search

cancerit / CaVEMan

SNV expectation maximisation based mutation calling algorithm aimed at detecting somatic mutations in paired (tumour/normal) cancer samples. Supports both bam and cram format via htslib
http://cancerit.github.io/CaVEMan
GNU Affero General Public License v3.0
60 stars 13 forks source link

Error during mstep for whole exome sequencing data #118

Open pawanchk opened 9 months ago

pawanchk commented 9 months ago

Hi,

I am using caveman.pl with Whole Exome sequencing (WXS) data however the execution ends in an error in the mstep process (please see the error below), this error occurs for various regions and then the execution stops.

I am using the singularity image created from the docker image to execute caveman.pl. For WGS data, the analysis completes successfully but for WXS data, it ends in this error.

Can I please know how to overcome this error ?

STDERR: + /opt/wtsi-cgp/bin/caveman mstep -i 49 -f ~/result/tmpCaveman/caveman.cfg.ini
STDERR: [ERROR] (src/bam_access.c: reads_at_pos_callback:843 errno: No such file or directory) Error, strand was already found for this readname HWI-ST118_106002510:3:23:14898:144717.
STDERR: [ERROR] (src/bam_access.c: bam_access_get_sorted_reads_at_this_pos:1240 errno: None) Error running callback
STDERR: [ERROR] (src/bam_access.c: bam_access_get_reads_at_this_pos:1276 errno: None) Error retrieving normal reads.
STDERR: [ERROR] (src/algos.c: algos_mstep_read_position:168 errno: None) Error retrieving read positions for section 10:24619409-24669409.
STDERR: caveman: src/List.c:60: List_clear_destroy: Assertion `list != NULL' failed.
STDERR: ~/result/tmpCaveman/logs/Sanger_CGP_Caveman_Implement_caveman_mstep.49.sh: line 3: 755750 Aborted                 (core dumped) /opt/wtsi-cgp/bin/caveman mstep -i 49 -f ~/result/tmpCaveman/caveman.cfg.ini
STDERR: Command exited with non-zero status 134
STDERR: 2.86user 0.25system 0:03.59elapsed 86%CPU (0avgtext+0avgdata 290328maxresident)k
STDERR: 0inputs+0outputs (0major+71668minor)pagefaults 0swaps
Thread 103 terminated abnormally: 
THREAD_EXITED: Wrapper script message:
"/usr/bin/time bash ~/result/tmpCaveman/logs/Sanger_CGP_Caveman_Implement_caveman_mstep.49.sh 1> ~/result/tmpCaveman/logs/Sanger_CGP_Caveman_Implement_caveman_mstep.49.out 2> ~/result/tmpCaveman/logs/Sanger_CGP_Caveman_Implement_caveman_mstep.49.err" unexpectedly returned exit value 134 at /opt/wtsi-cgp/lib/perl5/PCAP/Threaded.pm line 244 thread 103.
 at /opt/wtsi-cgp/lib/perl5/PCAP/Threaded.pm line 235
davidrajones commented 8 months ago

Hi there, I am no longer on this project, but it looks like there may be an issue with the data. STDERR: [ERROR] (src/bam_access.c: reads_at_pos_callback:843 errno: No such file or directory) Error, strand was already found for this readname HWI-ST118_106002510:3:23:14898:144717. implies that this read may have both ends mapped to the same orientation, which wouldn't be 'properly paired' by definition. Are you able to check sam flags for this read pair? Have you altered the acceptable flags required for CaVEMan to include a read in analysis at all?

If you can share the contents of ~/result/tmpCaveman/caveman.cfg.ini or your initial commandline for the setup step they may help.