Closed insilicool closed 1 year ago
I think you parsed the wrong file into the algorithm. You need to use this file: known_fusions.37.bedpe (not the csv file)
Apologizes. Indeed that was the issue.
Unfortunately, a new issue has arisen when applying the algorithm to the reference using the tumor junctions.
Command: module purge && module load mugqic/java/openjdk-jdk-19 && \ java -XX:+UseParallelGC -XX:ParallelGCThreads=1 -Dsamjdk.buffer_size=4194304 -Xmx36G \ -jar bin/sv-prep_v1.0.1.jar -threads 6 \ -sample COLO829N \ -bam_file COLO829N.sorted.dup.recal.bam \ -ref_genome Homo_sapiens.hs37d5.fa \ -ref_genome_version 37 \ -blacklist_bed hg37.blacklist.bed \ -known_fusion_bed known_fusions.37.bedpe \ -existing_junction_file SVariants/COLO829/GRIDSS/sv_prep/COLO829T.sv_prep.junctions.csv \ -output_dir SVariants/GRIDSS/COLO829/sv_prep
Error:
08:21:07.360 [main] [INFO ] SvPrep version: 1.0.1
08:21:07.370 [main] [INFO ] refGenome(V37), bam(alignment/COLO829N/COLO829N.sorted.dup.recal.bam)
08:21:07.371 [main] [INFO ] output(SVariants/GRIDSS/COLO829/sv_prep/)
08:21:07.380 [main] [INFO ] loaded 458 known hotspot records from file
08:21:07.381 [main] [INFO ] loaded 164 blacklist locations from file
08:21:07.387 [main] [ERROR] failed to create junction writer: java.io.FileNotFoundException: SVariants/GRIDSS/COLO829/sv_prep/COLO829N.sv_prep.junctions.csv (No such file or directory)
Exception in thread "main" htsjdk.samtools.util.RuntimeIOException: Error opening file: file:///lb/project/mugqic/projects/DREAM_techdev/SVariants/GRIDSS/COLO829/sv_prep/COLO829N.sv_prep.bam
at htsjdk.samtools.SAMFileWriterFactory.makeBAMWriter(SAMFileWriterFactory.java:306)
at htsjdk.samtools.SAMFileWriterFactory.makeBAMWriter(SAMFileWriterFactory.java:276)
at htsjdk.samtools.SAMFileWriterFactory.makeBAMWriter(SAMFileWriterFactory.java:252)
at com.hartwig.hmftools.svprep.BamWriter.initialise(BamWriter.java:44)
at com.hartwig.hmftools.svprep.BamWriter.
Best, Rob E.
Hi Rob,
SvPrep doesn't currently create new output directories, so it will need the one you specify to exist:
-output_dir SVariants/GRIDSS/COLO829/sv_prep
Can you try that?
thanks,
Charles
Hi,
I am testing sv_prep for potential addition to our workflow, but unfortunately, it is erroring out when parsing the known_fusion_data.37.csv from the latest resource bundle
Error: 15:40:46.242 [main] [INFO ] SvPrep version: 1.0.1 15:40:46.262 [main] [INFO ] refGenome(V37), bam(alignment/COLO829T/COLO829T.sorted.dup.recal.bam) 15:40:46.263 [main] [INFO ] output(SVariants/COLO829/GRIDSS/sv_prep/) 15:40:46.306 [main] [ERROR] invalid hotspot entry: Type,FiveGene,ThreeGene,CancerTypes,PubMedId,KnownExonTranscript,KnownExonUpRange,KnownExonDownRange,HighImpactPromiscuous,Overrides 15:40:46.359 [main] [INFO ] loaded 6402 blacklist locations from file
Command:
module purge && module load mugqic/java/openjdk-jdk-19 && \ mkdir -p SVariants/COLO829/GRIDSS/sv_prep && \ java -XX:+UseParallelGC -XX:ParallelGCThreads=1 -Dsamjdk.buffer_size=4194304 -Xmx36G \ -jar bin/sv-prep_v1.0.1.jar -threads 6 \ -sample COLO829T \ -bam_file alignment/COLO829T/COLO829T.sorted.dup.recal.bam \ -ref_genome Homo_sapiens.hs37d5.fa \ -ref_genome_version 37 \ -blacklist_bed hmf_pipeline_resources.37_v5.31/sv/sv_prep_blacklist.37.bed \ -known_fusion_bed hmf_pipeline_resources.37_v5.31/sv/known_fusion_data.37.csv \ -output_dir SVariants/COLO829/GRIDSS/sv_prep
Also, I've added -log_debug and -log_level DEBUG to the argument and it outputs the same uninformative error.
Any help would be much appreciated.
Thanks, Rob Eveleigh