Look for potential processed pseudogenes by applying filters to deletion SVs called by DELLY, then counting the number of events per gene.
Some code adapted from Ronak Shah's check_cDNA_contamination.py.
usage: python processed_pseudogenes.py [-h] [-s sample_name] [-t n]
[--iAnnotateSV iAnnotateSV.py]
[--genome {hg18,hg19,hg38}]
structural_variants.vcf
/path/to/output_directory/
positional arguments:
structural_variants.vcf
VCF with deletion event calls
/path/to/output_directory/
Directory to put output in
optional arguments:
-h, --help show this help message and exit
-s sample_name, --sample-name sample_name
Prefix [default: filename of in_vcf]
-t n, --threshold n Output pseudogenes must have at least this many
qualifying deletion events. [default: 2]
--iAnnotateSV iAnnotateSV.py
path to iAnnotateSV.py script [default: /home/shahr2/g
it/iAnnotateSV/iAnnotateSV/iAnnotateSV.py]
--genome {hg18,hg19,hg38}
Reference genome version to use during annotation
[default: hg19]
$ python processed_pseudogenes.py example.vcf ./ -s my_sample
<output directory>/<sample name>.pseudogenes.txt: a one-line, tab-delimited list of sample name and its processed pseudogenes- Intermediate files in
<output directory>/scratch/
- Python 2
- pandas
- gridmap
- iAnnotateSV and its required modules
- iCallSV
- PyVCF
- Filter the input VCF for PRECISE deletions with paired read count > 10
- Convert the filtered VCF to tab-delimited input format for iAnnotateSV, using iCallSV.dellyVcf2Tab
- Annotate using iAnnotateSV
- For each gene, count the number of events that aren't in exons, "in frame" or "out of frame".If a gene has more than the threshold number of events, then conclude it's a processed pseudogene.
Wrapper to run processed_psuedogene_analysis.py
python run_processed-psuedogene_analysis.py -h
usage: run_processed-psuedogene_analysis.py [options]
Run processed-psuedogene_analysis on selected pools/samples using MSK data
optional arguments:
-h, --help show this help message and exit
-mif folders.fof, --metaInformationFile folders.fof
Full path folders and sample names of files.
-qc /some/path/qcLocation, --qcLocation /some/path/qcLocation
Full path qc files.
-P /somepath/python, --python /somepath/python
Full path Pyhton executables.
-ppg /somepath/process-pusedogene.py, --processpsuedogene /somepath/process-pusedogene.py
Full path processpsuedogene.py executables.
-ias /somepath/iAnnotateSV.py, --iAnnotateSV /somepath/iAnnotateSV.py
Full path iAnnotate.py executables.
-q all.q or clin.q, --queue all.q or clin.q
Name of the SGE queue
-qsub /somepath/qsub, --qsubPath /somepath/qsub
Full Path to the qsub executables of SGE.
-t 5, --threads 5 Number of Threads to be used to run process-psuedogene
analysis on
-v, --verbose make lots of noise [default]
-o /somepath/output, --outDir /somepath/output
Full Path to the output dir.
$ python run_processed-psuedogene_analysis.py -mif /location/to/meatdata.txt -qc /dmp/qc/location/path -ppg /path/to/processed-pseudogenes/processed_pseudogenes.py -ias /path/to/iAnnotateSV.py -q sge_queue_name -qsub path/to/qsub -t 1 -o /path/to/output/directory -v
Tab separated file with header:
| Run | MRN | Mnumber | LIMS_ID | DMP_ASSAY_ID | 12245 |
|---|---|---|---|---|---|
| PoolName | SomeID | SomeID | SomeID | SomeID | Flag(0,1) |
From here: The Run and LIMS_ID are required.