def parse_paml(pamout, model):
'''
parser function for codeml files,
with values of w,dN,dS etc... dependending of the model
tested.
'''
# if multiple dataset in same file we divide the outfile and model.name+x
if not '*' in str(model.properties['params']['ndata']):
divide_data(pamout, model)
return
all_lines = open(pamout).readlines()
# if we do not have tree, load it
if model._tree == None:
from ete3.evol import EvolTree
model._tree = EvolTree(re.findall('\(.*\);', ''.join(all_lines))[2])
model._tree._label_as_paml()
# starts parsing
for i, line in enumerate(all_lines):
if line is '\n':
continue
# codon frequency
if line.startswith('Codon frequencies under model'):
model.stats['codonFreq'] = []
for j in range(16):
line = list(
map(float, re.findall('\d\.\d+', all_lines[i + j + 1])))
model.stats['codonFreq'] += [line]
continue
if 'codonFreq' not in model.stats:
continue
######################
# start serious staff
line = line.rstrip()
# lnL and number of parameters
if line.startswith('lnL'):
try:
line = re.sub('.* np: *(\d+)\): +(-\d+\.\d+).*', '\\1 \\2',
line)
model.stats['np'] = int(line.split()[0])
model.stats['lnL'] = float(line.split()[1])
except ValueError:
line = re.sub('.* np: *(\d+)\): +(nan).*', '\\1 \\2', line)
model.stats['np'] = int(line.split()[0])
model.stats['lnL'] = float('-inf')
continue
# get labels of internal branches
if line.count('..') >= 2:
labels = re.findall('\d+\.\.\d+', line + ' ')
_check_paml_labels(model._tree, labels, pamout, model)
continue
# retrieve kappa
if line.startswith('kappa '):
try:
model.stats['kappa'] = float(
re.sub('.*(\d+\.\d+).*', '\\1', line))
except ValueError:
model.stats['kappa'] = 'nan'
# retrieve dS dN t w N S and if present, errors. from summary table
if line.count('..') == 1 and line.startswith(' '):
if not re.match(' +\d+\.\.\d+ +\d+\.\d+ ', line):
if re.match(' +( +\d+\.\d+){8}', all_lines[i + 1]):
_get_values(model,
line.split()[0] + ' ' + all_lines[i + 1])
continue
_get_values(model, line)
continue
#MY_PATH = '/home/francisco/toolbox/ete3-codeml/doc/tutorial/examples/'
MY_PATH = ''
TREE_PATH = MY_PATH + re.sub('\./', '', TREE_PATH)
ALG_PATH = MY_PATH + re.sub('\./', '', ALG_PATH )
###
# load tree
print '\n ----> we create a EvolTree object, and give to him a topology, from',
print TREE_PATH
out = True
while out == True:
try:
T = EvolTree(TREE_PATH)
out = False
except:
sys.stderr.write('Bad path for working directory. Enter new path or quit("Q"):\n')
PATH = raw_input('')
if PATH.startswith('q') or PATH.startswith('Q'):
sys.exit()
TREE_PATH = "./measuring_%s_tree.nw" % (typ)
ALG_PATH = "./alignment_%s_measuring_evol.fasta" % (typ)
TREE_PATH = PATH + re.sub('\./', '', TREE_PATH)
ALG_PATH = PATH + re.sub('\./', '', ALG_PATH )
print T
print '\n ----> and an alignment from: \n'+ALG_PATH+'\n\n'
T.link_to_alignment(ALG_PATH)
#MY_PATH = '/home/francisco/toolbox/ete3-codeml/doc/tutorial/examples/'
MY_PATH = ''
TREE_PATH = MY_PATH + re.sub('\./', '', TREE_PATH)
ALG_PATH = MY_PATH + re.sub('\./', '', ALG_PATH)
###
# load tree
print '\n ----> we create a EvolTree object, and give to him a topology, from',
print TREE_PATH
out = True
while out == True:
try:
T = EvolTree(TREE_PATH)
out = False
except:
sys.stderr.write(
'Bad path for working directory. Enter new path or quit("Q"):\n')
PATH = raw_input('')
if PATH.startswith('q') or PATH.startswith('Q'):
sys.exit()
TREE_PATH = "./measuring_%s_tree.nw" % (typ)
ALG_PATH = "./alignment_%s_measuring_evol.fasta" % (typ)
TREE_PATH = PATH + re.sub('\./', '', TREE_PATH)
ALG_PATH = PATH + re.sub('\./', '', ALG_PATH)
print T
print '\n ----> and an alignment from: \n' + ALG_PATH + '\n\n'
T.link_to_alignment(ALG_PATH)
from ete3.evol import EvolTree
import sys, re
typ = 'S'
#while typ != 'L' and typ != 'S':
# typ = raw_input (\
# "choose kind of example [L]ong or [S]hort, hit [L] or [S]:\n")
TREE_PATH = "data/S_example/measuring_%s_tree.nw" % (typ)
ALG_PATH = "data/S_example/alignment_%s_measuring_evol.fasta" % (typ)
WORKING_PATH = "data/S_example/paml/"
MY_PATH = ''
TREE_PATH = MY_PATH + re.sub('\./', '', TREE_PATH)
ALG_PATH = MY_PATH + re.sub('\./', '', ALG_PATH )
T = EvolTree (TREE_PATH)
T.link_to_alignment (ALG_PATH)
T.workdir = (WORKING_PATH)
T.link_to_evol_model(T.workdir + '/fb/out','fb')
T.link_to_evol_model(T.workdir + '/M1/out','M1')
T.link_to_evol_model(T.workdir + '/M2/out','M2')
T.show(histfaces=['M2'])
sys.stderr.write('\n\nThe End.\n\n')
