def execDSSP(pdb, outputname=None, outputdir=None, stderr=True):
"""Execute DSSP for given *pdb*. *pdb* can be a PDB identifier or a PDB
file path. If *pdb* is a compressed file, it will be decompressed using
Python :mod:`gzip` library. When no *outputname* is given, output name
will be :file:`pdb.dssp`. :file:`.dssp` extension will be appended
automatically to *outputname*. If :file:`outputdir` is given, DSSP
output and uncompressed PDB file will be written into this folder.
Upon successful execution of :command:`dssp pdb > out` command, output
filename is returned. On Linux platforms, when *stderr* is false,
standard error messages are suppressed, i.e.
``dssp pdb > outputname 2> /dev/null``.
For more information on DSSP see http://swift.cmbi.ru.nl/gv/dssp/.
If you benefited from DSSP, please consider citing [WK83]_.
.. [WK83] Kabsch W, Sander C. Dictionary of protein secondary structure:
pattern recognition of hydrogen-bonded and geometrical features.
*Biopolymers* **1983** 22:2577-2637."""
dssp = which('mkdssp')
if dssp is None:
dssp = which('dssp')
if dssp is None:
raise EnvironmentError('command not found: dssp executable is not '
'found in one of system paths')
assert outputname is None or isinstance(outputname, str),\
'outputname must be a string'
assert outputdir is None or isinstance(outputdir, str),\
'outputdir must be a string'
if not os.path.isfile(pdb):
pdb = fetchPDB(pdb, compressed=False)
if pdb is None:
raise ValueError('pdb is not a valid PDB identifier or filename')
if os.path.splitext(pdb)[1] == '.gz':
if outputdir is None:
pdb = gunzip(pdb, os.path.splitext(pdb)[0])
else:
pdb = gunzip(pdb, os.path.join(outputdir,
os.path.split(os.path.splitext(pdb)[0])[1]))
if outputdir is None:
outputdir = '.'
if outputname is None:
out = os.path.join(outputdir,
os.path.splitext(os.path.split(pdb)[1])[0] +
'.dssp')
else:
out = os.path.join(outputdir, outputname + '.dssp')
if not stderr and PLATFORM != 'Windows':
status = os.system('{0} {1} > {2} 2> /dev/null'.format(
dssp, pdb, out))
else:
status = os.system('{0} {1} > {2}'.format(dssp, pdb, out))
if status == 0:
return out
def showSequenceTree(hits):
"""Returns a plot that contains a dendrogram of the sequence similarities among
the sequences in given hit list.
:arg hits: A dictionary that contains hits that are obtained from a blast record object.
:type hits: dict
"""
clustalw = which('clustalw')
if clustalw is None:
print(
"The executable for clustalw does not exists, install or add clustalw to path."
)
return
try:
from Bio import Phylo
except:
raise ImportError("Phylo is not installed properly.")
with open("hits.fasta", "w") as inp:
for z in hits:
inp.write(">" + str(z) + "\n")
inp.write(hits[z]['hseq'])
inp.write("\n")
cmd = clustalw + " hits.fasta"
os.system(cmd)
tree = Phylo.read("hits.dnd", "newick")
try:
import pylab
except:
raise ImportError("Pylab or matplotlib is not installed.")
Phylo.draw(tree)
return
def pathVMD(*path):
"""Returns VMD path, or set it to be a user specified *path*."""
if not path:
path = SETTINGS.get('vmd', None)
if isExecutable(path):
return path
else:
LOGGER.warning('VMD path is not set by user, looking for it.')
vmdbin = None
vmddir = None
if PLATFORM == 'Windows':
if PY3K:
import winreg
else:
import _winreg as winreg # PY3K: OK
for vmdversion in ('1.8.7', '1.9', '1.9.1'):
try:
key = winreg.OpenKey(
winreg.HKEY_LOCAL_MACHINE,
'Software\\University of Illinois\\VMD\\' +
vmdversion)
vmddir = winreg.QueryValueEx(key, 'VMDDIR')[0]
vmdbin = join(vmddir, 'vmd.exe')
except:
pass
try:
key = winreg.OpenKey(
winreg.HKEY_LOCAL_MACHINE,
'Software\\WOW6432node\\University of Illinois\\VMD\\'
+ vmdversion)
vmddir = winreg.QueryValueEx(key, 'VMDDIR')[0]
vmdbin = join(vmddir, 'vmd.exe')
except:
pass
else:
vmdbin = which('vmd')
if False:
pipe = os.popen('which vmd')
vmdbin = pipe.next().strip()
vmdfile = open(vmdbin)
for line in vmdfile:
if line.startswith('defaultvmddir='):
vmddir = line.split('=')[1].replace('"', '')
break
vmdfile.close()
if isExecutable(vmdbin):
setVMDpath(vmdbin)
return vmdbin
elif len(path) == 1:
path = path[0]
if isExecutable(path):
SETTINGS['vmd'] = path
SETTINGS.save()
LOGGER.info("VMD path is set to '{0}'.".format(path))
else:
raise OSError('{0} is not executable.'.format(str(path)))
else:
raise ValueError('specify a single path string')
def pathVMD(*path):
"""Return VMD path, or set it to be a user specified *path*."""
if not path:
path = SETTINGS.get('vmd', None)
if isExecutable(path):
return path
else:
LOGGER.warning('VMD path is not set by user, looking for it.')
vmdbin = None
vmddir = None
if PLATFORM == 'Windows':
if PY3K:
import winreg
else:
import _winreg as winreg # PY3K: OK
for vmdversion in ('1.8.7', '1.9', '1.9.1'):
try:
key = winreg.OpenKey(winreg.HKEY_LOCAL_MACHINE,
'Software\\University of Illinois\\VMD\\' +
vmdversion)
vmddir = winreg.QueryValueEx(key, 'VMDDIR')[0]
vmdbin = join(vmddir, 'vmd.exe')
except:
pass
try:
key = winreg.OpenKey(winreg.HKEY_LOCAL_MACHINE,
'Software\\WOW6432node\\University of Illinois\\VMD\\' +
vmdversion)
vmddir = winreg.QueryValueEx(key, 'VMDDIR')[0]
vmdbin = join(vmddir, 'vmd.exe')
except:
pass
else:
vmdbin = which('vmd')
if False:
pipe = os.popen('which vmd')
vmdbin = pipe.next().strip()
vmdfile = open(vmdbin)
for line in vmdfile:
if line.startswith('defaultvmddir='):
vmddir = line.split('=')[1].replace('"', '')
break
vmdfile.close()
if isExecutable(vmdbin):
setVMDpath(vmdbin)
return vmdbin
elif len(path) == 1:
path = path[0]
if isExecutable(path):
SETTINGS['vmd'] = path
SETTINGS.save()
LOGGER.info("VMD path is set to '{0}'.".format(path))
else:
raise OSError('{0} is not executable.'.format(str(path)))
else:
raise ValueError('specify a single path string')
def getVMDpath():
"""Return VMD path set by user or one identified automatically."""
path = SETTINGS.get("vmd", None)
if isExecutable(path):
return path
else:
LOGGER.warning("VMD path is not set by user, looking for it.")
from types import StringType, UnicodeType
vmdbin = None
vmddir = None
if PLATFORM == "Windows":
import _winreg
for vmdversion in ("1.8.7", "1.9", "1.9.1"):
try:
key = _winreg.OpenKey(
_winreg.HKEY_LOCAL_MACHINE, "Software\\University of Illinois\\VMD\\" + vmdversion
)
vmddir = _winreg.QueryValueEx(key, "VMDDIR")[0]
vmdbin = os.path.join(vmddir, "vmd.exe")
except:
pass
try:
key = _winreg.OpenKey(
_winreg.HKEY_LOCAL_MACHINE, "Software\\WOW6432node\\University of Illinois\\VMD\\" + vmdversion
)
vmddir = _winreg.QueryValueEx(key, "VMDDIR")[0]
vmdbin = os.path.join(vmddir, "vmd.exe")
except:
pass
else:
vmdbin = which("vmd")
if False:
pipe = os.popen("which vmd")
vmdbin = pipe.next().strip()
vmdfile = open(vmdbin)
for line in vmdfile:
if line.startswith("defaultvmddir="):
vmddir = line.split("=")[1].replace('"', "")
break
vmdfile.close()
if (
False
and isinstance(vmdbin, (StringType, UnicodeType))
and isinstance(vmddir, (StringType, UnicodeType))
and os.path.isfile(vmdbin)
and os.path.isdir(vmddir)
):
pass # return vmdbin, vmddir
if isExecutable(vmdbin):
setVMDpath(vmdbin)
return vmdbin
def getVMDpath():
"""Return VMD path set by user or one identified automatically."""
path = SETTINGS.get('vmd', None)
if isExecutable(path):
return path
else:
LOGGER.warning('VMD path is not set by user, looking for it.')
from types import StringType, UnicodeType
vmdbin = None
vmddir = None
if PLATFORM == 'Windows':
import _winreg
for vmdversion in ('1.8.7', '1.9', '1.9.1'):
try:
key = _winreg.OpenKey(_winreg.HKEY_LOCAL_MACHINE,
'Software\\University of Illinois\\VMD\\' +
vmdversion)
vmddir = _winreg.QueryValueEx(key, 'VMDDIR')[0]
vmdbin = os.path.join(vmddir, 'vmd.exe')
except:
pass
try:
key = _winreg.OpenKey(_winreg.HKEY_LOCAL_MACHINE,
'Software\\WOW6432node\\University of Illinois\\VMD\\' +
vmdversion)
vmddir = _winreg.QueryValueEx(key, 'VMDDIR')[0]
vmdbin = os.path.join(vmddir, 'vmd.exe')
except:
pass
else:
vmdbin = which('vmd')
if False:
pipe = os.popen('which vmd')
vmdbin = pipe.next().strip()
vmdfile = open(vmdbin)
for line in vmdfile:
if line.startswith('defaultvmddir='):
vmddir = line.split('=')[1].replace('"', '')
break
vmdfile.close()
if False and \
isinstance(vmdbin, (StringType, UnicodeType)) and \
isinstance(vmddir, (StringType, UnicodeType)) and \
os.path.isfile(vmdbin) and os.path.isdir(vmddir):
pass#return vmdbin, vmddir
if isExecutable(vmdbin):
setVMDpath(vmdbin)
return vmdbin
def execSTRIDE(pdb, outputname=None, outputdir=None):
"""Execute STRIDE program for given *pdb*. *pdb* can be an identifier or
a PDB file path. If *pdb* is a compressed file, it will be decompressed
using Python :mod:`gzip` library. When no *outputname* is given, output
name will be :file:`pdb.stride`. :file:`.stride` extension will be
appended automatically to *outputname*. If :file:`outputdir` is given,
STRIDE output and uncompressed PDB file will be written into this folder.
Upon successful execution of :command:`stride pdb > out` command, output
filename is returned.
For more information on STRIDE see http://webclu.bio.wzw.tum.de/stride/.
If you benefited from STRIDE, please consider citing [DF95]_.
.. [DF95] Frishman D, Argos P. Knowledge-Based Protein Secondary Structure
Assignment. *Proteins* **1995** 23:566-579."""
stride = which('stride')
if stride is None:
raise EnvironmentError('command not found: stride executable is not '
'found in one of system paths')
assert outputname is None or isinstance(outputname, str),\
'outputname must be a string'
assert outputdir is None or isinstance(outputdir, str),\
'outputdir must be a string'
if not os.path.isfile(pdb):
pdb = fetchPDB(pdb, compressed=False)
if pdb is None:
raise ValueError('pdb is not a valid PDB identifier or filename')
if os.path.splitext(pdb)[1] == '.gz':
if outputdir is None:
pdb = gunzip(pdb, os.path.splitext(pdb)[0])
else:
pdb = gunzip(
pdb,
os.path.join(outputdir,
os.path.split(os.path.splitext(pdb)[0])[1]))
if outputdir is None:
outputdir = '.'
if outputname is None:
out = os.path.join(
outputdir,
os.path.splitext(os.path.split(pdb)[1])[0] + '.stride')
else:
out = os.path.join(outputdir, outputname + '.stride')
status = os.system('{0} {1} > {2}'.format(stride, pdb, out))
if status == 0:
return out
def execSTRIDE(pdb, outputname=None, outputdir=None):
"""Execute STRIDE program for given *pdb*. *pdb* can be an identifier or
a PDB file path. If *pdb* is a compressed file, it will be decompressed
using Python :mod:`gzip` library. When no *outputname* is given, output
name will be :file:`pdb.stride`. :file:`.stride` extension will be
appended automatically to *outputname*. If :file:`outputdir` is given,
STRIDE output and uncompressed PDB file will be written into this folder.
Upon successful execution of :command:`stride pdb > out` command, output
filename is returned.
For more information on STRIDE see http://webclu.bio.wzw.tum.de/stride/.
If you benefited from STRIDE, please consider citing [DF95]_.
.. [DF95] Frishman D, Argos P. Knowledge-Based Protein Secondary Structure
Assignment. *Proteins* **1995** 23:566-579."""
stride = which('stride')
if stride is None:
raise EnvironmentError('command not found: stride executable is not '
'found in one of system paths')
assert outputname is None or isinstance(outputname, str),\
'outputname must be a string'
assert outputdir is None or isinstance(outputdir, str),\
'outputdir must be a string'
if not os.path.isfile(pdb):
pdb = fetchPDB(pdb, compressed=False)
if pdb is None:
raise ValueError('pdb is not a valid PDB identifier or filename')
if os.path.splitext(pdb)[1] == '.gz':
if outputdir is None:
pdb = gunzip(pdb, os.path.splitext(pdb)[0])
else:
pdb = gunzip(pdb, os.path.join(outputdir,
os.path.split(os.path.splitext(pdb)[0])[1]))
if outputdir is None:
outputdir = '.'
if outputname is None:
out = os.path.join(outputdir,
os.path.splitext(os.path.split(pdb)[1])[0] +
'.stride')
else:
out = os.path.join(outputdir, outputname + '.stride')
status = os.system('{0} {1} > {2}'.format(stride, pdb, out))
if status == 0:
return out
def conf_opt_setup(name):
namd2 = which('namd2')
par = os.path.join('/usr/local/lib/vmd/plugins/noarch/tcl/readcharmmpar1.2', 'par_all27_prot_lipid_na.inp')
dir_name = name[0:4] + '_opt'
if os.path.exists(dir_name):
shutil.rmtree(dir_name)
os.makedirs(dir_name)
conf = open('min.conf').read()
for pdb in glob.glob(os.path.join(name[0:4] + '_ens', '*.pdb')):
fn = os.path.splitext(os.path.split(pdb)[1])[0]
pdb = os.path.join('..', pdb)
out = open(os.path.join(dir_name, fn + '.conf'), 'w')
out.write(conf.format(
out=fn, pdb=pdb, par=par))
out.close()
def conf_opt_setup(name):
namd2 = which('namd2')
par = os.path.join('/usr/local/lib/vmd/plugins/noarch/tcl/readcharmmpar1.2', 'par_all27_prot_lipid_na.inp')
dir_name = name[0:4] + '_opt'
if os.path.exists(dir_name):
shutil.rmtree(dir_name)
os.makedirs(dir_name)
conf = open('min.conf').read()
for pdb in glob.glob(os.path.join(name[0:4] + '_ens', '*.pdb')):
fn = os.path.splitext(os.path.split(pdb)[1])[0]
pdb = os.path.join('..', pdb)
out = open(os.path.join(dir_name, fn + '.conf'), 'w')
out.write(conf.format(
out=fn, pdb=pdb, par=par))
out.close()
class TestDSSPFunctions(unittest.TestCase):
@dec.slow
def setUp(self):
"""Setup the testing framework."""
self.pdbs = [DATA_FILES['dssp']]
@dec.slow
@unittest.skipIf(which('dssp') is None, 'dssp is not found')
def testDSSPBridgePartners(self):
"""Check if the DSSP bridge-partners were correctly parsed and
assigned."""
for pdb in self.pdbs:
prot_ag = parseDatafile(pdb['file'], folder=TEMPDIR)
dssp = execDSSP(pathDatafile(pdb['file']),
outputdir=TEMPDIR,
stderr=False)
parseDSSP(dssp, prot_ag, parseall=True)
# Map a dssp_resnum to its Residue object.
dssp_dict = {}
for chain in prot_ag.select("protein").getHierView():
for res in chain:
dssp_resnum = res.getData("dssp_resnum")[0]
dssp_dict[dssp_resnum] = res
for res in dssp_dict.values():
bp1 = res.getData("dssp_bp1")[0]
bp2 = res.getData("dssp_bp2")[0]
if bp1 != 0:
msg_ = "BP1 (dssp_resnum: %d) of %s is missing" % \
(bp1, str(res))
self.assertIn(bp1, dssp_dict, msg=msg_)
if bp2 != 0:
msg_ = "BP2 (dssp_resnum: %d) of %s is missing" % \
(bp2, str(res))
self.assertIn(bp2, dssp_dict, msg=msg_)
def buildMSA(sequences, title='Unknown', labels=None, **kwargs):
"""
Aligns sequences with clustalw or clustalw2 and returns the resulting MSA.
:arg sequences: a file, MSA object or a list or array containing sequences
as Atomic objects with :func:`getSequence` or Sequence objects or strings.
If strings are used then labels must be provided using ``labels``
:type sequences: :class:`Atomic`, :class:`.MSA`,
:class:`~numpy.ndarray`, str
:arg title: the title for the MSA and it will be used as the prefix for output files.
:type title: str
:arg labels: a list of labels to go with the sequences
:type labels: list
:arg align: whether to align the sequences
default True
:type align: bool
:arg method: alignment method, one of either biopython.align.globalms or clustalw(2).
default 'clustalw'
:type align: str
"""
align = kwargs.get('align', True)
method = kwargs.pop('method', 'clustalw')
# 1. check if sequences are in a fasta file and if not make one
if isinstance(sequences, str):
filename = sequences
elif not isinstance(sequences, MSA):
try:
max_len = 0
for sequence in sequences:
if isinstance(sequence, Atomic):
if len(sequence.ca.copy()) > max_len:
max_len = len(sequence.ca.copy())
elif isinstance(sequence, MSA):
if len(sequence[0]) > max_len:
max_len = len(sequence[0])
else:
if len(sequence) > max_len:
max_len = len(sequence)
msa = []
fetched_labels = []
for i, sequence in enumerate(sequences):
if isinstance(sequence, Atomic):
strseq = sequence.ca.getSequence()
label = sequence.getTitle()
elif isinstance(sequence, Sequence):
strseq = str(sequence)
label = sequence.getLabel()
elif isinstance(sequence, MSA):
strseq = str(sequence[0])
label = sequence.getLabel(0)
LOGGER.warn(
'Only the first sequence in the MSA at entry {0} is used.'
.format(i))
elif isinstance(sequence, str):
strseq = sequence
label = str(i + 1)
else:
raise TypeError('sequences should be a list of strings, '
'Atomic, or Sequence instances')
strseq = strseq + '-' * (max_len - len(strseq))
msa.append(array(list(strseq)))
fetched_labels.append(label)
sequences = array(msa)
except:
raise TypeError('sequences should be iterable')
# "if a list" is a pythonic way to check if a list is empty or not (or none)
if not labels and fetched_labels:
labels = fetched_labels
label = [label.replace(' ', '_') for label in labels]
# labels checkers are removed because they will be properly handled in MSA class initialization
msa = MSA(msa=sequences, title=title, labels=labels)
if align and 'clustal' in method:
filename = writeMSA(title + '.fasta', msa)
if align:
# 2. find and run alignment method
if 'biopython' in method:
if len(sequences) == 2:
msa, _, _ = alignTwoSequencesWithBiopython(
sequences[0], sequences[1], **kwargs)
else:
raise ValueError(
"Provide only two sequences or another method. \
Biopython pairwise alignment can only be used \
to build an MSA with two sequences.")
elif 'clustalw' in method:
clustalw = which('clustalw')
if clustalw is None:
if which('clustalw2') is not None:
clustalw = which('clustalw2')
else:
raise EnvironmentError(
"The executable for clustalw was not found, \
install clustalw or add it to the path."
)
os.system('"%s" %s -OUTORDER=INPUT' % (clustalw, filename))
# 3. parse and return the new MSA
msa = parseMSA(title + '.aln')
else:
alignTool = which(method)
if alignTool is None:
raise EnvironmentError("The executable for {0} was not found, \
install it or add it to the path.".
format(alignTool))
os.system('"%s" %s -OUTORDER=INPUT' % (clustalw, filename))
# 3. parse and return the new MSA
msa = parseMSA(title + '.aln')
return msa
def buildMSA(sequences, title='Unknown', labels=None, **kwargs):
"""
Aligns sequences with clustalw or clustalw2 and returns the resulting MSA.
:arg sequences: a file, MSA object or a list or array containing sequences
as Atomic objects with :func:`getSequence` or Sequence objects or strings.
If strings are used then labels must be provided using ``labels``
:type sequences: :class:`Atomic`, :class:`.MSA`,
:class:`~numpy.ndarray`, str
:arg title: the title for the MSA and it will be used as the prefix for output files.
:type title: str
:arg labels: a list of labels to go with the sequences
:type labels: list
:arg align: whether to align the sequences
default True
:type align: bool
:arg method: alignment method, one of either biopython.align.globalms or clustalw(2).
default 'clustalw'
:type align: str
"""
align = kwargs.get('align', True)
method = kwargs.pop('method', 'clustalw')
# 1. check if sequences are in a fasta file and if not make one
if isinstance(sequences, str):
filename = sequences
elif not isinstance(sequences, MSA):
try:
max_len = 0
for sequence in sequences:
if isinstance(sequence, Atomic):
if len(sequence.ca.copy()) > max_len:
max_len = len(sequence.ca.copy())
elif isinstance(sequence, MSA):
if len(sequence[0]) > max_len:
max_len = len(sequence[0])
else:
if len(sequence) > max_len:
max_len = len(sequence)
msa = []
fetched_labels = []
for i, sequence in enumerate(sequences):
if isinstance(sequence, Atomic):
strseq = sequence.ca.getSequence()
label = sequence.getTitle()
elif isinstance(sequence, Sequence):
strseq = str(sequence)
label = sequence.getLabel()
elif isinstance(sequence, MSA):
strseq = str(sequence[0])
label = sequence.getLabel(0)
LOGGER.warn('Only the first sequence in the MSA at entry {0} is used.'
.format(i))
elif isinstance(sequence, str):
strseq = sequence
label = str(i + 1)
else:
raise TypeError('sequences should be a list of strings, '
'Atomic, or Sequence instances')
strseq = strseq + '-'*(max_len - len(strseq))
msa.append(array(list(strseq)))
fetched_labels.append(label)
sequences = array(msa)
except:
raise TypeError('sequences should be iterable')
# "if a list" is a pythonic way to check if a list is empty or not (or none)
if not labels and fetched_labels:
labels = fetched_labels
label = [label.replace(' ','_') for label in labels]
# labels checkers are removed because they will be properly handled in MSA class initialization
msa = MSA(msa=sequences, title=title, labels=labels)
if align and 'clustal' in method:
filename = writeMSA(title + '.fasta', msa)
if align:
# 2. find and run alignment method
if 'biopython' in method:
if len(sequences) == 2:
msa, _, _ = alignTwoSequencesWithBiopython(sequences[0], sequences[1], **kwargs)
else:
raise ValueError("Provide only two sequences or another method. \
Biopython pairwise alignment can only be used \
to build an MSA with two sequences.")
elif 'clustalw' in method:
clustalw = which('clustalw')
if clustalw is None:
if which('clustalw2') is not None:
clustalw = which('clustalw2')
else:
raise EnvironmentError("The executable for clustalw was not found, \
install clustalw or add it to the path.")
os.system('"%s" %s -OUTORDER=INPUT'%(clustalw, filename))
# 3. parse and return the new MSA
msa = parseMSA(title + '.aln')
else:
alignTool = which(method)
if alignTool is None:
raise EnvironmentError("The executable for {0} was not found, \
install it or add it to the path.".format(alignTool))
os.system('"%s" %s -OUTORDER=INPUT'%(clustalw, filename))
# 3. parse and return the new MSA
msa = parseMSA(title + '.aln')
return msa
def buildMSA(sequences, title='Unknown', labels=None, **kwargs):
"""
Aligns sequences with clustalw or clustalw2 and returns the resulting MSA.
:arg sequences: a file, MSA object or a list or array containing sequences
as Atomic objects with :func:`getSequence` or Sequence objects or strings.
If strings are used then labels must be provided using ``labels``
:type sequences: :class:`Atomic`, :class:`.MSA`,
:class:`~numpy.ndarray`, str
:arg title: the title for the MSA and it will be used as the prefix for output files.
:type title: str
:arg labels: a list of labels to go with the sequences
:type labels: list
:arg align: whether to do alignment with clustalw(2)
default True
:type align: bool
"""
align = kwargs.get('align', True)
# 1. check if sequences are in a fasta file and if not make one
if isinstance(sequences, str):
filename = sequences
elif not isinstance(sequences, MSA):
try:
max_len = 0
for sequence in sequences:
if len(sequence) > max_len:
max_len = len(sequence)
msa = []
fetched_labels = []
for i, sequence in enumerate(sequences):
if isinstance(sequence, Atomic):
strseq = sequence.getSequence()
label = sequence.getTitle()
elif isinstance(sequence, Sequence):
strseq = str(sequence)
label = sequence.getLabel()
elif isinstance(sequence, str):
strseq = sequence
label = str(i + 1)
else:
raise TypeError('sequences should be a list of strings, '
'Atomic, or Sequence instances')
strseq = strseq + '-' * (max_len - len(strseq))
msa.append(array(list(strseq)))
fetched_labels.append(label)
sequences = array(msa)
except:
raise TypeError('sequences should be iterable')
# "if a list" is a pythonic way to check if a list is empty or not (or none)
if not labels and fetched_labels:
labels = fetched_labels
# labels checkers are removed because they will be properly handled in MSA class initialization
msa = MSA(msa=sequences, title=title, labels=labels)
if align:
filename = writeMSA(title + '.fasta', msa)
if align:
# 2. find and run alignment method
clustalw = which('clustalw')
if clustalw is None:
if which('clustalw2') is not None:
clustalw = which('clustalw2')
else:
raise EnvironmentError(
"The executable for clustalw was not found, \
install clustalw or add it to the path."
)
os.system('"%s" %s' % (clustalw, filename))
# 3. parse and return the new MSA
msa = parseMSA(title + '.aln')
return msa
from os.path import sep as dirsep
import inspect
import tempfile
try:
import unittest2 as unittest
from unittest2 import TestCase, skipIf, skipUnless
except ImportError:
import unittest
from unittest import TestCase, skipIf, skipUnless
from prody.utilities import PLATFORM
from prody import LOGGER
from prody.utilities import which
NOPRODYCMD = which('prody') is None
WINDOWS = PLATFORM == 'Windows'
try:
import matplotlib
matplotlib.use('Agg')
except ImportError:
MATPLOTLIB = False
else:
try:
from matplotlib import pyplot
except ImportError:
MATPLOTLIB = False
else:
MATPLOTLIB = True
def scanPockets(self):
'Generates ESSA z-scores for pockets and parses pocket features. It requires both Fpocket 3.0 and Pandas being installed in your system.'
from re import findall
fpocket = which('fpocket')
if fpocket is None:
LOGGER.warning(
'Fpocket (version >= 3.0) was not found, please install it.')
return None
try:
from pandas import Index, DataFrame
except ImportError as ie:
LOGGER.warning(ie.__str__() + ' was found, please install it.')
return None
rcr = {(i, j): k if self._rib else self._ri[k]
for i, j, k in zip(self._ca.getChids(), self._ca.getResnums(),
self._ca.getResindices())}
writePDB('{}_pro'.format(self._title), self._heavy)
direc = '{}_pro_out'.format(self._title)
if not isdir(direc):
system('fpocket -f {}_pro.pdb'.format(self._title))
chdir(direc + '/pockets')
l = [x for x in listdir('.') if x.endswith('.pdb')]
l.sort(key=lambda x: int(x.partition('_')[0][6:]))
ps = []
for x in l:
with open(x, 'r') as f:
tmp0 = f.read()
tmp1 = [(x[1].strip(), float(x[2])) for x in findall(
r'(\w+\s\w+\s*-\s*)(.+):\s*([\d.-]+)(\n)', tmp0)]
fea, sco = list(zip(*tmp1))
ps.append(sco)
pdbs = parsePDB(l)
chdir('../..')
# ----- # ----- #
ps = array(ps)
pcn = {
int(pdb.getTitle().partition('_')[0][6:]):
set(zip(pdb.getChids().tolist(),
pdb.getResnums().tolist()))
for pdb in pdbs
}
pi = {p: [rcr[x] for x in crn] for p, crn in pcn.items()}
pzs_max = {k: max(self._zscore[v]) for k, v in pi.items()}
pzs_med = {k: median(self._zscore[v]) for k, v in pi.items()}
# ----- # ----- #
indices = Index(range(1, ps.shape[0] + 1), name='Pocket #')
columns = Index(fea, name='Feature')
self._df = DataFrame(index=indices, columns=columns, data=ps)
# ----- # ----- #
columns_zs = Index(['ESSA_max', 'ESSA_med', 'LHD'], name='Z-score')
zps = c_[list(pzs_max.values())]
zps = hstack((zps, c_[list(pzs_med.values())]))
zps = hstack(
(zps, zscore(self._df[['Local hydrophobic density Score']])))
self._df_zs = DataFrame(index=indices, columns=columns_zs, data=zps)
from os.path import sep as dirsep
import inspect
import tempfile
try:
import unittest2 as unittest
from unittest2 import TestCase, skipIf, skipUnless
except ImportError:
import unittest
from unittest import TestCase, skipIf, skipUnless
from prody.utilities import PLATFORM
from prody import LOGGER
from prody.utilities import which
NOPRODYCMD = which('prody') is None
WINDOWS = PLATFORM == 'Windows'
try:
import matplotlib
matplotlib.use('Agg')
except ImportError:
MATPLOTLIB = False
else:
try:
from matplotlib import pyplot
except ImportError:
MATPLOTLIB = False
else:
MATPLOTLIB = True