An implementation of the leapdna toolkit in Python.
leapdna is compatible with Python 3.9 and greater. Install it from PyPI with
pip install leapdna
leapdna aims to provide a way to represent a lot of the data that is normally dealt with in genetics, with a focus on forensic applications. The usual suspects here are loci, alleles, profiles, allele frequency studies... All of these are represented in leapdna by a block. A block is a Python class which can be stored as a JSON file with the leapdna format. All blocks inherit from the Base block class and define some attributes and functionality of their own. For instance, a Locus block has a name attribute, in addition to a band attribute, which specifies the chromosome band where the locus is. Other Locus blocks will have different values for these attributes while other types of blocks, such as Allele, will have a different set of attributes altogether. The documentation details which attributes are supported (and sometimes required) for a particular block type.
from leapdna.blocks import Allele, Locus
l1 = Locus('L1')
a1 = Allele(name='a1', locus=l1)leapdna aims to be extensible, so it allows you to specify your own attributes in any kind of block. If you wanted to mark a locus as a potential carrier of a diseas you could do
l1.user['carrier'] = TrueTo run the leapdna test suite execute
python3 -m unittestfrom the root of the leapdna repository. Test coverage can be calculated with the coverage Python package. To do so, first run the tests with coverage run --source leapdna -m unittest discover and then get the report with coverage report -m.
In addition, leapdna comes with type annotations which allows for running it through a type checker such as mypy. This can be done by executing mypy leapdna from the root of the repository. If you do not have mypy installed you may do so with python3 -m pip install mypy.
Copyright 2021 Elias Hernandis
Leapdna is free and open-source sofware released under the MIT License. Everyone is free to copy, modify and redistribute the sofware, even for commercial purposes. While no attribution is required, it is very much appreciated. The full text of the license is available here.