def split(tree_file, size, nexus=False, dir=None):
print file, size
if nexus:
tree = Nexus(tree_file).trees[0]
tree2 = Nexus(tree_file).trees[0]
else:
with open(tree_file) as handle:
tree_str = handle.read()
tree = Tree(tree_str)
tree2 = Tree(tree_str)
# with open(align_file) as handle:
# alignment = AlignIO.read(handle, "phylip")
splitter = TreeSplitter(tree,
max_size=size,
annotater=UnrootedShortestPath)
subs = list(splitter.subtrees())
runtime().debug("Found", len(subs), subs)
dir = dir if dir else os.path.dirname(tree_file)
for i, tree in enumerate(subs):
nodes = [tree.node(node) for node in tree.all_ids()]
taxa = set(
[node.data.taxon for node in nodes if node.data.taxon != None])
for terminal in tree2.get_terminals():
node = tree2.node(terminal)
if node.data.taxon in taxa:
node.data.taxon = "%i-" % i + node.data.taxon
# sub_taxa = tree.get_taxa()
# sub_alignment = Alignment(alphabet=alignment._alphabet)
# sub_alignment._records = [r for r in alignment._records if r.id in sub_taxa]
# assert len(sub_taxa)==len(sub_alignment._records)
## align_out = "%s.%i" % (os.path.join(dir,os.path.basename(align_file)),i)
# with open(align_out,"w") as handle:
# AlignIO.write([sub_alignment], handle, "phylip")
# from hpf.phylip import interleave
# interleave(align_out)
with open(
"%s.%i" % (os.path.join(dir, os.path.basename(tree_file)), i),
"w") as handle:
print >> handle, tree.to_string(plain_newick=True,
branchlengths_only=False) + ";"
with open("%s.annotated" % os.path.join(dir, os.path.basename(tree_file)),
"w") as handle:
print >> handle, tree2.to_string(plain_newick=True,
branchlengths_only=False) + ";"
trefname = dirbase + bpg[0:6] + '/' + bpg[0:9] + '/user/' + bpg[
0:9] + '.nj'
handle = open(trefname, 'r')
treestr = ''
for line in handle:
treestr += line.strip()
handle.close()
for oldid in oldid_newid:
treestr = treestr.replace('bpgseq%d' % oldid, oldid_newid[oldid],
1)
mytreeobj = Tree(tree=treestr, rooted=True)
### prune taxa we don't want ###
alltaxa = mytreeobj.get_taxa()
badtaxa = []
slowest_inparalogs = {}
for taxon in alltaxa:
if taxon not in oldid_newid.values():
badtaxa.append(taxon)
else:
sp = taxon.split('_bpgseq')[0]
if sp in slowest_inparalogs:
