Scripts for Regulatory Genome analysis.
Before running tests, sample data
of large size should be placed under
large_sample_data/.
hg38.fa: UCSC hg38 reference genome.
Test scripts are under tests/.
Use the following command to
run test for each scripts.
SCRIPT=#test script to run
python -m unittest ${SCRIPT}
For detailed input/output of each script,
please use the --help flag for each script.
Here are a few concepts the scripts center on.
A set of sequences that are independent from the reference genome. In our convention, exogeneous sequences are stored as a combination of a few files.
fasta: A fasta file stores the sequence information. The boundry of the sequence do not necessarily represents the boundry of exogeneous elements, as they are defined by the bed region.bed: A bed file stores the region information, aka the element boundry.npy: Zero or more npy file that store the necessary information, storing numpy array of data in the same order of the bed file.
As constricted by the numpy array data structure, informations are of the same length. For example, when storing data such as singal track (e.g. PROcap signals), elements will be padded to the same length.
Although we usually put one fasta entry for each element, elements are defined by the bed file and can be defined on the same fasta entry.
A special type of exogeneous sequences. The
fasta file for genomic regions are restricted to
a reference genome.
Another special type of exogeneous sequences and can
be processed by exogeneous_tool. This data type has an
additional constraint that each bed entry must correspond
to a fasta entry, so that results can be processed
using pattern matching with regular expressions.