The script fails at the SNV callers step with the following error:
Run SNV callers module for PBTA data
Setting up Database
python3: can't open file 'analyses/snv-callers/scripts/01-setup_db.py': [Errno 2] No such file or directory
I believe the source of the error is that there is nothing in the shell script analyses/snv-callers/run_caller_consensus_analysis-pbta.sh that causes it to have a consistent working directory. So, when it gets called from scripts/generate-analysis-files-for-release.sh, its working directory is scripts/ which does not contain analyses/snv-callers/scripts/01-setup_db.py.
It would be consistent with other aspects of the project to set the shell script's working directory to where the file is located in the project, e.g., with:
Then we'd change all of the paths to the scripts in analyses/snv-callers/run_caller_consensus_analysis-pbta.sh to be relative to analyses/snv-callers. If I recall correctly, all the scripts in the module already assume the paths to files (e.g., MAF files) supplied as arguments are relative to the root of the repository, so we won't have to do anything there.
Note: I haven't look at the TCGA shell script in snv-callers, but I'm willing to bet we'd have to do the same thing.
@runjin326 is running the
generate-analysis-files-for-release.sh
script described here: https://github.com/AlexsLemonade/OpenPBTA-analysis/issues/1399#issuecomment-1124329628The script fails at the SNV callers step with the following error:
Which stems from this line: https://github.com/jaclyn-taroni/OpenPBTA-analysis/blob/f95853e50c92fa850a156f00cd914808e88cf098/scripts/generate-analysis-files-for-release.sh#L63
And subsequently:
https://github.com/AlexsLemonade/OpenPBTA-analysis/blob/c9b0f4456066f821f2c67983bdd822ed209d8b7a/analyses/snv-callers/run_caller_consensus_analysis-pbta.sh#L30
I believe the source of the error is that there is nothing in the shell script
analyses/snv-callers/run_caller_consensus_analysis-pbta.sh
that causes it to have a consistent working directory. So, when it gets called fromscripts/generate-analysis-files-for-release.sh
, its working directory isscripts/
which does not containanalyses/snv-callers/scripts/01-setup_db.py
.It would be consistent with other aspects of the project to set the shell script's working directory to where the file is located in the project, e.g., with:
https://github.com/AlexsLemonade/OpenPBTA-analysis/blob/c9b0f4456066f821f2c67983bdd822ed209d8b7a/analyses/interaction-plots/01-create-interaction-plots.sh#L14
Then we'd change all of the paths to the scripts in
analyses/snv-callers/run_caller_consensus_analysis-pbta.sh
to be relative toanalyses/snv-callers
. If I recall correctly, all the scripts in the module already assume the paths to files (e.g., MAF files) supplied as arguments are relative to the root of the repository, so we won't have to do anything there.Note: I haven't look at the TCGA shell script in
snv-callers
, but I'm willing to bet we'd have to do the same thing.