This module calculate the read coverage for each loci, given nucleotide reads and loci of interest,
pip install git+https://www.github.com/zhoulytwinyu/coverage-py
You can import functions to calculate coveraged:
from coverage import get_coverages_bisect
get_coverages_bisect
A command line script is included in the package. Once the package is installed using pip, you can call:
get_coverages.py <READS_CSV> <LOCI_CSV> <OUTPUT_CSV>
where
READS_CSV
is a csv containing "start" and "length"LOCI_CSV
is a csv containing "position"OUTPUT_CSV
is the output, a csv containing "position" and "coverage"
You can verify the module works correctly by running unittest we provided. In command line:
git clone https://www.github.com/zhoulytwinyu/coverage-py.git
cd coverage-py
python3 setup.py test
The unittest randomly generate reads and loci, in silico and then crosscheck the results from two methods: get_coverages_bisect
and get_coverages_loop
.
The idea is that random, simulated test may cover more corner cases and the two methods provide the same result only when both of them works correctly.