I've run scalpel-discovery in single mode on 22 samples. The samples were run in parallel with identical commands; this is the discovery command I used:
One of those samples is failing repeatedly. The relevant part of the output looks like this:
START ANALYSIS
-- Print parameters to ./outdir/parameters.txt
Loading targets from BED file...12769 targets (filtered 0 overlapping).
Loading genome from FASTA file...7 sequences.
Assembly Exons
start assembly of 803632 regions.
stepSize: 803632 0 started, pid: 353
1. [0..803631] 0 just got out of the pool [PID: 353 | exit code: 0 | exit signal: 0 | core dump: 0]
Merge DBs...
sdbm: cannot insert after SPLTMAX attempts.
sdbm store returned -1, errno 22, key "d24d48574943ee6798bd1e4a41fadc56" at /local/bin/scalpel-0.5.3/MLDBM/Sync/SDBM_File.pm line 80.
Command failure: FindVariants (~/local/bin/scalpel-0.5.3/FindVariants.pl --bam ./sample.bam --bed GENES.bed --ref CHROMOSOMES.fa --kmer 25 --covthr 5 --lowcov 2 --mincov 5 --covratio 0.01 --outratio 0.05 --radius 100 --window 500 --maxregcov 10000 --step 100 --mapscore 1 --mismatches 3 --pathlimit 1000000 --dir /outdir --numprocs 1 --format vcf --coords null --logs)...
child exited with value 22
Comparing the output directory of the failed sample to output directories of samples that completed successfully, I see that the failed sample has files "variants.1.db.<ext>" and "variants.db.<ext>," with extensions .pag, .dir, and .lock, while the completed samples only have the "variants.db.<ext>" files. This makes sense in light of the apparent database merging error.
Does this reflect a bug in the program or something wrong with this particular sample? If the latter, is there a way I can resolve this to run scalpel-discovery on this sample?
Thanks.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've run scalpel-discovery in single mode on 22 samples. The samples were run in parallel with identical commands; this is the discovery command I used:
cd ${sample}
~/local/bin/scalpel-0.5.3/scalpel-discovery --single \ --bam ./${sample}.bam \ --bed ${bed} \ --ref ${ref} \ --window 500 \ --logs
--mapscore 20
One of those samples is failing repeatedly. The relevant part of the output looks like this:
START ANALYSIS
-- Print parameters to ./outdir/parameters.txt
Loading targets from BED file...12769 targets (filtered 0 overlapping).
Loading genome from FASTA file...7 sequences.
Assembly Exons
start assembly of 803632 regions.
stepSize: 803632
0 started, pid: 353
1. [0..803631] 0 just got out of the pool [PID: 353 | exit code: 0 | exit signal: 0 | core dump: 0]
Merge DBs...
sdbm: cannot insert after SPLTMAX attempts.
sdbm store returned -1, errno 22, key "d24d48574943ee6798bd1e4a41fadc56" at /local/bin/scalpel-0.5.3/MLDBM/Sync/SDBM_File.pm line 80.
Command failure: FindVariants (~/local/bin/scalpel-0.5.3/FindVariants.pl --bam ./sample.bam --bed GENES.bed --ref CHROMOSOMES.fa --kmer 25 --covthr 5 --lowcov 2 --mincov 5 --covratio 0.01 --outratio 0.05 --radius 100 --window 500 --maxregcov 10000 --step 100 --mapscore 1 --mismatches 3 --pathlimit 1000000 --dir /outdir --numprocs 1 --format vcf --coords null --logs)...
child exited with value 22
Comparing the output directory of the failed sample to output directories of samples that completed successfully, I see that the failed sample has files "variants.1.db.<ext>" and "variants.db.<ext>," with extensions .pag, .dir, and .lock, while the completed samples only have the "variants.db.<ext>" files. This makes sense in light of the apparent database merging error.
Does this reflect a bug in the program or something wrong with this particular sample? If the latter, is there a way I can resolve this to run scalpel-discovery on this sample?
Thanks.