ImmunoTyper
A Immunoglobulin Heavy Chain Variable Gene genotyping and CNV analysis tool for WGS PacBio Long Reads using ILP Optimization.
Dependencies:
Running ImmunoTyper:
ImmunoTyper is run from the src/immunotyper.py
script via the command python immunotyper.py
.
```
usage: immunotyper.py [-h] [–dsf DSF] [–poa POA] [–spoa SPOA]
[–minimap MINIMAP] [–blasr BLASR]
[–max_copy MAX_COPY] [–debug_log_path DEBUG_LOG_PATH]
[–no_coverage_estimation] [–skip_extraction]
reads_path expected_coverage output_dir
ImmunoTyper: IGHV Genotyping using PacBio Long Reads
positional arguments:
reads_path Input FASTA file of PacBio reads
expected_coverage Sequencing depth
output_dir Path to output directory. The directory must exist. A
fasta is created for each allele call containing the
subreads assigned to that call
optional arguments:
-h, –help show this help message and exit
–dsf DSF Path to Dense Subgraph Finder tool (usually located in
ig_repertoire_constructor/dense_subgraph_finder.py)
–poa POA Path to POA tool (usually located at poaV2/poa).
Default location is ../tools/poaV2/poa.
–spoa SPOA Path to SPOA tool (usually located at
spoa/build/bin/spoa). Default location is
../tools/spoa/build/bin/spoa
–minimap MINIMAP Path to minimap tool
–blasr BLASR Path to blasr tool
–max_copy MAX_COPY Maximum number of allele copies to call
–debug_log_path DEBUG_LOG_PATH
Path to write log
–no_coverage_estimation
Disables empirical coverage
–skip_extraction Disables extractinbg subreads from input FASTA reads.
Used when subreads have already been isolated (i.e. in
simulated data)