def _stash_rewrite_and_call(self, fname, test_cases):
"""Safely run a series of tests on a parsed and rewritten file.
Specifically: Parse a file, rename the source file to a backup, rewrite
the file from the parsed object, check the rewritten file with the
given series of test functions, then restore the original by renaming
the backup copy.
Python 2.4 support: This would make more sense as a context manager
that simply handles renaming and finally restoring the original.
"""
phx = PhyloXMLIO.read(fname)
if os.path.exists(fname + "~"):
os.remove(fname + "~")
os.rename(fname, fname + "~")
try:
PhyloXMLIO.write(phx, fname)
for cls, tests in test_cases:
inst = cls("setUp")
for test in tests:
getattr(inst, test)()
finally:
# XXX not safe!
if os.path.exists(fname):
os.remove(fname)
os.rename(fname + "~", fname)
def _rewrite_and_call(self, orig_fname, test_cases):
"""Parse, rewrite and retest a phyloXML example file."""
infile = open(orig_fname, "rb")
phx = PhyloXMLIO.read(infile)
infile.close()
outfile = open(DUMMY, "w+b")
PhyloXMLIO.write(phx, outfile)
outfile.close()
for cls, tests in test_cases:
inst = cls("setUp")
for test in tests:
getattr(inst, test)()
def _rewrite_and_call(self, orig_fname, test_cases):
"""Parse, rewrite and retest a phyloXML example file."""
infile = open(orig_fname, 'r')
phx = PhyloXMLIO.read(infile)
infile.close()
outfile = open(DUMMY, 'w')
PhyloXMLIO.write(phx, outfile)
outfile.close()
for cls, tests in test_cases:
inst = cls('setUp')
for test in tests:
getattr(inst, test)()
def prepare_species_tree(FILE_TREE_IN,FILE_TREE_OUT):
clan_taxa = {}
treexml = PhyloXMLIO.read(open(FILE_TREE_IN, 'r'))
tree = treexml[0]
treexml.attributes.pop('schemaLocation', None) # not supported by Forester
tree.rooted = True
leaf_dict = {}
for node in tree.clade.find_clades():
if node.name:
tax_id = node.name
if tax_id.startswith('INT'):
tax_id = tax_id[3:]
taxon = PhyloXML.Taxonomy(id=PhyloXML.Id(tax_id, provider='ncbi_taxonomy'))
try:
taxon.scientific_name = find_tax_name(tax_id)
except KeyError:
taxon.scientific_name = '(NA)'
node._set_taxonomy(taxon)
node.name = None
else:
pass
PhyloXMLIO.write(treexml, FILE_TREE_OUT)
def prepare_species_tree(FILE_TREE_IN, FILE_TREE_OUT):
clan_taxa = {}
treexml = PhyloXMLIO.read(open(FILE_TREE_IN, 'r'))
tree = treexml[0]
treexml.attributes.pop('schemaLocation', None) # not supported by Forester
tree.rooted = True
leaf_dict = {}
for node in tree.clade.find_clades():
if node.name:
tax_id = node.name
if tax_id.startswith('INT'):
tax_id = tax_id[3:]
taxon = PhyloXML.Taxonomy(
id=PhyloXML.Id(tax_id, provider='ncbi_taxonomy'))
try:
taxon.scientific_name = find_tax_name(tax_id)
except KeyError:
taxon.scientific_name = '(NA)'
node._set_taxonomy(taxon)
node.name = None
else:
pass
PhyloXMLIO.write(treexml, FILE_TREE_OUT)
def reconcile_tree(gene_tree_file, reconciled_file, rec_tag, pfam_id, db):
if (os.path.isfile(rec_tag + 'ids.pickle')) and (pplacer_flag == 1):
id_information = pickle.load(open(rec_tag + 'ids.pickle', 'rb'))
existing_genes = id_information['existing_genes']
Sequnces = []
p_ids = []
new_genes = set([w['id'] for w in pplacer_queries[pfam_id]])
if not (new_genes - set(existing_genes)):
print "All %s Genes for family %s have already been placed in the reconciled tree." % (
len(new_genes), pfam_id)
print "Skip Reconciliation for %s" % pfam_id
return
txid_file = rec_tag + 'txid.xml'
if not (os.path.isfile(rec_tag + 'ids.pickle')) or not (
os.path.isfile(reconciled_file + '.gz')) or (pplacer_flag == 1):
print "Running Reconciliation for: %s" % pfam_id
rand_id = random.randint(1000000, 9999999)
subprocess.check_call("gunzip -c %s/%s.nw.gz > %s.%d" %
(tree_folder, pfam_id, gene_tree_file, rand_id),
shell=True)
tree = ete2.PhyloTree('%s.%d' % (gene_tree_file, rand_id), format=0)
tree.resolve_polytomy()
tree.write(format=0, outfile=txid_file + '.tmp.nw')
if os.path.exists('%s.%d' % (gene_tree_file, rand_id)):
subprocess.check_call("rm %s.%d" % (gene_tree_file, rand_id),
shell=True)
Phylo.convert(txid_file + '.tmp.nw', 'newick', txid_file + '.tmp.xml',
'phyloxml')
treexml = PhyloXMLIO.read(open(txid_file + '.tmp.xml', 'r'))
tree = treexml[0]
treexml.attributes.pop('schemaLocation',
None) # not supported by Forester
tree.rooted = True
my_ids = set([])
my_query_by_taxid = {}
for leaf in tree.clade.find_clades(terminal=True):
up_name = leaf.name.split('/')[0]
tax_id, tax_name = find_tax_id_unip(up_name, db)
if tax_id not in all_species_txids:
if tax_id in merged_taxid.keys():
tax_id = merged_taxid[tax_id]
tax_name = find_tax_name(tax_id, db)
if tax_id in best_taxid_map.keys():
tax_id = best_taxid_map[tax_id]
tax_name = find_tax_name(tax_id, db)
else:
tax_id0 = tax_id
tax_id, tax_name = find_best_taxid(tax_id, db)
if tax_id > 0:
best_taxid_map[tax_id0] = tax_id
if tax_id < 0:
if (-tax_id) in merged_taxid.keys():
tax_id = merged_taxid[-tax_id]
tax_name = find_tax_name(tax_id, db)
if tax_id in my_query_by_taxid:
my_query_by_taxid[tax_id].append(up_name)
else:
my_query_by_taxid[tax_id] = [up_name]
my_ids.add(tax_id)
my_tax_id = PhyloXML.Id(tax_id, provider='ncbi_taxonomy')
taxon = PhyloXML.Taxonomy(id=my_tax_id)
taxon.scientific_name = tax_name
leaf._set_taxonomy(taxon)
PhyloXMLIO.write(treexml, open(txid_file, 'w'))
os.system('rm ' + txid_file + '.tmp.nw')
os.system('rm ' + txid_file + '.tmp.xml')
print "Taxid file done for: %s" % pfam_id
existing_ids = list(set(my_ids) & set(all_species_txids))
existing_genes = [
g for txid in my_query_by_taxid.keys()
for g in my_query_by_taxid[txid] if txid in existing_ids
]
pickle.dump(
{
'pfam_id': pfam_id,
'existing_ids': existing_ids,
'existing_genes': existing_genes
}, open(rec_tag + 'ids.pickle', 'wb'))
print "Pickle file done for: %s" % pfam_id
if os.path.exists(reconciled_file):
os.system('rm ' + reconciled_file)
os.system(
"java -Xmx4g -cp %s/forester_1038.jar org.forester.application.gsdi -g %s %s/ncbi_2_fixed.xml %s"
% (lib_path, txid_file, species_tree_data_path, reconciled_file))
if os.path.exists(reconciled_file):
if os.path.exists(reconciled_file + '.gz'):
subprocess.check_call("rm %s.gz" % (reconciled_file), shell=True)
subprocess.check_call("gzip %s" % (reconciled_file), shell=True)
os.system('rm ' + rec_tag + 'reconciled_species_tree_used.xml')
os.system('rm ' + rec_tag + 'reconciled_gsdi_log.txt')
os.system('rm ' + txid_file)
print "Reconciliation file done for: %s" % pfam_id
def reconcile_tree(gene_tree_file,reconciled_file,rec_tag,pfam_id,db):
if (os.path.isfile(rec_tag+'ids.pickle')) and (pplacer_flag==1):
id_information = pickle.load(open(rec_tag+'ids.pickle', 'rb'))
existing_genes=id_information['existing_genes']
Sequnces=[]
p_ids=[]
new_genes=set([w['id'] for w in pplacer_queries[pfam_id]])
if not (new_genes-set(existing_genes)):
print "All %s Genes for family %s have already been placed in the reconciled tree."%(len(new_genes),pfam_id)
print "Skip Reconciliation for %s"%pfam_id
return
txid_file=rec_tag+'txid.xml'
if not(os.path.isfile(rec_tag+'ids.pickle')) or not(os.path.isfile(reconciled_file+'.gz')) or (pplacer_flag==1):
print "Running Reconciliation for: %s"%pfam_id
rand_id=random.randint(1000000,9999999)
subprocess.check_call("gunzip -c %s/%s.nw.gz > %s.%d"%(tree_folder,pfam_id,gene_tree_file,rand_id),shell=True)
tree = ete2.PhyloTree('%s.%d'%(gene_tree_file,rand_id), format=0)
tree.resolve_polytomy()
tree.write(format=0, outfile=txid_file+'.tmp.nw')
if os.path.exists('%s.%d'%(gene_tree_file,rand_id)):
subprocess.check_call("rm %s.%d"%(gene_tree_file,rand_id),shell=True)
Phylo.convert(txid_file+'.tmp.nw', 'newick', txid_file+'.tmp.xml', 'phyloxml')
treexml = PhyloXMLIO.read(open(txid_file+'.tmp.xml','r'))
tree = treexml[0]
treexml.attributes.pop('schemaLocation', None) # not supported by Forester
tree.rooted = True
my_ids=set([])
my_query_by_taxid={}
for leaf in tree.clade.find_clades(terminal=True):
up_name = leaf.name.split('/')[0]
tax_id,tax_name=find_tax_id_unip(up_name,db)
if tax_id not in all_species_txids:
if tax_id in merged_taxid.keys():
tax_id=merged_taxid[tax_id]
tax_name=find_tax_name(tax_id,db)
if tax_id in best_taxid_map.keys():
tax_id=best_taxid_map[tax_id]
tax_name=find_tax_name(tax_id,db)
else:
tax_id0=tax_id
tax_id,tax_name=find_best_taxid(tax_id,db)
if tax_id>0:
best_taxid_map[tax_id0]=tax_id
if tax_id<0:
if (-tax_id) in merged_taxid.keys():
tax_id=merged_taxid[-tax_id]
tax_name=find_tax_name(tax_id,db)
if tax_id in my_query_by_taxid:
my_query_by_taxid[tax_id].append(up_name)
else:
my_query_by_taxid[tax_id]=[up_name]
my_ids.add(tax_id)
my_tax_id = PhyloXML.Id(tax_id, provider='ncbi_taxonomy')
taxon=PhyloXML.Taxonomy(id=my_tax_id)
taxon.scientific_name = tax_name
leaf._set_taxonomy(taxon)
PhyloXMLIO.write(treexml, open(txid_file,'w'))
os.system('rm '+txid_file+'.tmp.nw')
os.system('rm '+txid_file+'.tmp.xml')
print "Taxid file done for: %s"%pfam_id
existing_ids=list(set(my_ids)&set(all_species_txids))
existing_genes=[g for txid in my_query_by_taxid.keys() for g in my_query_by_taxid[txid] if txid in existing_ids]
pickle.dump({'pfam_id':pfam_id,'existing_ids':existing_ids,'existing_genes':existing_genes}, open(rec_tag+'ids.pickle', 'wb'))
print "Pickle file done for: %s"%pfam_id
if os.path.exists(reconciled_file):
os.system('rm '+reconciled_file)
os.system("java -Xmx4g -cp %s/forester_1038.jar org.forester.application.gsdi -g %s %s/ncbi_2_fixed.xml %s"%(lib_path, txid_file, species_tree_data_path, reconciled_file))
if os.path.exists(reconciled_file):
if os.path.exists(reconciled_file+'.gz'):
subprocess.check_call("rm %s.gz"%(reconciled_file),shell=True)
subprocess.check_call("gzip %s"%(reconciled_file),shell=True)
os.system('rm '+rec_tag+'reconciled_species_tree_used.xml')
os.system('rm '+rec_tag+'reconciled_gsdi_log.txt')
os.system('rm '+txid_file)
print "Reconciliation file done for: %s"%pfam_id
print(aln)
# Calculate the distance matrix
# calculator = DistanceCalculator('identity')
# dm = calculator.get_distance(aln)
#
# with open("data/project/dist_mat.p", "wb") as f:
# pickle.dump(dm, f)
with open("data/project/dist_mat.p", "rb") as f:
dm = pickle.load(f)
# Construct the phylogenetic tree using UPGMA algorithm
constructor = DistanceTreeConstructor()
tree = constructor.upgma(dm)
tree_xml = tree.as_phyloxml()
PhyloXMLIO.write(tree_xml, "data/project/tree1.xml")
def labels(c):
if not c.is_terminal():
return ""
else:
return " ".join(c.name.split()[:2])
fig = plt.figure(figsize=(25, 25), dpi=100)
axes = fig.add_subplot(1, 1, 1)
Phylo.draw(tree, axes=axes, do_show=False, label_func=labels)
plt.savefig("plots/tree1.svg", format="svg")
Phylo.draw_ascii(tree)