except AssertionError:
print 'Can\'t get partitions to match'
return -1
return partition_list
sort_key = lambda item: tuple((int(num) if num else alpha) for (num,alpha) in re.findall(r'(\d+)|(\D+)', item.name))
print 'loading recs...'
recs = sorted(load_records(sys.argv[1], sys.argv[2]), key=sort_key)
print 'loading crecs...'
crecs = load_records(sys.argv[3])
d = make_target_dict(crecs)
print 'done.'
t = [1]*15+[2]*15+[3]*15+[4]*15
c = SequenceCollection()
c.records = recs
for metric in ['sym', 'euc', 'geo']:
c.put_distance_matrices(metric, gtp_path = '/homes/kgori/research/clustering_project/class_files')
for method in ['single', 'complete', 'average', 'ward', 'MDS', 'spectral', 'kmedoids']:
p = order(rebuild_partitions(recs, d, metric, method))
c.clustering.partitions[(metric, method, 4)] = p
c.clustering.partitions['true'] = t
c.put_clusters()
print '(Done).'
for rec in c.get_cluster_records():
try:
rec.tree = d[rec.name].tree
except KeyError:
5) Compare scores derived from clusters to random permutation of the original data
either by making a copy of the SequenceCollection object, with clusters made up
of the same number of genes with the same number of characters, or by randomising
the alignments and performing hierarchical clustering on the randomised data
if the former, do rand1 = col.make_randomised_copy
if the latterm do rand2 = SequenceCollection(records=col.get_randomised_alignments(),
datatype = 'protein')
"""
# indir = '/Users/kgori/git/kevin/yeast_data/MSA'
indir = '/Users/kgori/git/kevin/data/simulated_data/eight/MSA'
col = SequenceCollection(indir, datatype='protein')
ran = SequenceCollection(records=col.get_randomised_alignments(), datatype='protein')
col.put_trees_parallel()
ran.put_trees_parallel()
col.put_partitions(metrics=['euc','rf','sym'], linkages=['ward'], nclasses=[2,3,4,5,6,7,8,9,10])
ran.put_partitions(metrics=['euc','rf','sym'], linkages=['ward'], nclasses=[2,3,4,5,6,7,8,9,10])
col.put_clusters()
col.put_cluster_trees_parallel()
ran.put_clusters()
ran.put_cluster_trees_parallel()
rn2 = col.make_randomised_copy()
r1 = ran.get_clusters()
r2 = rn2.get_clusters()
cl = col.get_clusters()
'\t')[1])
for rec in phymlrecords:
rec.datatype = 'dna'
for rec in bionjrecords:
rec.datatype = 'dna'
try:
assert len(phymlrecords) == len(bionjrecords) == 60
except:
print 'Missing records in {0}'.format(indir)
sys.exit(1)
phyml_sc = SequenceCollection(records=phymlrecords,
datatype='dna',
helper=os.environ['DARWINHELPER'],
tmpdir=tmpdir,
get_distances=False)
bionj_sc = SequenceCollection(records=bionjrecords,
datatype='dna',
helper=os.environ['DARWINHELPER'],
tmpdir=tmpdir,
get_distances=False)
phyml_sc.put_partitions(
['geo', 'euc', 'sym'],
['single', 'complete', 'average', 'ward', 'kmedoids', 'MDS', 'spectral'],
4,
gtp_path=
'/net/isilon7/nobackup/research/goldman/kevin/clustering_project/class_files',
tmpdir=tmpdir)
progname = re.compile('[A-Za-z0-9.-_]+').search(sys.argv[0]).group()
desc = 'Read in a SequenceCollection from disk and dump records'
input_help = 'Filepath+name of gzipped SequenceCollection object'
output_help = 'Directory to dump files in'
choice_help = \
'\n'.join(['Choose to dump post-clustering concatenated records',
'instead of pre-clustering single records'])
parser = argparse.ArgumentParser(prog=progname, description=desc)
parser.add_argument('-i', dest='input_file', help=input_help, type=str)
parser.add_argument('-o', dest='output_dir', help=output_help, type=str)
parser.add_argument('-c', dest='cluster_recs', action='store_true')
args = parser.parse_args()
input_file = args.input_file
output_dir = args.output_dir.rstrip('/')
cluster_recs = args.cluster_recs
filecheck_and_quit(input_file)
directorycheck_and_make(output_dir)
from sequence_collection import SequenceCollection
sc = SequenceCollection.gunzip(input_file)
if cluster_recs:
records = sc.get_cluster_records()
sc.dump_records(output_dir, records)
else:
records = sc.get_records() # should be default anyway, but explicit
sc.dump_records(output_dir,
records) # is better than implicit, and all that
])
else:
helper = os.environ['DARWINHELPER']
try:
TMPDIR = os.environ['TEMPORARY_DIRECTORY']
except:
TMPDIR = '/tmp'
### MAIN
if treeprog == 'treecollection':
get_distances = True
else:
get_distances = False
sc = SequenceCollection(indir, file_format=file_format,
gtp_path=gtp_path, datatype=datatype,
get_distances=get_distances, tmpdir=TMPDIR, helper=helper)
sc.put_trees(program=treeprog)
sc.put_partitions(['geo', 'euc', 'rf'], [
'average',
'complete',
'kmedoids',
'MDS',
'single',
'spectral00',
'spectral01',
'spectral10',
'spectral11',
'ward',
], nclasses, recalculate=True)
'average',
'ward',
'kmedoids',
'spectral',
'MDS',
]
calc_varinf = False
if os.path.isfile(outf):
sys.exit(1)
sc = SequenceCollection(
seqdir,
tmpdir=TMPDIR,
gtp_path=GTP_PATH,
helper=HELPER,
file_format=format,
datatype=datatype,
parallel_load=False,
get_distances=False,
)
sc.put_trees(program='bionj', model='GTR', tmpdir=TMPDIR, ncat=4,
datatype='nt')
# sc.put_distance_matrices(['rf'])
# print sc.get_distance_matrices()['rf']
# sys.exit()
sc.put_partitions('rf', methods, nclasses, recalculate=True)
sc.put_partitions('euc', methods, nclasses, recalculate=True)
sc.put_partitions('geo', methods, nclasses, recalculate=True)
#!/usr/bin/python
# -*- coding: utf-8 -*-
from sequence_collection import SequenceCollection
import cPickle
import time
indir = '/Users/kgori/git/kevin/data/simulated_data/small/MSA'
print 'test directory = ', indir
load_start = time.time()
print 'loading sequences (parallel)'
col = SequenceCollection(indir, datatype='protein')
print col
load_end = time.time()
tcpar_start = time.time()
print 'putting TC trees (parallel)'
col.put_trees_parallel(program='treecollection', tmpdir='/tmp')
for rec in col.records:
print rec.name
print rec.tree
tcpar_end = time.time()
par_start = time.time()
print 'Putting partitions'
col.put_partitions(metrics=['sym', 'euc'],
linkages=['ward', 'single'],
nclasses=[3, 4, 5, 6])
print col.get_partitions()
from pylab import *
print 'done.'
def print_dict(d):
for k in sorted(d):
print d
np.set_printoptions(linewidth=200, precision=3)
sc = SequenceCollection(
'/Users/kgori/scratch/chk/aa_alignments/',
get_distances=False,
file_format='phylip',
helper='/Users/kgori/git/kevin/clustering_project/class_files/DV_wrapper.drw'
,
parallel_load=True,
gtp_path='/Users/kgori/git/kevin/clustering_project/class_files/',
tmpdir='/tmp',
datatype='protein',
)
sc_yeast = SequenceCollection(
'/Users/kgori/scratch/yeast_MSA',
get_distances=False,
file_format='phylip',
helper='/Users/kgori/git/kevin/clustering_project/class_files/DV_wrapper.drw'
,
parallel_load=True,
gtp_path='/Users/kgori/git/kevin/clustering_project/class_files/',
tmpdir='/tmp/yeast',
from tree import Tree
from clustering import Clustering
import cPickle
import time
import os
import copy
import numpy as np
np.set_printoptions(precision=2, linewidth=200)
indir = '/Users/kgori/git/kevin/data/simulated_data/small/MSA'
print 'test directory = ', indir
load_start = time.time()
print 'loading sequences (parallel)'
col = SequenceCollection(indir, datatype='protein')
print col
load_end = time.time()
tcseq_start = time.time()
print 'getting TC trees (sequential)'
col.get_trees(program='treecollection', tmpdir='/tmp')
for rec in col.records:
print rec.name
print rec.tree
tcseq_end = time.time()
tcpar_start = time.time()
print 'getting TC trees (parallel)'
col.get_trees_parallel(program='treecollection', tmpdir='/tmp')
for rec in col.records:
from pylab import *
print 'done.'
def print_dict(d):
for k in sorted(d):
print d
np.set_printoptions(linewidth=200, precision=3)
sc = SequenceCollection(
'/Users/kgori/scratch/chk/aa_alignments/',
get_distances=False,
file_format='phylip',
helper=
'/Users/kgori/git/kevin/clustering_project/class_files/DV_wrapper.drw',
parallel_load=True,
gtp_path='/Users/kgori/git/kevin/clustering_project/class_files/',
tmpdir='/tmp',
datatype='protein',
)
sc_yeast = SequenceCollection(
'/Users/kgori/scratch/yeast_MSA',
get_distances=False,
file_format='phylip',
helper=
'/Users/kgori/git/kevin/clustering_project/class_files/DV_wrapper.drw',
parallel_load=True,
gtp_path='/Users/kgori/git/kevin/clustering_project/class_files/',
tmpdir='/tmp/yeast',
#!/usr/bin/python
# -*- coding: utf-8 -*-
from sequence_collection import SequenceCollection
import cPickle
import time
indir = "/Users/kgori/git/kevin/data/simulated_data/small/MSA"
print "test directory = ", indir
load_start = time.time()
print "loading sequences (parallel)"
col = SequenceCollection(indir, datatype="protein")
print col
load_end = time.time()
tcpar_start = time.time()
print "putting TC trees (parallel)"
col.put_trees_parallel(program="treecollection", tmpdir="/tmp")
for rec in col.records:
print rec.name
print rec.tree
tcpar_end = time.time()
par_start = time.time()
print "Putting partitions"
col.put_partitions(metrics=["sym", "euc"], linkages=["ward", "single"], nclasses=[3, 4, 5, 6])
print col.get_partitions()
par_end = time.time()
type=fpath,
default='/tmp')
parser.add_argument('-data', '--datatype', help='datatype', default=None)
args = vars(parser.parse_args())
outdir = args['directory']
program = args['program']
model = args['model']
ncat = args['ncat']
datatype = args['datatype']
gtp_path = os.environ['GTP_PATH']
tmpdir = os.environ['TEMPORARY_DIRECTORY']
tmpdir = args['tmpdir']
print 'Reading alignments into SequenceRecord object'
seq = SequenceCollection('{0}/dna_alignments'.format(outdir),
datatype='dna',
tmpdir=tmpdir,
helper=os.environ['DARWINHELPER'])
print 'Calculating trees'
print program, model, datatype, ncat, tmpdir
seq.put_trees(program=program,
model=model,
datatype=datatype,
ncat=ncat,
tmpdir=tmpdir)
print 'doing geodesic distance matrices'
seq.put_distance_matrices('geo', gtp_path=gtp_path, tmpdir=tmpdir)
print 'doing euc distance matrices'
seq.put_distance_matrices('euc')
print 'doing sym distance matrices'
seq.put_distance_matrices('sym')
tree.read_from_file('{0}/trees/{1}.nwk'.format(working_dir,
name))
dv_matrix_strip_header = '\n'.join(dv_matrix.split('\n'
)[2:]).rstrip()
labels_strip_header = labels.split('\n')[1].rstrip()
record = TCSeqRec()
record.dv = [(dv_matrix_strip_header, labels_strip_header)]
record.tree = tree
record.name = name
record.headers = labels_strip_header.split()
record.sequences = ['' for _ in record.headers]
record._update()
records.append(record)
collection = SequenceCollection(records=records, get_distances=False,
gtp_path=os.environ['GTP_PATH'])
collection.put_distance_matrices('rf')
T = \
collection.Clustering.run_spectral_rotate(collection.distance_matrices['rf'
])
collection.partitions[T] = Partition(T)
collection.clusters_to_partitions[('rf', 'spectral_rotate', max(T))] = T
collection.concatenate_records()
cluster_recs = collection.get_cluster_records()
number_of_clusters = len(cluster_recs)
for j in range(number_of_clusters):
record = cluster_recs[j]
record_dv = record.dv[0]
labels = record.dv[1]
datatype = args['datatype']
score = args['score']
directorycheck_and_quit(input_dir)
directorycheck_and_make(tmpdir)
gtp_path = os.environ['GTP_PATH']
helper = os.environ['DARWINHELPER']
from sequence_collection import SequenceCollection
sc = SequenceCollection(
input_dir,
file_format='phylip',
datatype=datatype,
helper=helper,
gtp_path=gtp_path,
tmpdir=tmpdir,
overwrite=True,
)
sc.load_phyml_results(input_dir, program=None)
sc.quality_scores = {}
for dist in distance:
(_, qs) = sc.autotune(dist,
max_groups=max_clusters,
min_groups=min_clusters)
sc.quality_scores[dist] = qs
cluster_range = range(min_clusters, max_clusters + 1)
if min_clusters > 1:
cluster_range.insert(0, 1)
(simdir, 'bionj_clustering'))),
key=sort_key)
else:
records = sorted(load_records('/'.join((simdir, record_dir)),
'*.ml.pickle'),
key=sort_key)
cluster_records = sorted(load_records('/'.join(
(simdir, 'phyml_clustering'))),
key=sort_key)
cluster_dic = make_target_dict(cluster_records)
print '(Done).'
#rebuild sequenceCollection object
sc = SequenceCollection()
sc.records = records
print 'Generating distance matrices...'
for metric in ['euc', 'sym', 'geo']:
sc.put_distance_matrices(
metric,
gtp_path='/homes/kgori/research/clustering_project/class_files',
tmpdir=tmpdir)
for method in [
'single', 'complete', 'ward', 'average', 'spectral', 'MDS',
'kmedoids'
]:
partition = rebuild_partitions(records,
cluster_dic,
metric=metric,
'average',
'ward',
'kmedoids',
'spectral',
'MDS',
]
calc_varinf = False
if os.path.isfile(outf):
sys.exit(1)
sc = SequenceCollection(
seqdir,
tmpdir=TMPDIR,
gtp_path=GTP_PATH,
helper=HELPER,
file_format=format,
datatype=datatype,
parallel_load=False,
get_distances=False,
)
sc.put_trees(program='bionj',
model='GTR',
tmpdir=TMPDIR,
ncat=4,
datatype='nt')
# sc.put_distance_matrices(['rf'])
# print sc.get_distance_matrices()['rf']
# sys.exit()
parser.add_argument('-m', '--model', help='which model to use in phylogenetic inference', default=None)
parser.add_argument('-n', '--ncat', help='number of categories for gamma distributed rates', default=1)
parser.add_argument('-t', '--tmpdir', help='temporary directory', type=fpath, default='/tmp')
parser.add_argument('-data', '--datatype', help='datatype', default=None)
args = vars(parser.parse_args())
outdir = args['directory']
program = args['program']
model = args['model']
ncat = args['ncat']
datatype = args['datatype']
gtp_path = os.environ['GTP_PATH']
tmpdir = os.environ['TEMPORARY_DIRECTORY']
tmpdir = args['tmpdir']
print 'Reading alignments into SequenceRecord object'
seq = SequenceCollection('{0}/dna_alignments'.format(outdir), datatype='dna', tmpdir=tmpdir, helper=os.environ['DARWINHELPER'])
print 'Calculating trees'
print program,model,datatype,ncat,tmpdir
seq.put_trees(program=program, model=model, datatype=datatype, ncat=ncat, tmpdir=tmpdir)
print 'doing geodesic distance matrices'
seq.put_distance_matrices('geo', gtp_path=gtp_path, tmpdir=tmpdir)
print 'doing euc distance matrices'
seq.put_distance_matrices('euc')
print 'doing sym distance matrices'
seq.put_distance_matrices('sym')
print 'getting score for true clustering'
with open('{0}/treedistances.txt'.format(outdir)) as file:
T = file.readline().rstrip().split('\t')[1][1:-1].split(', ')
seq.clustering.partitions['true'] = T
seq.put_clusters()
bionjrecords = sorted([cPickle.load(file(x)) for x in bionjpickles], key=lambda x:sort_key(x.name))
true = eval(open('{0}/treedistances.txt'.format(indir)).read().split('\n')[0].split('\t')[1])
for rec in phymlrecords:
rec.datatype='dna'
for rec in bionjrecords:
rec.datatype='dna'
try:
assert len(phymlrecords) == len(bionjrecords) == 60
except:
print 'Missing records in {0}'.format(indir)
sys.exit(1)
phyml_sc = SequenceCollection(records=phymlrecords, datatype='dna', helper=os.environ['DARWINHELPER'],tmpdir=tmpdir, get_distances=False)
bionj_sc = SequenceCollection(records=bionjrecords, datatype='dna', helper=os.environ['DARWINHELPER'],tmpdir=tmpdir, get_distances=False)
phyml_sc.put_partitions(['geo','euc','sym'],['single','complete','average','ward','kmedoids','MDS','spectral'], 4, gtp_path='/net/isilon7/nobackup/research/goldman/kevin/clustering_project/class_files', tmpdir=tmpdir)
bionj_sc.put_partitions(['geo','euc','sym'],['single','complete','average','ward','kmedoids','MDS','spectral'], 4, gtp_path='/net/isilon7/nobackup/research/goldman/kevin/clustering_project/class_files', tmpdir=tmpdir)
phyml_sc.clustering.partitions['true']=true
bionj_sc.clustering.partitions['true']=true
phyml_sc.put_clusters()
bionj_sc.put_clusters()
if not os.path.isdir('{0}/phyml_clustering'.format(indir)):
os.mkdir('{0}/phyml_clustering'.format(indir))
if not os.path.isdir('{0}/bionj_clustering'.format(indir)):
os.mkdir('{0}/bionj_clustering'.format(indir))
col.put_cluster_trees_parallel()
5) Compare scores derived from clusters to random permutation of the original data
either by making a copy of the SequenceCollection object, with clusters made up
of the same number of genes with the same number of characters, or by randomising
the alignments and performing hierarchical clustering on the randomised data
if the former, do rand1 = col.make_randomised_copy
if the latterm do rand2 = SequenceCollection(records=col.get_randomised_alignments(),
datatype = 'protein')
"""
# indir = '/Users/kgori/git/kevin/yeast_data/MSA'
indir = '/Users/kgori/git/kevin/data/simulated_data/eight/MSA'
col = SequenceCollection(indir, datatype='protein')
ran = SequenceCollection(records=col.get_randomised_alignments(),
datatype='protein')
col.put_trees_parallel()
ran.put_trees_parallel()
col.put_partitions(metrics=['euc', 'rf', 'sym'],
linkages=['ward'],
nclasses=[2, 3, 4, 5, 6, 7, 8, 9, 10])
ran.put_partitions(metrics=['euc', 'rf', 'sym'],
linkages=['ward'],
nclasses=[2, 3, 4, 5, 6, 7, 8, 9, 10])
col.put_clusters()
col.put_cluster_trees_parallel()
ran.put_clusters()
ran.put_cluster_trees_parallel()
rn2 = col.make_randomised_copy()
"""
print 'loading sequences...'
col = SequenceCollection(indir, helper=helper, tmpdir=tmpdir)
print 'getting trees...'
col.put_trees_parallel(program='phyml',tmpdir=tmpdir)
print 'getting partitions...'
col.put_partitions(metrics=['sym'],linkages=['ward'],nclasses=[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17])
#print 'randomizing bytes...'
col.put_clusters()
#print 'immanentizing the eschaton...'
col.put_cluster_trees_parallel(program='phyml',tmpdir=tmpdir)
"""
plottable = []
#plottable.append(add_to_plot(col,'sym','ward'))
col = cPickle.load(file('col.pickle'))
#print 'whipping into frenzy...'
for i in range(1):
r = SequenceCollection(records=col.get_randomised_alignments(), helper=helper, tmpdir=tmpdir)
r.put_trees_parallel(program='phyml',tmpdir=tmpdir)
r.put_partitions(metrics=['sym'],linkages=['ward'],nclasses=[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17])
r.put_clusters()
r.put_cluster_trees_parallel(program='phyml',tmpdir=tmpdir)
plottable.append(add_to_plot(r, 'sym', 'ward'))
cPickle.dump(plottable, file('plottable{0}.pickle'.format(index),'w'))
#cPickle.dump(col, file('col.pickle','w'))
#!/usr/bin/python
# -*- coding: utf-8 -*-
from sequence_collection import SequenceCollection
import time
indir = '/Users/kgori/git/kevin/data/real_data/yeast_data/MSA'
print 'test directory = ', indir
load_start = time.time()
print 'loading sequences (parallel)'
col = SequenceCollection(indir, datatype='dna')
print col
load_end = time.time()
col.put_trees_parallel()
col.put_partitions(metrics=['sym','geodesic'], linkages=['ward'], nclasses=[2,3,4,5,6,7,8,9,10,11,12,13,14,15,16])
col.put_clusters()
col.put_cluster_trees_parallel()
timings = [
load_end - load_start
]
print 'time = {0:.3f}'.format(*timings)
from tree import Tree
from clustering import Clustering
import cPickle
import time
import os
import copy
import numpy as np
np.set_printoptions(precision=2, linewidth=200)
indir = '/Users/kgori/git/kevin/data/simulated_data/small/MSA'
print 'test directory = ', indir
load_start = time.time()
print 'loading sequences (parallel)'
col = SequenceCollection(indir, datatype='protein')
print col
load_end = time.time()
tcseq_start = time.time()
print 'getting TC trees (sequential)'
col.get_trees(program='treecollection', tmpdir='/tmp')
for rec in col.records:
print rec.name
print rec.tree
tcseq_end = time.time()
tcpar_start = time.time()
print 'getting TC trees (parallel)'
col.get_trees_parallel(program='treecollection', tmpdir='/tmp')
for rec in col.records:
datatype = args['datatype']
score = args['score']
directorycheck_and_quit(input_dir)
directorycheck_and_make(tmpdir)
gtp_path = os.environ['GTP_PATH']
helper = os.environ['DARWINHELPER']
from sequence_collection import SequenceCollection
sc = SequenceCollection(
input_dir,
file_format='phylip',
datatype=datatype,
helper=helper,
gtp_path=gtp_path,
tmpdir=tmpdir,
overwrite=True,
)
sc.load_phyml_results(input_dir, program=None)
sc.quality_scores = {}
for dist in distance:
(_, qs) = sc.autotune(dist, max_groups=max_clusters,
min_groups=min_clusters)
sc.quality_scores[dist] = qs
cluster_range = range(min_clusters, max_clusters+1)
if min_clusters > 1:
cluster_range.insert(0, 1)
sc.put_partitions(distance, method, cluster_range)
print 'Reading records...'
# some initialisation
if program == 'bionj':
records = sorted(load_records('/'.join((simdir, record_dir)), '*.nj.pickle'), key=sort_key)
cluster_records = sorted(load_records('/'.join((simdir, 'bionj_clustering'))), key=sort_key)
else:
records = sorted(load_records('/'.join((simdir, record_dir)), '*.ml.pickle'), key=sort_key)
cluster_records = sorted(load_records('/'.join((simdir, 'phyml_clustering'))), key=sort_key)
cluster_dic = make_target_dict(cluster_records)
print '(Done).'
#rebuild sequenceCollection object
sc = SequenceCollection()
sc.records = records
print 'Generating distance matrices...'
for metric in ['euc', 'sym', 'geo']:
sc.put_distance_matrices(metric, gtp_path = '/homes/kgori/research/clustering_project/class_files', tmpdir=tmpdir)
for method in ['single', 'complete', 'ward', 'average', 'spectral', 'MDS', 'kmedoids']:
partition = rebuild_partitions(records, cluster_dic, metric=metric, method=method)
sc.clustering.partitions[(metric, method, 4)] = partition
sc.clustering.partitions['true'] = [1]*15 + [2]*15 + [3]*15 + [4]*15
sc.put_clusters()
print '(Done).'
for rec in sc.get_cluster_records():
import argparse
import re
import sys
from errors import filecheck_and_quit, directorycheck_and_quit
progname = re.compile('[A-Za-z0-9.-_]+').search(sys.argv[0]).group()
desc = 'Read in a SequenceCollection from disk and print scores'
input_help = 'Filepath+name of gzipped SequenceCollection object'
category_choices = ['Observed', 'Randomised', 'Simulated', 'NA']
parser = argparse.ArgumentParser(prog=progname, description=desc)
parser.add_argument('-i', dest='input_file', help=input_help, type=str)
parser.add_argument('-t', dest='phyml_dir', help=input_help, type=str)
args = parser.parse_args()
input_file = args.input_file
phyml_dir = args.phyml_dir.rstrip('/')
filecheck_and_quit(input_file)
directorycheck_and_quit(phyml_dir)
from sequence_collection import SequenceCollection
sc = SequenceCollection.gunzip(input_file)
cluster_records = sc.get_cluster_records()
sc.load_phyml_results(phyml_dir, records=cluster_records,
use_hashname=True)
sc.update_scores()
sc.gzip(input_file)
print indir
rob.r('library(ape)')
rob.r('library(phangorn)')
print 'r libraries loaded'
alignments_dir = '{0}/dna_alignments'.format(indir)
if os.path.isfile('{0}/tmpPickle.pkl'.format(indir)):
sc = cPickle.load(file('{0}/tmpPickle.pkl'.format(indir)))
else:
sc = SequenceCollection(
alignments_dir,
file_format='fasta',
datatype='dna',
gtp_path=gtp_path,
helper=helper,
tmpdir=tmpdir,
)
sc.put_trees(program='bionj')
sc.put_partitions('geo', 'spectral', 4)
sc.concatenate_records()
sc.put_cluster_trees(program='bionj')
cPickle.dump(sc, open('{0}/tmpPickle.pkl'.format(indir), 'w'))
print 'SC object available'
# Plot the heatmap of the distance matrix
dm = sc.get_distance_matrices()['geo']
p = sc.partitions[sc.clusters_to_partitions[('geo', 'spectral', 4)]]