def run_evol_py(tree,
alg,
branch_model,
site_models,
workir='data/evol_output',
tool_dir="ete3_apps/bin"):
print(tree, alg, branch_model, site_models)
builtin_apps_path = None
builtin_apps_path = os.path.join(os.path.split(ete3_path)[0], tool_dir)
tree = EvolTree(tree, binpath=builtin_apps_path)
tree.link_to_alignment(alg)
tree.workdir = workir
###branch model
if branch_model:
branch_model = str(branch_model)
tree.run_model(branch_model)
print(tree.get_evol_model(branch_model))
### site model
for site_model in site_models:
tree.run_model(site_model)
#tree.run_model('SLR.lele')
global evol_output_dir, final_evol_tree
evol_output_dir = workir
final_evol_tree = evol_output_dir + '/tree_evol_result.png'
tree.render(final_evol_tree,
layout=evol_clean_layout,
histfaces=site_models)
return tree
def run(self, pamlsrc, output_folder, model='M1'):
"""Run PAML using ETE.
The default model is M1 as it is best for orthology inference in
our case. You can use models `M2`, `M0`, `M3`.
Ensure that you have the correct path to your codeml binary. It should
be in the paml `/bin`.
:param pamlsrc: Path to the codemly binary.
:param output_folder: The name of the output folder.
:param model: The model to be used. (Default value = 'M1')
"""
# Import the newick tree
tree = EvolTree('temptree.nw')
# Import the alignment
tree.link_to_alignment(self.alignmentfile)
tree.workdir = self.workdir
# Set the binpath of the codeml binary
tree.execpath = pamlsrc
# Run the model M1, M2, M3, or M0
model_path = model + '.' + output_folder
tree.run_model(model_path)
self.ete3paml_log.info('Codeml is generating data in %s.' % model_path)
def main(self):
"""The main function for running the test."""
print("Running model %s paml on input." % str(self.defaultmodel))
tree = EvolTree(self.tree) # Import the newick tree
tree.link_to_alignment(self.alignment) # Import the alignment
tree.workdir = self.workdir # Set the working directory
tree.execpath = self.pamlpath # Set the binpath of the codeml binary
tree.run_model(self.defaultmodel) # Run the codeml model
def count_omega(align_file, gene_name):
print(gene_name)
tree = EvolTree(tree_file)
tree.link_to_alignment(align_file)
#
# #free branch ratio count
tree.run_model('fb')
fb_results = tree.get_evol_model('fb')
print(fb_results)
with open(temp, 'w') as temp_file:
temp_file.write(str(fb_results))
write_in_table(gene_name)
def ete3paml(gene, paml_path, workdir='data/paml-output/', model='M1'):
""" Use ETE3's integration with PAML"""
# Import the species tree to compare species that are present in alignment
# file
t = Tree('data/initial-data/species_tree.nw', format=1)
orgsfile = pd.read_csv('data/initial-data/organisms.csv', header=None)
# Create a list name/variable and use list()
orgs = list(orgsfile[0])
organismslist = formatlist(orgs)
# Import alignment file as string
alignment_file = open(
'data/clustal-output/' + gene + '_Aligned/' + gene +
'_aligned_cds_nucl.fasta', 'r')
alignment_str = alignment_file.read()
alignment_file.close()
# Keep the branches in the species tree for species in the alignment file
# Some species may not be present in the alignment file
branches2keep = []
for organism in organismslist:
if organism in alignment_str:
#print('Yup.')
branches2keep.append(organism)
else:
pass
#print('Nope.') Make an error code in the log
# Input a list of branches to keep on the base tree
speciestree = t.prune(branches2keep, preserve_branch_length=True)
# Import the newick tree
tree = EvolTree(speciestree)
# Import the alignment
tree.link_to_alignment('data/clustal-output/' + gene + '_Aligned/' + gene +
'_aligned_cds_nucl.fasta')
tree.workdir = workdir
# Set the binpath of the codeml binary
tree.execpath = paml_path
# Run the codeml model
tree.run_model(model + '.' + gene)
def pamlSite(alnFile, treeFile, lModels, pamlParams, outDir, baseName, logger):
tree = EvolTree(treeFile)
os.mkdir(outDir + "paml_site/")
tree.workdir = outDir + "paml_site/"
tree.link_to_alignment(alnFile, "Fasta")
logger.info("PAML codeml")
dModelRun = {}
for model in lModels:
if model in ["M0", "M1", "M2", "M7", "M8"]:
logger.info("Running {:s}".format(model))
dModelRun[model] = tree.run_model(model)
if "M1" and "M2" in dModelRun:
p12 = tree.get_most_likely("M2", "M1")
logger.info("LRT of M1 vs M2 = {}".format(p12))
if "M7" and "M8" in dModelRun:
p78 = tree.get_most_likely("M8", "M7")
logger.info("LRT of M7 vs M8 = {}".format(p78))
"""
def main():
"""
main function
"""
tree = EvolTree(WRKDIR + 'tree.nw')
tree.workdir = 'data/protamine/PRM1/paml/'
random_swap(tree)
tree.link_to_evol_model(WRKDIR + 'paml/fb/fb.out', 'fb')
check_annotation(tree)
tree.link_to_evol_model(WRKDIR + 'paml/M1/M1.out', 'M1')
tree.link_to_evol_model(WRKDIR + 'paml/M2/M2.out', 'M2')
tree.link_to_evol_model(WRKDIR + 'paml/M7/M7.out', 'M7')
tree.link_to_evol_model(WRKDIR + 'paml/M8/M8.out', 'M8')
tree.link_to_alignment(WRKDIR + 'alignments.fasta_ali')
print 'pv of LRT M2 vs M1: ',
print tree.get_most_likely('M2', 'M1')
print 'pv of LRT M8 vs M7: ',
print tree.get_most_likely('M8', 'M7')
tree.show(histfaces=['M2'])
print 'The End.'
def run_codeml(mark_id, aln_file, tree_file, sleep):
logger.info('sub-process: {0}'.format(str(mark_id)))
time.sleep(round(sleep / args.threads, 2))
run_dir = os.path.join(output_dir, str(mark_id))
os.makedirs(run_dir)
tree = EvolTree(tree_file, format=0)
tree.link_to_alignment(aln_file)
tree.run_model('M0')
tree.workdir = run_dir
tree.mark_tree([mark_id], marks=['#1'])
tree.run_model('bsA.' + str(mark_id))
tree.run_model('bsA1.' + str(mark_id))
ps = tree.get_most_likely('bsA.' + str(mark_id), 'bsA1.' + str(mark_id))
rx = tree.get_most_likely('bsA1.' + str(mark_id), 'M0')
bsA = tree.get_evol_model('bsA.' + str(mark_id))
p_bsA = bsA.classes['proportions'][2]
wfrg2a = bsA.classes['foreground w'][2]
if ps < 0.05 and float(wfrg2a) > 1:
result = [mark_id, ps, rx, p_bsA, 'positive selection']
elif rx < 0.05 and ps >= 0.05:
result = [mark_id, ps, rx, p_bsA, 'relaxation']
else:
result = [mark_id, ps, rx, p_bsA, 'no signal']
return result
def main(args):
if args.BinPath:
tree = EvolTree(args.Tree, binpath=args.BinPath)
else:
tree = EvolTree(args.Tree)
if args.MSA[:-3] == ".phy":
tree.link_to_alignment(args.MSA, format="phylip")
elif args.MSA:
tree.link_to_alignment(args.MSA)
print(tree)
tree.workdir = os.getcwd()
if args.LoadedModels:
load_model(args.LoadedModels, tree)
compare_models(models=args.LoadedModels, tree=tree, args=args)
if args.Models:
run_models(args.models, tree)
if args.Compare:
if args.TreeStruct:
tree_structure = parse_structure_file(args.TreeStruct)
compare_models(models=args.Compare,
tree=tree,
tree_structure=tree_structure)
else:
compare_models(models=args.Compare, tree=tree, args=args)
def main():
"""
main function
"""
tree = EvolTree (WRKDIR + 'tree.nw')
tree.workdir = 'data/protamine/PRM1/paml/'
random_swap(tree)
tree.link_to_evol_model (WRKDIR + 'paml/fb/fb.out', 'fb')
check_annotation (tree)
tree.link_to_evol_model (WRKDIR + 'paml/M1/M1.out', 'M1')
tree.link_to_evol_model (WRKDIR + 'paml/M2/M2.out', 'M2')
tree.link_to_evol_model (WRKDIR + 'paml/M7/M7.out', 'M7')
tree.link_to_evol_model (WRKDIR + 'paml/M8/M8.out', 'M8')
tree.link_to_alignment (WRKDIR + 'alignments.fasta_ali')
print 'pv of LRT M2 vs M1: ',
print tree.get_most_likely ('M2','M1')
print 'pv of LRT M8 vs M7: ',
print tree.get_most_likely ('M8','M7')
tree.show (histfaces=['M2'])
print 'The End.'
15 Nov 2010
simple example to mark a tree and compute branch-site test of positive selection
"""
__author__ = "Francois-Jose Serra"
__email__ = "*****@*****.**"
__licence__ = "GPLv3"
__version__ = "0.0"
from ete3 import EvolTree
tree = EvolTree("data/S_example/measuring_S_tree.nw")
tree.link_to_alignment('data/S_example/alignment_S_measuring_evol.fasta')
print tree
raw_input('\n tree and alignment loaded\nHit some key, to start computation of branch site models A and A1 on each branch.\n')
print 'running model M0, for comparison with branch-site models...'
tree.run_model('M0')
# each node/leaf has two kind of identifiers node_id and paml_id, to mark nodes we have to specify
# the node_id of the nodes we want to mark, and the kind of mark in this way:
for leaf in tree:
leaf.node_id
print '\n---------\nNow working with leaf ' + leaf.name
tree.mark_tree([leaf.node_id], marks=['#1'])
np.array(all_dists),
2) # for the case that list is only 3 items long
closest_seq_ids = [seqid]
for d in idxes_of_3_smallest:
closest_seq_ids.append(seqids_of_other_species[d])
# ete3 has codeml handling implemented!! No need for own functions.
subtree = t.copy()
subtree.prune(closest_seq_ids, preserve_branch_length=True)
subtree.unroot()
evotree = EvolTree(subtree.write())
subfasta = make_clean_fasta(closest_seq_ids, seqdatadict)
if not subfasta:
omega_list.append("NA")
continue
else:
evotree.link_to_alignment(subfasta)
workdirname = './codeml_' + "__".join(closest_seq_ids)
evotree.workdir = workdirname
list_of_tempdirs.append(workdirname)
# mark the foreground branch
foreground_leafnode = evotree & seqid
# print (seqid)
# print(foreground_leafnode.node_id)
# print (evotree.write())
evotree.mark_tree([foreground_leafnode.node_id], ['#1'])
# print (evotree.write())
evotree.run_model('b_free.run')
b_free_fit = evotree.get_evol_model('b_free.run')
out_branches_dict = b_free_fit.branches
for b in out_branches_dict:
from ete3 import EvolTree
from string import ascii_letters
# CREATE TREE
fasta_lines = open("./whales.fasta", "r").readlines()
taxa = [l.replace('>', '').strip() for l in fasta_lines if l.startswith('>')]
taxa_map = { t: ascii_letters[i] for i, t in enumerate(taxa) }
taxa_string = '(' * (len(taxa) - 1) + '%s,%s)' % (ascii_letters[0], ascii_letters[1])
for t in ascii_letters[2:len(taxa)]:
taxa_string = taxa_string + ',%s)' % t
taxa_string = taxa_string + ';'
align = ''.join(fasta_lines)
for t in taxa:
align = align.replace(t, taxa_map[t])
tree = EvolTree(taxa_string)
tree.link_to_alignment(align)
#tree.link_to_evol_model("M2")
#tree.get_evol_model("M2")
print(tree.run_model.__doc__)
tree.run_model("fb")
tree = EvolTree(tree_file)
out_tree_name = os.path.basename(tree_file)
out_tree_name = os.path.splitext(out_tree_name)[0]
out_tree_name = out_tree_name + '_' + gene_name + '.tre'
# If there is a new alignment, prune the tree down to the taxa that remain in
# the new alignment and write a new tree because EvolTree is shit and can't
# use the pruned tree saved in memory
if empty_seq_count >= 1:
if len(taxa_in_alignment) >= 1:
tree.prune(taxa_in_alignment, preserve_branch_length=True)
tree.unroot()
tree.write(outfile=out_tree_name, format=0)
tree = EvolTree(out_tree_name)
tree.link_to_alignment(alignment_file)
tree.workdir = os.getcwd()
# Record list of all node_ids in the tree for later retrieving omega from a
# background branch in the b_free model
list_of_node_ids = []
for node in tree.traverse('postorder'):
list_of_node_ids.append(node.node_id)
test_taxa = []
with open(test_taxa_file, 'r') as test_taxa_list:
for taxon in test_taxa_list:
taxon = taxon.rstrip()
test_taxa.append(taxon)
'can be run from within a paml directory' from ete3 import EvolTree import os tree_file = "testTree.tre" alignment_file = "testAlignment.fasta" model = "./model/out" model_name = "bsD.bl_0.2w" # model_name = os.path.basename(os.getcwd()) testTree = EvolTree(tree_file) testTree.link_to_alignment(alignment_file) testTree.link_to_evol_model(model, model_name) testTree.show()
import sys, os, subprocess
import argparse
from ete3 import EvolTree
tree = EvolTree("tree.nw",
binpath="/home/edu/miniconda3/envs/ete3/bin/ete3_apps/bin")
tree.link_to_alignment("infile.phy", alg_format="phylip")
tree.workdir = os.getcwd()
print(tree)
print('running model M0, for comparison with branch-site models...')
tree.run_model('M0', keep=True)
#tree.link_to_evol_model("/home/edu/Desktop/Bioinformatica/Mitogenomics/Chondrichthyes/Phylogenetic_Tree","M0")
chimaeriformes = tree.get_common_ancestor("HM147138.1", "HM147135.1")
#chimaeriformes =tree.get_common_ancestor("Human_ECP","Goril_ECP")
for leaf in chimaeriformes:
tree.mark_tree([leaf.node_id], marks=["#1"])
#tree.run_model("bsA." + chimaeriformes)
#tree.mark_tree([leaf.node_id], marks = ["#1"])
print("Running")
print(tree.write())
tree.run_model('bsA.Chimaeriformes')
tree.run_model("bsA1.Chimaeriformes")
print('p-value of positive selection for sites on this branch is: ')
ps = tree.get_most_likely('bsA.Chimaeriformes', 'bsA1.Chimaeriformes')
print(str(ps))
rx = tree.get_most_likely('bsA1.Chimaeriformes', 'M0')