if not n.is_leaf():
n.name = n.S
else:
sp_dict[n.name] = n.S
logger.debug("sp_dict")
logger.debug(sp_dict)
# Let's reconcile our genetree with the species tree
recon_tree, events = genetree.reconcile(sptree)
# a new "reconcilied tree" is returned. As well as the list of
# inferred events.
ntrees, ndups, sptrees = genetree.get_speciation_trees()
logger.debug( "Found %d species trees and %d duplication nodes", ntrees, ndups)
HomologySummary=["DUPLICATIONS : " + str(ndups) ]
logger.debug( "Orthology and Paralogy relationships:")
for ev in events:
if ev.etype == "S":
logger.debug("".join(['ORTHOLOGY RELATIONSHIP:', ','.join(ev.inparalogs), " <===> ", ','.join(ev.orthologs)]))
HomologySummary.append("".join(['ORTHOLOGY RELATIONSHIP: ', ', '.join(ev.inparalogs), " <===> ", ','.join(ev.orthologs)]))
elif ev.etype == "D":
logger.debug("".join(['PARALOGY RELATIONSHIP: ', ', '.join(ev.inparalogs), " <===>" , ','.join(ev.outparalogs)]))
HomologySummary.append("".join(['PARALOGY RELATIONSHIP: ', ', '.join(ev.inparalogs), " <===> ", ','.join(ev.outparalogs)]))
HomologyFile = OutPrefixName + ".orthologs.txt"
with open(HomologyFile,"w") as File:
for ev in evev:
if ev.etype == "S":
print(ev.orthologs)
# find evolutionary events
evev = phy.get_descendant_evol_events(sos_thr=0.9)
# all events
for ev in evev:
print(ev.etype, ','.join(ev.in_seqs), "<====>", ','.join(ev.out_seqs))
# all events involving either Hsap or Drer
fseqs = lambda slist: [
s for s in slist if s.startswith("Drer") or s.startswith("Hsap")
]
for ev in evev:
if ev.etype == "D":
print('Paralog: ', ','.join(fseqs(ev.in_seqs)), "<====>",
','.join(fseqs(ev.out_seqs)))
for ev in evev:
if ev.etype == "S":
print('Ortholog:', ','.join(fseqs(ev.in_seqs)), "<====>",
','.join(fseqs(ev.out_seqs)))
# obtain duplication events
ntrees, ndups, sptrees = phy.get_speciation_trees()
print("Found %d species trees and %d duplication nodes" % (ntrees, ndups))
for spt in sptrees:
print(spt)
def main():
usage = "usage: %prog --genetree <genetree-file> --speciestree <speciestree-file> [options]"
parser = optparse.OptionParser(usage=usage)
parser.add_option('--genetree', help='GeneTree in nhx format')
parser.add_option('--speciestree', help='Species Tree in nhx format')
parser.add_option('--ingroup', help='Species Tree in nhx format')
parser.add_option('--outgroup', help='Species Tree in nhx format')
parser.add_option('--species_format',
type='int',
default=8,
help='Species Tree input format (0-9)')
parser.add_option('--gene_node',
type='int',
default=0,
help='Gene node format 0=gene_species, 1=species_gene')
parser.add_option('--gainlose',
action='store_true',
default=False,
help='Find out gene gain/lose')
parser.add_option('--split',
type='choice',
choices=['dups', 'treeko', 'species'],
dest="split",
default='dups',
help='Choose GeneTree splitting algorithms')
parser.add_option('--output_format',
type='int',
default=9,
help='GeneTree output format (0-9)')
parser.add_option(
'-d',
'--dir',
type='string',
default="",
help=
"Absolute or relative path to output directory. If directory does not exist it will be created"
)
options, args = parser.parse_args()
if options.dir and not os.path.exists(options.dir):
os.makedirs(options.dir)
if options.genetree is None:
parser.error(
"--genetree option must be specified, GeneTree in nhx format")
if os.stat(options.genetree).st_size == 0:
sys.exit()
with open(options.genetree, 'r') as f:
contents = f.read()
# Remove empty NHX features that can be produced by TreeBest but break ete3
contents = contents.replace('[&&NHX]', '')
# reads single gene tree
genetree = PhyloTree(contents)
# sets species naming function
if options.gene_node == 0:
genetree.set_species_naming_function(parse_sp_name)
# reconcile species tree with gene tree to help find out gene gain/lose
if options.gainlose:
if options.speciestree is None:
parser.error(
"--speciestree option must be specified, species tree in nhx format"
)
# reads species tree
speciestree = PhyloTree(options.speciestree,
format=options.species_format)
# Removes '*' from Species names comes from Species tree configrured for TreeBest
for leaf in speciestree:
leaf.name = leaf.name.strip('*')
genetree, events = genetree.reconcile(speciestree)
if options.split == "dups":
# splits tree by duplication events which returns the list of all subtrees resulting from splitting current tree by its duplication nodes.
for cluster_id, node in enumerate(genetree.split_by_dups(), start=1):
outfile = '{}_genetree.nhx'.format(cluster_id)
if options.dir:
outfile = os.path.join(options.dir, outfile)
with open(outfile, 'w') as f:
f.write(node.write(format=options.output_format))
elif options.split == "treeko":
# splits tree using the TreeKO algorithm.
ntrees, ndups, sptrees = genetree.get_speciation_trees()
for cluster_id, spt in enumerate(sptrees, start=1):
outfile = '{}_genetree.nhx'.format(cluster_id)
if options.dir:
outfile = os.path.join(options.dir, outfile)
with open(outfile, 'w') as f:
f.write(spt.write(format=options.output_format))
elif options.split == "species":
ingroup = options.ingroup.split(",")
outgroup = options.outgroup.split(",")
cluster_id = 0
def split_tree_by_species(tree, ingroup, outgroup):
nonlocal cluster_id
if len(outgroup) > 0:
outgroup_bool = check_outgroup(tree, outgroup)
else:
outgroup_bool = True
if outgroup_bool and check_ingroup(tree, ingroup):
child1, child2 = tree.children
split_tree_by_species(child1, ingroup, outgroup)
split_tree_by_species(child2, ingroup, outgroup)
else:
cluster_id += 1
outfile = '{}_genetree.nhx'.format(cluster_id)
if options.dir:
outfile = os.path.join(options.dir, outfile)
with open(outfile, 'w') as f:
f.write(tree.write(format=options.output_format))
split_tree_by_species(genetree, ingroup, outgroup)
