def construct_tree(taxon, db, includespecies, taxalist=None):
conn = sqlite3.connect(db)
c = conn.cursor()
includelist = None
if taxalist != None:
tl = set()
tlf = open(taxalist, "r")
for i in tlf:
tl.add(i)
tlf.close()
includelist = get_all_included(tl, c)
species = []
stack = []
done = set()
rt = None
nodes = {} # id is key, value is node
if (taxon.isdigit()):
c.execute("select ncbi_id from taxonomy where ncbi_id = ?", (taxon, ))
else:
c.execute(
"select ncbi_id from taxonomy where name = ? and node_rank != 'species'",
(taxon, ))
for j in c:
stack.append(str(j[0]))
rt = node.Node()
rt.label = taxon + "_" + str(j[0])
rt.data["id"] = str(j[0])
nodes[str(j[0])] = rt
while len(stack) > 0:
id = stack.pop()
if id in done:
continue
done.add(id)
if includespecies:
c.execute(
"select ncbi_id,name,name_class,edited_name from taxonomy where parent_ncbi_id = ?",
(id, ))
else:
c.execute(
"select ncbi_id,name,name_class,edited_name from taxonomy where parent_ncbi_id = ? and node_rank != 'species'",
(id, ))
childs = []
for j in c:
tid = str(j[0])
if includelist != None and tid not in includelist:
continue
childs.append(tid)
stack.append(tid)
if str(j[2]) == "scientific name":
name = str(j[1])
edname = str(j[3])
nn = node.Node()
nn.label = clean_name(edname) + "_" + str(tid)
nn.data["id"] = tid
nodes[tid] = nn
nn.parent = nodes[id]
nodes[id].add_child(nn)
if len(childs) == 0 and id not in species:
species.append(id)
return rt
os.system(cmd)
return newalns
if __name__ == "__main__":
if len(sys.argv) != 2:
print "python " + sys.argv[0] + " startdir"
sys.exit(0)
cld = sys.argv[1]
#take off the trailing slash if there is one
if cld[-1] == "/":
cld = cld[0:-1]
count = 0
tree = node.Node()
nodes = {}
firstnode = True
#build a tree from the directory
for root, dirs, files in os.walk(cld, topdown=True):
if "clusters" in root:
continue
if "clusters" in dirs:
if firstnode == True:
tree.label = root.split("/")[-1]
firstnode = False
nodes[root.split("/")[-1]] = tree
nd = nodes[root.split("/")[-1]]
nd.data["dir"] = root
nd.data["names"] = set()
def construct_tree_only_ids(baseid, c, ids):
species = []
stack = []
done = set()
rt = None
includelist = get_all_included(ids, c)
# node_ids = {} # id is key, value is parent id
nodes = {} # id is key, value is node
stack.append(str(baseid))
rt = node.Node()
rt.label = baseid + "_" + str(baseid)
rt.data["id"] = str(baseid)
nodes[str(baseid)] = rt
while len(stack) > 0:
id = stack.pop()
if id in done:
continue
done.add(id)
c.execute(
"select ncbi_id,name,name_class,edited_name from taxonomy where parent_ncbi_id = ?",
(id, ))
childs = []
for j in c:
tid = str(j[0])
if includelist != None and tid not in includelist:
continue
childs.append(tid)
stack.append(tid)
if str(j[2]) == "scientific name":
name = str(j[1])
edname = str(j[3])
nn = node.Node()
nn.label = clean_name(name) #+"_"+str(tid)
nn.data["id"] = tid
nodes[tid] = nn
# node_ids[tid] = id
nn.parent = nodes[id]
nodes[id].add_child(nn)
if len(childs) == 0 and id not in species:
species.append(id)
for i in rt.iternodes():
if len(i.children) == 0:
continue
else:
i.label = ""
going = True
while going:
found = False
for i in rt.iternodes():
if i.parent != None and len(i.children) == 1 and i.label == "":
par = i.parent
ch = i.children[0]
par.remove_child(i)
par.add_child(ch)
found = True
break
if found == False:
going = False
break
return rt
sys.stderr.write("here\n")
while len(diffnms) > 0:
for j in diffnms:
going = True
cn = diffnds[j]
while going:
par = cn.parent
pln = set(par.lvsnms()).intersection(rootnms)
if len(pln) > 0:
amrca = tree_utils.get_mrca_wnms(pln,tree1)
#if VERBOSE:
# sys.stderr.write("add at this node"+" "+par.get_newick_repr(False)+" "+amrca.get_newick_repr(False)+"\n")
if len(pln) == 1:
amrca = tree1.get_leaf_by_name(list(pln)[0])
#print "f",amrca.get_newick_repr(True)
nn = node.Node()
if EDITLEN:
nn.length = amrca.length/2.
nn.height = amrca.height+amrca.length/2.
amrca.length = nn.length
amrca.parent.add_child(nn)
amrca.parent.remove_child(amrca)
nn.add_child(amrca)
amrca = nn
for k in par.children:
if len(set(k.lvsnms()).intersection(rootnms)) > 0:
continue
else:
#tree_utils.set_heights(amrca)
#print "a",k.get_newick_repr(True),amrca.length,amrca.get_newick_repr(True),amrca.height
if EDITLEN:
def construct_tree_only_ids(baseid, c, ids):
species = []
stack = []
done = set()
rt = None
includelist = get_all_included(ids, c)
# node_ids = {} # id is key, value is parent id
nodes = {} # id is key, value is node
stack.append(str(baseid))
rt = node.Node()
rt.label = baseid + "_" + str(baseid)
rt.data["id"] = str(baseid)
nodes[str(baseid)] = rt
while len(stack) > 0:
id = stack.pop()
if id in done:
continue
done.add(id)
c.execute(
"select ncbi_id,name_class,name,node_rank from taxonomy where parent_ncbi_id = ?",
(id, ))
for j in c:
tid = str(j[0])
if includelist != None and tid not in includelist:
continue
stack.append(tid)
if str(j[1]) == "scientific name" and (noinclude == False
or str(j[3]) != stopat):
nn = node.Node()
nn.label = str(tid)
nn.data["id"] = tid
nn.data["rank"] = str(j[3])
nodes[tid] = nn
# node_ids[tid] = id
if id in ids or "incertae" in str(
j[2]) or "unidentified" in str(
j[2]) or "unplaced" in str(j[2]):
#should lose these and not constrain
nodes[tid] = nodes[id]
#nn.parent = nodes[id].parent
#nodes[id].parent.add_child(nn)
else:
nn.parent = nodes[id]
nodes[id].add_child(nn)
elif (noinclude == True and str(j[3]) == stopat) and str(
j[1]) == "scientific name":
if id in ids:
nodes[tid] = nodes[id].parent
else:
nodes[tid] = nodes[id]
if useonly: #remove any constraint that insn't in the list
toremove = set()
for i in rt.iternodes():
if len(i.children) == 0 or i == rt:
continue
else:
if i.data["rank"] not in useonlylist:
toremove.add(i)
for i in toremove:
p = i.parent
if p == None:
continue
p.remove_child(i)
i.parent = None
for j in i.children:
p.add_child(j)
j.parent = p
for i in rt.iternodes():
if len(i.children) == 0:
continue
else:
i.label = ""
going = True
while going:
found = False
for i in rt.iternodes():
if i.parent != None and len(i.children) == 1 and i.label == "":
par = i.parent
ch = i.children[0]
par.remove_child(i)
par.add_child(ch)
found = True
break
if found == False:
going = False
break
if len(rt.children) == 1:
rt = rt.children[0]
return rt