search

jodyphelan / TBProfiler

Profiling tool for Mycobacterium tuberculosis to detect ressistance and strain type from WGS data
GNU General Public License v3.0
105 stars 43 forks source link

Analysis fails when using bam files as input #323

Open KWollenbergPhylo opened 10 months ago

KWollenbergPhylo commented 10 months ago

I am trying to run a TBProfiler 4.2.0 analysis on our Linux HPC cluster using trimmed, aligned, and deduplicated bam files as input (-a option). All of my runs have been failing at the delly analysis step. Here is what I am getting in the system error log:

Running command: set -u pipefail; delly call -t DEL -g /sysapps/cluster/software/Anaconda3/5.3.0/envs/tbprofiler-3.0.4-python3/share/tbprofiler/tbdb.fasta ./ExternalData/STaPHIdedup2/STaPHIdedup/ERR5863687_NC_000962.3sort.dedup2.bam -o ./ExternalData/STaPHItbpvar/288d2fa6-85a5-4071-a9ee-2a196d6f4f6a.delly.bcf Can't find ./ExternalData/STaPHItbpvar/288d2fa6-85a5-4071-a9ee-2a196d6f4f6a.delly.bcf

The directory is present and accessable. It appears that delly is failing but I do not get a TBProfiler error log for this. The runs complete if I use the --no_delly option, but then I don't get lineage calls, which I need. I am using bam files as input because about 1/3 of these samples have technical replicates which are merged to create the final bam file for each sample.

jodyphelan commented 7 months ago

Could you retry this using version >=6.2? Let me know if this is still an isue. The lineage information is not calculated with any of the delly results so this should not be an issue to run with --no_delly.