dogma-data 0.2.19

Data processing for Dogma

Dogma Data

Dogma Data is a Python package built for fast and efficient parsing of FASTA files, optimized for high-performance computing. It leverages multi-threading to fully utilize all available system threads, enabling parallel processing. Additionally, the package supports exporting parsed data to the HDF5 file format for easy storage and access.

Installation

To install Dogma Data, you can use pip:

pip install dogma-data

Usage

import dogma_data

vocab = {
    'a': 0,
    'g': 1,
    'c': 2,
    't': 3,
    ...
}

mapping = dogma_data.FastaMapping(vocab, vocab['a'])
(tokens, sequences, (taxons)) = dogma_data.parse_fasta('input_path.fa', dogma_data.HeaderType.TaxonId, mapping)

header_info = {"taxons": taxons}

dogma_data.export_hdf5(
    'output_path.h5',
    dogma_data.Splitter(
        train_prop=0.95,
        val_prop=0.025,
        test_prop=0.025,
        length=len(sequences) - 1,
    ),
    tokens,
    sequences,
    header_info,
    mapping
)

Requirements

• Python 3.10

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contact

For any questions, feel free to reach out:

• Marcel Rød - [e-mail]
• Miha Krajnc - [e-mail]