def __init__(
self,
records=None,
input_dir=None,
file_format='fasta',
datatype=None,
tmpdir='/tmp',
calc_distances=False,
compression=None,
analysis=None,
):
self.tmpdir = directorycheck(tmpdir)
if records:
self.records = records
self.datatype = datatype or records[0].datatype
optioncheck(self.datatype, ['dna', 'protein'])
elif input_dir:
directorycheck(input_dir)
self.datatype = optioncheck(datatype, ['dna', 'protein'])
optioncheck(file_format, ['fasta', 'phylip'])
self.records = self.read_files(input_dir, file_format, compression)
else:
print 'Provide a list of records, or the path to a set of alignments'
if not self.records:
raise NoRecordsError(file_format, input_dir, compression)
if calc_distances:
self.calc_distances()
def embedding_plotter(
self, coordinates, dimensions, partition=None, add_sphere=False,
xlab='PCo1', ylab='PCo2', zlab='PCo3',
title='Trees embedded in dimension-reduced space',
outfile=False,
):
""" Points are coloured according to cluster membership specified
by Partition object (or all black if no Partition specified) """
optioncheck(dimensions, [2,3])
partition = partition or Partition(tuple([6]*len(self.collection.records)))
colours = 'bgrcmyk'
colour_mapping = np.array([colours[i] for i in partition.partition_vector])
fig = plt.figure()
if dimensions == 3:
ax = fig.add_subplot(111, projection='3d',
xlabel=xlab, ylabel=ylab, zlabel=zlab, title=title)
if add_sphere:
ax = self.sphere(ax)
else:
ax = fig.add_subplot(111,
xlabel=xlab, ylabel=ylab, title=title)
ax.scatter(*coordinates.T, color=colour_mapping)
# ax.set_aspect(1)
if outfile:
fig.savefig('{0}.pdf'.format(outfile))
return fig
def get_decomp(self, method='MDS', **kwargs):
optioncheck(method, ['MDS', 'spectral'])
cl = Clustering(self.dm)
if method == 'MDS':
return cl.MDS_decomp()
if method == 'spectral':
return cl.spectral_decomp(**kwargs)
def read_files(self, input_dir, file_format, compression=None):
""" Get list of alignment files from an input directory *.fa, *.fas and
*.phy files only
Stores in self.files """
optioncheck(compression, [None, 'gz', 'bz2'])
if file_format == 'fasta':
extensions = ['fa', 'fas', 'fasta']
elif file_format == 'phylip':
extensions = ['phy']
if compression:
extensions = ['.'.join([x, compression]) for x in extensions]
files = fileIO.glob_by_extensions(input_dir, extensions)
files.sort(key=sort_key)
return [
TrClSeq(f,
file_format=file_format,
datatype=self.datatype,
name=get_name(f),
tmpdir=self.tmpdir) for f in files
]
def __init__(
self,
records=None,
input_dir=None,
file_format='fasta',
datatype=None,
tmpdir='/tmp',
calc_distances=False,
compression=None,
analysis=None,
):
self.tmpdir = directorycheck(tmpdir)
if records:
self.records = records
self.datatype = datatype or records[0].datatype
optioncheck(self.datatype, ['dna', 'protein'])
elif input_dir:
directorycheck(input_dir)
self.datatype = optioncheck(datatype, ['dna', 'protein'])
optioncheck(file_format, ['fasta', 'phylip'])
self.records = self.read_files(input_dir, file_format, compression)
else:
print 'Provide a list of records, or the path to a set of alignments'
if not self.records:
raise NoRecordsError(file_format, input_dir, compression)
if calc_distances:
self.calc_distances()
def get_decomp(self, method='MDS', **kwargs):
optioncheck(method, ['MDS', 'spectral'])
cl = Clustering(self.dm)
if method == 'MDS':
return cl.MDS_decomp()
if method == 'spectral':
return cl.spectral_decomp(**kwargs)
def __new__(
cls,
trees,
metric,
tmpdir='/tmp',
dtype=float,
add_noise=False,
normalise=False,
):
optioncheck(metric, ['euc', 'geo', 'rf', 'wrf'])
input_array = get_distance_matrix(trees, metric, tmpdir,
normalise=normalise)
obj = np.asarray(input_array, dtype).view(cls)
obj.metric = metric
obj.tmpdir = tmpdir
if add_noise:
obj = obj.add_noise()
return obj
def embedding_plotter(
self,
coordinates,
dimensions,
partition=None,
add_sphere=False,
xlab='PCo1',
ylab='PCo2',
zlab='PCo3',
title='Trees embedded in dimension-reduced space',
outfile=False,
):
""" Points are coloured according to cluster membership specified
by Partition object (or all black if no Partition specified) """
optioncheck(dimensions, [2, 3])
partition = partition or Partition(
tuple([6] * len(self.collection.records)))
colours = 'bgrcmyk'
colour_mapping = np.array(
[colours[i] for i in partition.partition_vector])
fig = plt.figure()
if dimensions == 3:
ax = fig.add_subplot(111,
projection='3d',
xlabel=xlab,
ylabel=ylab,
zlabel=zlab,
title=title)
if add_sphere:
ax = self.sphere(ax)
else:
ax = fig.add_subplot(111, xlabel=xlab, ylabel=ylab, title=title)
ax.scatter(*coordinates.T, color=colour_mapping)
# ax.set_aspect(1)
if outfile:
fig.savefig('{0}.pdf'.format(outfile))
return fig
def __init__(
self,
records,
analysis,
max_guidetrees=10,
tmpdir=None,
datatype=None,
verbosity=0,
):
self.analysis = optioncheck(analysis, ['ml', 'nj', 'TreeCollection'])
self.max_guidetrees = max_guidetrees
self.records = records
self.datatype = datatype or records[0].datatype
self.verbosity = verbosity
optioncheck(self.datatype, ['protein', 'dna'])
self.tmpdir = tmpdir or records[0].tmpdir
self.concats = {}
self.history = []
def __init__(
self,
records,
analysis,
max_guidetrees=10,
tmpdir=None,
datatype=None,
verbosity=0,
):
self.analysis = optioncheck(analysis, ['ml', 'nj',
'TreeCollection'])
self.max_guidetrees = max_guidetrees
self.records = records
self.datatype = datatype or records[0].datatype
self.verbosity = verbosity
optioncheck(self.datatype, ['protein', 'dna'])
self.tmpdir = tmpdir or records[0].tmpdir
self.concats = {}
self.history = []
def __new__(
cls,
trees,
metric,
tmpdir='/tmp',
dtype=float,
add_noise=False,
normalise=False,
):
optioncheck(metric, ['euc', 'geo', 'rf', 'wrf'])
input_array = get_distance_matrix(trees,
metric,
tmpdir,
normalise=normalise)
obj = np.asarray(input_array, dtype).view(cls)
obj.metric = metric
obj.tmpdir = tmpdir
if add_noise:
obj = obj.add_noise()
return obj
def read_files(self, input_dir, file_format, compression=None):
""" Get list of alignment files from an input directory *.fa, *.fas and
*.phy files only
Stores in self.files """
optioncheck(compression, [None, 'gz', 'bz2'])
if file_format == 'fasta':
extensions = ['fa', 'fas', 'fasta']
elif file_format == 'phylip':
extensions = ['phy']
if compression:
extensions = ['.'.join([x, compression]) for x in extensions]
files = fileIO.glob_by_extensions(input_dir, extensions)
files.sort(key=sort_key)
return [TrClSeq(f, file_format=file_format, datatype=self.datatype,
name=get_name(f), tmpdir=self.tmpdir) for f in files]
def simulate(self, index_list, model=None):
""" Simulate a group of sequence alignments using ALF. Uses one of
{(GCB, JTT, LG, WAG - protein), (CPAM, ECM and ECMu - DNA)}, WAG by
default. TO DO: add parameterised models when I have a robust (probably
PAML) method of estimating them from alignment+tree """
if self.datatype == 'protein': # set some defaults
model = model or 'WAG'
optioncheck(model, [
'CPAM',
'ECM',
'ECMu',
'WAG',
'JTT',
'GCB',
'LG',
])
else:
model = model or 'ECM'
try:
optioncheck(model, ['CPAM', 'ECM', 'ECMu'])
except OptionError, e:
print 'Choose a DNA-friendly model for simulation:\n', e
return
def simulate(self, index_list, model=None):
""" Simulate a group of sequence alignments using ALF. Uses one of
{(GCB, JTT, LG, WAG - protein), (CPAM, ECM and ECMu - DNA)}, WAG by
default. TO DO: add parameterised models when I have a robust (probably
PAML) method of estimating them from alignment+tree """
if self.datatype == 'protein': # set some defaults
model = model or 'WAG'
optioncheck(model, [
'CPAM',
'ECM',
'ECMu',
'WAG',
'JTT',
'GCB',
'LG',
])
else:
model = model or 'ECM'
try:
optioncheck(model, ['CPAM', 'ECM', 'ECMu'])
except OptionError, e:
print 'Choose a DNA-friendly model for simulation:\n', e
return
import argparse
parser = argparse.ArgumentParser(description='Clustering optimiser')
parser.add_argument('-n', '--nclusters', type=int)
parser.add_argument('-f', '--format', default='phylip')
parser.add_argument('-d', '--datatype', default='protein')
parser.add_argument('-i', '--input_dir', default='./')
parser.add_argument('-c', '--compression', default=None)
parser.add_argument('-t', '--tmpdir', default='/tmp/')
parser.add_argument('-m', '--method', default='s')
parser.add_argument('-o', '--output', default=None)
args = parser.parse_args()
optioncheck(args.method, ['s', 'spectral', 'h', 'hierarchical', 'k', 'kmedoids', 'MDS', 'mds'])
def id_generator(size=6, chars=string.ascii_uppercase + string.digits):
return ''.join(random.choice(chars) for x in range(size))
new_tmpdir = tempfile.mkdtemp(prefix='tmpwrap_mgp_', dir=args.tmpdir)
c = Collection(input_dir=args.input_dir,
compression=args.compression, file_format=args.format, datatype=args.datatype,
tmpdir=new_tmpdir)
c.calc_NJ_trees()
dm = c.distance_matrix('euc')
cl = Clustering(dm)
def decomp_to_coords(self, decomp, dimensions, normalise=False):
optioncheck(dimensions, [2,3])
coords = decomp.coords_by_dimension(dimensions)[0]
return coords.normalise_rows() if normalise else coords
def decomp_to_coords(self, decomp, dimensions, normalise=False):
optioncheck(dimensions, [2, 3])
coords = decomp.coords_by_dimension(dimensions)[0]
return coords.normalise_rows() if normalise else coords
