Source: python-mhcflurry
Section: science
Maintainer: Steffen Moeller <moeller@debian.org>
Build-Depends:
 debhelper-compat (= 14),
 dh-sequence-python3,
 python3-all,
 python3-appdirs,
 python3-matplotlib,
 python3-mhcgnomes (>= 3.33.6),
 python3-numpy (>= 1.22.4),
 python3-packaging <!nocheck>,
 python3-pandas (>= 2.0),
 python3-pytest <!nocheck>,
 python3-scipy,
 python3-setuptools,
 python3-sklearn,
 python3-threadpoolctl,
 python3-torch (>= 2.0.0),
 python3-tqdm,
 python3-yaml,
Standards-Version: 4.7.4
Homepage: https://github.com/openvax/mhcflurry
Vcs-Browser: https://salsa.debian.org/med-team/python-mhcflurry
Vcs-Git: https://salsa.debian.org/med-team/python-mhcflurry.git
Testsuite: autopkgtest-pkg-pybuild

Package: python3-mhcflurry
Section: python
Architecture: all
Depends:
 ${python3:Depends},
 python3-appdirs,
 python3-matplotlib,
 python3-mhcgnomes (>= 3.33.6),
 python3-numpy (>= 1.22.4),
 python3-pandas (>= 2.0),
 python3-scipy,
 python3-sklearn,
 python3-threadpoolctl,
 python3-torch (>= 2.0.0),
 python3-tqdm,
 python3-yaml,
Description: MHC class I peptide presentation predictor
 MHCflurry predicts which peptides are likely to be displayed by major
 histocompatibility complex (MHC) class I molecules.  It provides neural
 network models for binding affinity, antigen processing, and presentation,
 as well as Python interfaces for training models and scanning protein
 sequences for candidate epitopes.
 .
 This package provides the Python 3 library.  The command-line tools are in
 the mhcflurry package.

Package: mhcflurry
Architecture: all
Depends:
 python3-mhcflurry (= ${binary:Version}),
Description: MHC class I peptide presentation prediction tools
 MHCflurry predicts which peptides are likely to be displayed by major
 histocompatibility complex (MHC) class I molecules.  It provides neural
 network models for binding affinity, antigen processing, and presentation,
 as well as tools for training models and scanning protein sequences for
 candidate epitopes.
 .
 This package provides the command-line tools.  Trained model weights are
 distributed separately and can be retrieved with the mhcflurry downloads
 command.