# coding=utf-8
"""
Command line object for bbcontacts contact filtering application
"""
__author__ = "Felix Simkovic"
__date__ = "10 Aug 2016"
__version__ = 0.1
from Bio.Application import _Argument
from Bio.Application import _Option
from Bio.Application import _Switch
from Bio.Application import AbstractCommandline
[docs]class BbcontactsCommandLine(AbstractCommandline):
"""
Command line object for bbcontacts [#]_ contact filtering application
https://github.com/soedinglab/bbcontacts
The bbcontacts program is a Python program predicting residue-level
contacts between beta-strands by detecting patterns in matrices of
predicted couplings. bbcontacts can make use of a secondary structure
assignment or a secondary structure prediction.
.. [#] Andreani J., Söding J. (2015). bbcontacts: prediction of beta-strand
pairing from direct coupling patterns. Bioinformatics 31(11), 1729-1737.
Examples
--------
To filter a contact map using a Multiple Sequence Alignment in
CCMpred format, use:
>>> from conkit.applications import BbcontactsCommandLine
>>> bbcontacts_cline = BbcontactsCommandLine(
... matfile='test.mat', diversity_score=0.482, prefix='test'
... )
>>> print(bbcontacts_cline)
bbcontacts
You would typically run the command line with :func:`bbcontacts_cline` or via
the Python subprocess module.
Notes
-----
In order to use bbcontacts, head over to the `GitHub repository
<https://github.com/soedinglab/bbcontacts>`_, download the latest version
and install it using python setup.py install.
"""
def __init__(self, cmd="bbcontacts", **kwargs):
# TODO: figure a way to group CL arguments as in `mutually_exclusive_group`
if 'dssp_file' in list(kwargs.keys()) and 'psipred_file' in list(kwargs.keys()):
msg = 'Provide only one of [dssp_file|psipred_file]!'
raise RuntimeError(msg)
elif not ('dssp_file' in list(kwargs.keys()) or 'psipred_file' in list(kwargs.keys())):
msg = 'Provide one of [dssp_file|psipred_file]!'
raise RuntimeError(msg)
self.parameters = [
_Option(['-c', 'config_file'],
'bbcontacts configuration file',
filename=True,
equate=False),
_Option(['-s', 'smoothing_size'],
'Perform local background correction of the coupling matrix '
'before decoding: from each coupling, subtract the average '
'coupling (smoothed background) over an area extending by '
'SMOOTHINGSIZE in each direction [default=10, use 0 for no '
'local background correction]',
equate=False),
_Switch(['-l', 'long_predictions'],
'Turn off (slow) prediction-shortening mode (this mode is on '
'by default but will only get triggered when long predictions occur)'),
_Option(['-n', 'pdb_name'],
'Provide a PDB identifier (when also using -e, this will be the '
'PDB name to look for in EVALUATIONFILE)',
equate=False),
_Option(['-e', 'evaluation_file'],
'Provide a file containing the true contacts (BetaSheet916.dat, '
'BetaSheet1452.dat or same format) for evaluation',
filename=True,
equate=False),
_Argument(['matfile'],
'CCMpred-like coupling matrix',
filename=True,
is_required=True),
_Argument(['diversity_score'],
'sequence-dependent diversity score',
is_required=True),
_Argument(['prefix'],
'output prefix',
is_required=True),
_Option(['-d', 'dssp_file'],
'DSSP secondary structure prediction file',
filename=True,
equate=False),
_Option(['-p', 'psipred_file'],
'PSIPRED secondary structure prediction file',
filename=True,
equate=False),
]
AbstractCommandline.__init__(self, cmd, **kwargs)