def test_calculation(self):
c = treeCl.Collection(input_dir=os.path.join(thisdir, 'data'),
param_dir=os.path.join(thisdir, 'data', 'cache'),
file_format='phylip',
show_progress=False)
dm = c.get_inter_tree_distances('geo', show_progress=False)
self.assertAlmostEqual(dm.df.values.sum(), 412.70677069540181)
def test_can_run_on_dna(self):
self.c = treeCl.Collection(input_dir=os.path.join(
thisdir, 'data', 'dna_alignments'),
file_format='phylip',
show_progress=False)
self.c.calc_trees(indices=[0], model='GTRGAMMA', show_progress=False)
self.assertFalse(self.c[0].parameters.ml_tree is None)
def setUp(self):
self.c = treeCl.Collection(input_dir=os.path.join(thisdir, 'data'),
trees_dir=os.path.join(
thisdir, 'data', 'trees'),
file_format='phylip',
show_progress=False)
self.tree1 = treeCl.Tree(self.c[0].tree)
self.tree2 = treeCl.Tree(self.c[1].tree)
def test_read_trees(self):
self.c = treeCl.Collection(input_dir=os.path.join(thisdir, 'data'),
trees_dir=os.path.join(
thisdir, 'data', 'trees'),
file_format='phylip',
show_progress=False)
rec = self.c[0]
self.assertEqual(
rec.parameters.ml_tree[:72],
'((((Sp1:1.48316688535948748573,(Sp4:1.16694627918414717271,((Sp8:0.00749'
)
def test_read_parameters(self):
self.c = treeCl.Collection(input_dir=os.path.join(thisdir, 'data'),
param_dir=os.path.join(
thisdir, 'data', 'cache'),
file_format='phylip',
show_progress=False)
rec = self.c[0]
self.assertEqual(
rec.parameters.nj_tree[:72],
'((((Sp1:1.47856,(Sp4:1.20999,((Sp8:0.00595845,Sp9:0.00469589):0.27853,Sp'
)
def test_scorer_can_write(self):
c = treeCl.Collection(input_dir=os.path.join(thisdir, 'data'),
param_dir=os.path.join(thisdir, 'data', 'cache'),
file_format='phylip',
show_progress=False)
raxml = treeCl.tasks.RaxmlTaskInterface()
sc = treeCl.Scorer(c, cache_dir=self.workingdir, task_interface=raxml)
p = treeCl.Partition([0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2])
sc.write_partition(p)
# check files were written
import glob
files = glob.glob(os.path.join(self.workingdir, '*.phy'))
self.assertTrue(len(files) > 0)
def generate_npbs(path, i):
c = treeCl.Collection(input_dir=path, file_format='phylip')
working_dir = get_dirs(path, i)['wdir']
# Check if work already done
work_done = True
for rec in c:
looking_for = '{}.phy'.format(os.path.join(working_dir, rec.name))
if not (os.path.exists(looking_for) and os.path.getsize(looking_for) > 0):
if not (os.path.exists(looking_for + '.bz2') and os.path.getsize(looking_for + '.bz2') > 0):
logger.error("File not found or is empty: {}".format(looking_for))
work_done = False
if not work_done:
npbs = c.permuted_copy()
if not os.path.exists(working_dir):
os.mkdir(working_dir)
for rec in npbs:
rec.write_alignment('{}.phy'.format(os.path.join(working_dir, rec.name)), 'phylip', True)
AlignIO.convert('{}.phy'.format(os.path.join(working_dir, rec.name)), 'phylip-relaxed', '{}.phy_'.format(os.path.join(working_dir, rec.name)), 'phylip-relaxed')
os.system('mv {} {}'.format('{}.phy_'.format(os.path.join(working_dir, rec.name)), '{}.phy'.format(os.path.join(working_dir, rec.name))))
#!/usr/bin/env python
import time
import treeCl
time.sleep(15)
c = treeCl.Collection(input_dir='/homes/kgori/scratch/simtest4',
file_format='phylip')
c.calc_pll_trees()
import treeCl
"""
The first point of call is the treeCl.Collection class.
This handles loading your data, and calculating the trees
and distances that will be used later.
This is how to load your data. This should be a directory
full of sequence alignments in fasta '*.fas' or phylip
'*.phy' formats. These can also be zipped using gzip or
bzip2, treeCl will load them directly.
"""
c = treeCl.Collection(input_dir='input_dir', file_format='phylip')
"""
Now it's time to calculate some trees. The simplest way to
do this is
"""
c.calc_trees()
"""
This uses RAxML to infer a tree for each alignment. We can
pass arguments to RAxML using keywords.
"""
c.calc_trees(
executable='raxmlHPC-PTHREADS-AVX', # specify raxml binary to use
threads=8, # use multithreaded raxml
model='PROTGAMMAWAGX', # this model of evolution
fast_tree=True) # use raxml's experimental fast tree search option
"""
We can use PhyML instead of RAxML. Switching programs is
done using a TaskInterface
"""
def setUp(self):
self.c = treeCl.Collection(input_dir=os.path.join(thisdir, 'data'),
param_dir=os.path.join(
thisdir, 'data', 'cache'),
file_format='phylip',
show_progress=False)
import treeCl c = treeCl.Collection(input_dir="/Users/kgori/scratch/simtest4", file_format="phylip") print c[5].get_distances() print c[5].chkdst() print c[5].get_bionj_tree() #conc = treeCl.Concatenation(c,[5,6,7,8,9]) #part = conc.qfile(protein_model='LGX') #p = conc.alignment.pll_get_instance(part, c[5].tree, 6)
def get_collection(path, i):
working_dir = get_dirs(path, i)['wdir']
return treeCl.Collection(input_dir=working_dir, file_format='phylip')
