_SN1120_0268_AC2E3DACXX_8, BAM says: 0 and tested is 206. Changed to tested value.
Insert sizes and stds tested by different RGs:
130712_SN1120_0268_AC2E3DACXX_8 206 109
28.5201 seconds

Processing DNA Normal...

Warning: for RG 130712_SN1120_0268_AC2E3DACXX_6, BAM says: 0 and tested is 194. Changed to tested value.
Insert sizes and stds tested by different RGs:
130712_SN1120_0268_AC2E3DACXX_6 194 98
27.9099 seconds

Printing results...
here 1
/var/spool/torque/mom_priv/jobs/1442332.cnode341.mdanderson.edu.SC: line 50: 2861 Segmentation fault (core dumped) Integrate fusion ./Homo_sapiens_assembly19.fasta ./annot.ucsc.txt ./bwts ./accepted_hits.bam ./unmapped.bam ./tumor.sorted.recalibed.bam ./normal.sorted.recalibed.bam
ERROR: Integrate Tue Dec 29 01:24:47 CST 2015
---------------------------------------------------------------Log Message--------------------------------------

[Jin]

It seems that something from the WGS alignments finally caused the issue. This is the first time of this issue. I will have to use your data to find the reason. But in the meanwhile, for this case, all the results about gene fusions have been printed out, and you can BLAT the spanning WGS reads in reads.txt if you need genomic breakpoints.

[Hoon Kim]

Thank you very much for your quick response and suggestion.

[Hoon Kim]

[Hoon Kim]

Dear Zhang,
As a test run, I did BLAT (default parameter) with spanning WGS reads whose breakpoints are provided in 'breakpoints.tsv'; however, my BLAT did not provide the same breakpoints as in 'breakpoints.tsv'.
I wonder if Integrate uses different parameter values when running BLAT. If it does, would you please let me know those parameter sets so that I can get the same breakpoints?

Thank you very much,

Hoon