def get_example_tree():
# Random tree
t = Tree()
t.populate(20, random_branches=True)
# Some random features in all nodes
for n in t.traverse():
n.add_features(weight=random.randint(0, 50))
# Create an empty TreeStyle
ts = TreeStyle()
# Set our custom layout function
ts.layout_fn = layout
# Draw a tree
ts.mode = "c"
# We will add node names manually
ts.show_leaf_name = False
# Show branch data
ts.show_branch_length = True
ts.show_branch_support = True
return t, ts
def get_example_tree():
# Set dashed blue lines in all leaves
nst1 = NodeStyle()
nst1["bgcolor"] = "LightSteelBlue"
nst2 = NodeStyle()
nst2["bgcolor"] = "Moccasin"
nst3 = NodeStyle()
nst3["bgcolor"] = "DarkSeaGreen"
nst4 = NodeStyle()
nst4["bgcolor"] = "Khaki"
t = Tree("((((a1,a2),a3), ((b1,b2),(b3,b4))), ((c1,c2),c3));")
for n in t.traverse():
n.dist = 0
n1 = t.get_common_ancestor("a1", "a2", "a3")
n1.set_style(nst1)
n2 = t.get_common_ancestor("b1", "b2", "b3", "b4")
n2.set_style(nst2)
n3 = t.get_common_ancestor("c1", "c2", "c3")
n3.set_style(nst3)
n4 = t.get_common_ancestor("b3", "b4")
n4.set_style(nst4)
ts = TreeStyle()
ts.layout_fn = layout
ts.show_leaf_name = False
ts.mode = "c"
ts.root_opening_factor = 1
return t, ts
def get_example_tree():
# Random tree
t = Tree()
t.populate(20, random_branches=True)
# Some random features in all nodes
for n in t.traverse():
n.add_features(weight=random.randint(0, 50))
# Create an empty TreeStyle
ts = TreeStyle()
# Set our custom layout function
ts.layout_fn = layout
# Draw a tree
ts.mode = "c"
# We will add node names manually
ts.show_leaf_name = False
# Show branch data
ts.show_branch_length = True
ts.show_branch_support = True
return t, ts
def get_example_tree():
# Set dashed blue lines in all leaves
nst1 = NodeStyle()
nst1["bgcolor"] = "LightSteelBlue"
nst2 = NodeStyle()
nst2["bgcolor"] = "Moccasin"
nst3 = NodeStyle()
nst3["bgcolor"] = "DarkSeaGreen"
nst4 = NodeStyle()
nst4["bgcolor"] = "Khaki"
t = Tree("((((a1,a2),a3), ((b1,b2),(b3,b4))), ((c1,c2),c3));")
for n in t.traverse():
n.dist = 0
n1 = t.get_common_ancestor("a1", "a2", "a3")
n1.set_style(nst1)
n2 = t.get_common_ancestor("b1", "b2", "b3", "b4")
n2.set_style(nst2)
n3 = t.get_common_ancestor("c1", "c2", "c3")
n3.set_style(nst3)
n4 = t.get_common_ancestor("b3", "b4")
n4.set_style(nst4)
ts = TreeStyle()
ts.layout_fn = layout
ts.show_leaf_name = False
ts.mode = "c"
ts.root_opening_factor = 1
return t, ts
def get_example_tree():
t = Tree()
ts = TreeStyle()
ts.layout_fn = layout
ts.mode = "r"
ts.show_leaf_name = False
t.populate(10)
return t, ts
def get_example_tree():
t = Tree()
ts = TreeStyle()
ts.layout_fn = layout
ts.mode = "r"
ts.show_leaf_name = False
t.populate(10)
return t, ts
def get_example_tree():
t = Tree()
ts = TreeStyle()
ts.layout_fn = layout
ts.mode = "c"
ts.show_leaf_name = True
ts.min_leaf_separation = 15
t.populate(100)
return t, ts
def get_example_tree():
t = Tree()
ts = TreeStyle()
ts.layout_fn = layout
ts.mode = "c"
ts.show_leaf_name = True
ts.min_leaf_separation = 15
t.populate(100)
return t, ts
def render_tree(tree, fname):
# Generates tree snapshot
npr_nodestyle = NodeStyle()
npr_nodestyle["fgcolor"] = "red"
for n in tree.traverse():
if hasattr(n, "nodeid"):
n.set_style(npr_nodestyle)
ts = TreeStyle()
ts.show_leaf_name = True
ts.show_branch_length = True
ts.show_branch_support = True
ts.mode = "r"
iterface = faces.TextFace("iter")
ts.legend.add_face(iterface, 0)
tree.dist = 0
tree.sort_descendants()
tree.render(fname, tree_style=ts, w=700)
t.add_face(fixed, column=1, position="branch-right") # Bind the precomputed style to the root node # ETE 2.1 has now support for general image properties I = TreeStyle() # You can add faces to the tree image (without any node # associated). They will be used as headers and foot notes of the # aligned columns (aligned faces) I.aligned_header.add_face(t1, column = 0) I.aligned_header.add_face(t1, 1) I.aligned_header.add_face(t1, 2) I.aligned_header.add_face(t1, 3) t1.hz_align = 1 # 0 left, 1 center, 2 right t1.border.width = 1 I.aligned_foot.add_face(t2, column = 0) I.aligned_foot.add_face(t2, 1) I.aligned_foot.add_face(t2, 2) I.aligned_foot.add_face(t2, 3) t2.hz_align = 1 # Set tree image style. Note that aligned header and foot is only # visible in "rect" mode. I.mode = "r" #(rectangular) I.tree_width= 50 t.show(mylayout, tree_style=I)
t, ts = Tree(), TreeStyle()
t.populate(5)
for n in t.traverse():
n.dist = 0
temp_tface = TreeFace(t, ts)
n = main_tree.add_child()
n.add_face(temp_tface, 0, "aligned")
ts.optimal_scale_level = "full"
temp_tface = TreeFace(t, ts)
n = main_tree.add_child()
n.add_face(temp_tface, 0, "aligned")
ts = TreeStyle()
t.populate(5)
ts.mode = "c"
temp_tface = TreeFace(t, ts)
n = main_tree.add_child()
n.add_face(temp_tface, 0, "aligned")
ts.optimal_scale_level = "full"
temp_tface = TreeFace(t, ts)
n = main_tree.add_child()
n.add_face(temp_tface, 0, "aligned")
t, ts = Tree(), TreeStyle()
temp_tface = TreeFace(Tree(), ts)
n = main_tree.add_child()
n.add_face(temp_tface, 0, "aligned")
t, ts = Tree(), TreeStyle()
def ncbi_consensus(self, ):
nsubtrees, ndups, subtrees = self.get_speciation_trees(map_features=["taxid"])
valid_subtrees, broken_subtrees, ncbi_mistakes, broken_branches, total_rf, broken_clades, broken_sizes = analyze_subtrees(t, subtrees, show_tree=SHOW_TREE)
avg_rf = []
rf_max = 0.0 # reft.robinson_foulds(reft)[1]
sum_size = 0.0
#reftree =
for tn, subt in enumerate(subtrees):
partial_rf = subt.robinson_foulds(reft, attr_t1="taxid")
sptree_size = len(set([n.taxid for n in subt.iter_leaves()]))
sum_size += sptree_size
avg_rf.append((partial_rf[0]/float(partial_rf[1])) * sptree_size)
common_names = len(partial_rf[3])
max_size = max(max_size, sptree_size)
rf_max = max(rf_max, partial_rf[1])
rf = numpy.sum(avg_rf) / float(sum_size) # Treeko dist
rf_std = numpy.std(avg_rf)
rf_med = numpy.median(avg_rf)
sizes_info = "%0.1f/%0.1f +- %0.1f" %( numpy.mean(broken_sizes), numpy.median(broken_sizes), numpy.std(broken_sizes))
iter_values = [os.path.basename(tfile), nsubtrees, ndups,
broken_subtrees, ncbi_mistakes, broken_branches, sizes_info, rf, rf_med,
rf_std, rf_max, common_names]
print >>OUT, '|'.join(map(lambda x: str(x).strip().ljust(15), iter_values))
fixed = sorted([n for n in prev_broken if n not in broken_clades])
new_problems = sorted(broken_clades - prev_broken)
fixed_string = color(', '.join(fixed), "green") if fixed else ""
problems_string = color(', '.join(new_problems), "red") if new_problems else ""
OUT.write(" Fixed clades: %s\n" %fixed_string) if fixed else None
OUT.write(" New broken: %s\n" %problems_string) if new_problems else None
prev_broken = broken_clades
ENTRIES.append([os.path.basename(tfile), nsubtrees, ndups,
broken_subtrees, ncbi_mistakes, broken_branches, sizes_info, fixed_string, problems_string])
OUT.flush()
if args.show_tree or args.render:
ts = TreeStyle()
ts.force_topology = True
#ts.tree_width = 500
ts.show_leaf_name = False
ts.layout_fn = ncbi_layout
ts.mode = "r"
t.dist = 0
if args.show_tree:
#if args.hide_monophyletic:
# tax2monophyletic = {}
# n2content = t.get_node2content()
# for node in t.traverse():
# term2count = defaultdict(int)
# for leaf in n2content[node]:
# if leaf.lineage:
# for term in leaf.lineage:
# term2count[term] += 1
# expected_size = len(n2content)
# for term, count in term2count.iteritems():
# if count > 1
print "Showing tree..."
t.show(tree_style=ts)
else:
t.render("img.svg", tree_style=ts, dpi=300)
print "dumping color config"
cPickle.dump(name2color, open("ncbi_colors.pkl", "w"))
if args.dump:
cPickle.dump(t, open("ncbi_analysis.pkl", "w"))
def main(argv):
parser = argparse.ArgumentParser(
description=__DESCRIPTION__,
formatter_class=argparse.RawDescriptionHelpFormatter)
# name or flags - Either a name or a list of option strings, e.g. foo or -f, --foo.
# action - The basic type of action to be taken when this argument is encountered at the command line. (store, store_const, store_true, store_false, append, append_const, version)
# nargs - The number of command-line arguments that should be consumed. (N, ? (one or default), * (all 1 or more), + (more than 1) )
# const - A constant value required by some action and nargs selections.
# default - The value produced if the argument is absent from the command line.
# type - The type to which the command-line argument should be converted.
# choices - A container of the allowable values for the argument.
# required - Whether or not the command-line option may be omitted (optionals only).
# help - A brief description of what the argument does.
# metavar - A name for the argument in usage messages.
# dest - The name of the attribute to be added to the object returned by parse_args().
parser.add_argument("--show",
dest="show_tree",
action="store_true",
help="""Display tree after the analysis.""")
parser.add_argument("--render",
dest="render",
action="store_true",
help="""Render tree.""")
parser.add_argument("--dump",
dest="dump",
action="store_true",
help="""Dump analysis""")
parser.add_argument(
"--explore",
dest="explore",
type=str,
help="""Reads a previously analyzed tree and visualize it""")
input_args = parser.add_mutually_exclusive_group()
input_args.required = True
input_args.add_argument("-t",
"--tree",
dest="target_tree",
nargs="+",
type=str,
help="""Tree file in newick format""")
input_args.add_argument("-tf",
dest="tree_list_file",
type=str,
help="File with the list of tree files")
parser.add_argument("--tax",
dest="tax_info",
type=str,
help="If the taxid attribute is not set in the"
" newick file for all leaf nodes, a tab file file"
" with the translation of name and taxid can be"
" provided with this option.")
parser.add_argument(
"--sp_delimiter",
dest="sp_delimiter",
type=str,
help=
"If taxid is part of the leaf name, delimiter used to split the string"
)
parser.add_argument(
"--sp_field",
dest="sp_field",
type=int,
default=0,
help="field position for taxid after splitting leaf names")
parser.add_argument("--ref",
dest="ref_tree",
type=str,
help="Uses ref tree to compute robinson foulds"
" distances of the different subtrees")
parser.add_argument("--rf-only",
dest="rf_only",
action="store_true",
help="Skip ncbi consensus analysis")
parser.add_argument(
"--outgroup",
dest="outgroup",
type=str,
nargs="+",
help="A list of node names defining the trees outgroup")
parser.add_argument("--is_sptree",
dest="is_sptree",
action="store_true",
help="Assumes no duplication nodes in the tree")
parser.add_argument("-o",
dest="output",
type=str,
help="Writes result into a file")
parser.add_argument("--tax2name", dest="tax2name", type=str, help="")
parser.add_argument("--tax2track", dest="tax2track", type=str, help="")
parser.add_argument("--dump_tax_info",
dest="dump_tax_info",
action="store_true",
help="")
args = parser.parse_args(argv)
if args.sp_delimiter:
GET_TAXID = lambda x: x.split(args.sp_delimiter)[args.sp_field]
else:
GET_TAXID = None
reftree_name = os.path.basename(args.ref_tree) if args.ref_tree else ""
if args.explore:
print >> sys.stderr, "Reading tree from file:", args.explore
t = cPickle.load(open(args.explore))
ts = TreeStyle()
ts.force_topology = True
ts.show_leaf_name = False
ts.layout_fn = ncbi_layout
ts.mode = "r"
t.show(tree_style=ts)
print >> sys.stderr, "dumping color config"
cPickle.dump(name2color, open("ncbi_colors.pkl", "w"))
sys.exit()
if args.output:
OUT = open(args.output, "w")
else:
OUT = sys.stdout
print >> sys.stderr, "Dumping results into", OUT
target_trees = []
if args.tree_list_file:
target_trees = [line.strip() for line in open(args.tree_list_file)]
if args.target_tree:
target_trees += args.target_tree
prev_tree = None
if args.tax2name:
tax2name = cPickle.load(open(args.tax2name))
else:
tax2name = {}
if args.tax2track:
tax2track = cPickle.load(open(args.tax2track))
else:
tax2track = {}
print len(tax2track), len(tax2name)
header = ("TargetTree", "Subtrees", "Ndups", "Broken subtrees",
"Broken clades", "Clade sizes", "RF (avg)", "RF (med)",
"RF (std)", "RF (max)", "Shared tips")
print >> OUT, '|'.join([h.ljust(15) for h in header])
if args.ref_tree:
print >> sys.stderr, "Reading ref tree from", args.ref_tree
reft = Tree(args.ref_tree, format=1)
else:
reft = None
SHOW_TREE = False
if args.show_tree or args.render:
SHOW_TREE = True
prev_broken = set()
ENTRIES = []
ncbi.connect_database()
for tfile in target_trees:
#print tfile
t = PhyloTree(tfile, sp_naming_function=None)
if GET_TAXID:
for n in t.iter_leaves():
n.name = GET_TAXID(n.name)
if args.outgroup:
if len(args.outgroup) == 1:
out = t & args.outgroup[0]
else:
out = t.get_common_ancestor(args.outgroup)
if set(out.get_leaf_names()) ^ set(args.outgroup):
raise ValueError("Outgroup is not monophyletic")
t.set_outgroup(out)
t.ladderize()
if prev_tree:
tree_compare(t, prev_tree)
prev_tree = t
if args.tax_info:
tax2name, tax2track = annotate_tree_with_taxa(
t, args.tax_info, tax2name, tax2track)
if args.dump_tax_info:
cPickle.dump(tax2track, open("tax2track.pkl", "w"))
cPickle.dump(tax2name, open("tax2name.pkl", "w"))
print "Tax info written into pickle files"
else:
for n in t.iter_leaves():
spcode = n.name
n.add_features(taxid=spcode)
n.add_features(species=spcode)
tax2name, tax2track = annotate_tree_with_taxa(
t, None, tax2name, tax2track)
# Split tree into species trees
#subtrees = t.get_speciation_trees()
if not args.rf_only:
#print "Calculating tree subparts..."
t1 = time.time()
if not args.is_sptree:
subtrees = t.split_by_dups()
#print "Subparts:", len(subtrees), time.time()-t1
else:
subtrees = [t]
valid_subtrees, broken_subtrees, ncbi_mistakes, broken_branches, total_rf, broken_clades, broken_sizes = analyze_subtrees(
t, subtrees, show_tree=SHOW_TREE)
#print valid_subtrees, broken_subtrees, ncbi_mistakes, total_rf
else:
subtrees = []
valid_subtrees, broken_subtrees, ncbi_mistakes, broken_branches, total_rf, broken_clades, broken_sizes = 0, 0, 0, 0, 0, 0
ndups = 0
nsubtrees = len(subtrees)
rf = 0
rf_max = 0
rf_std = 0
rf_med = 0
common_names = 0
max_size = 0
if reft and len(subtrees) == 1:
rf = t.robinson_foulds(reft, attr_t1="realname")
rf_max = rf[1]
rf = rf[0]
rf_med = rf
elif reft:
#print "Calculating avg RF..."
nsubtrees, ndups, subtrees = t.get_speciation_trees(
map_features=["taxid"])
#print len(subtrees), "Sub-Species-trees found"
avg_rf = []
rf_max = 0.0 # reft.robinson_foulds(reft)[1]
sum_size = 0.0
print nsubtrees, "subtrees", ndups, "duplications"
for ii, subt in enumerate(subtrees):
print "\r%d" % ii,
sys.stdout.flush()
try:
partial_rf = subt.robinson_foulds(reft, attr_t1="taxid")
except ValueError:
pass
else:
sptree_size = len(
set([n.taxid for n in subt.iter_leaves()]))
sum_size += sptree_size
avg_rf.append(
(partial_rf[0] / float(partial_rf[1])) * sptree_size)
common_names = len(partial_rf[3])
max_size = max(max_size, sptree_size)
rf_max = max(rf_max, partial_rf[1])
#print partial_rf[:2]
rf = numpy.sum(avg_rf) / float(sum_size) # Treeko dist
rf_std = numpy.std(avg_rf)
rf_med = numpy.median(avg_rf)
sizes_info = "%0.1f/%0.1f +- %0.1f" % (numpy.mean(broken_sizes),
numpy.median(broken_sizes),
numpy.std(broken_sizes))
iter_values = [
os.path.basename(tfile), nsubtrees, ndups, broken_subtrees,
ncbi_mistakes, broken_branches, sizes_info, rf, rf_med, rf_std,
rf_max, common_names
]
print >> OUT, '|'.join(
map(lambda x: str(x).strip().ljust(15), iter_values))
fixed = sorted([n for n in prev_broken if n not in broken_clades])
new_problems = sorted(broken_clades - prev_broken)
fixed_string = color(', '.join(fixed), "green") if fixed else ""
problems_string = color(', '.join(new_problems),
"red") if new_problems else ""
OUT.write(" Fixed clades: %s\n" % fixed_string) if fixed else None
OUT.write(" New broken: %s\n" %
problems_string) if new_problems else None
prev_broken = broken_clades
ENTRIES.append([
os.path.basename(tfile), nsubtrees, ndups, broken_subtrees,
ncbi_mistakes, broken_branches, sizes_info, fixed_string,
problems_string
])
OUT.flush()
if args.show_tree or args.render:
ts = TreeStyle()
ts.force_topology = True
#ts.tree_width = 500
ts.show_leaf_name = False
ts.layout_fn = ncbi_layout
ts.mode = "r"
t.dist = 0
if args.show_tree:
#if args.hide_monophyletic:
# tax2monophyletic = {}
# n2content = t.get_node2content()
# for node in t.traverse():
# term2count = defaultdict(int)
# for leaf in n2content[node]:
# if leaf.lineage:
# for term in leaf.lineage:
# term2count[term] += 1
# expected_size = len(n2content)
# for term, count in term2count.iteritems():
# if count > 1
print "Showing tree..."
t.show(tree_style=ts)
else:
t.render("img.svg", tree_style=ts, dpi=300)
print "dumping color config"
cPickle.dump(name2color, open("ncbi_colors.pkl", "w"))
if args.dump:
cPickle.dump(t, open("ncbi_analysis.pkl", "w"))
print
print
HEADER = ("TargetTree", "Subtrees", "Ndups", "Broken subtrees",
"Broken clades", "Broken branches", "Clade sizes",
"Fixed Groups", "New Broken Clades")
print_table(ENTRIES, max_col_width=50, row_line=True, header=HEADER)
if args.output:
OUT.close()
def get_example_tree():
t = Tree()
t.populate(10)
# Margins, alignment, border, background and opacity can now be set for any face
rs1 = faces.TextFace("branch-right\nmargins&borders",
fsize=12,
fgcolor="#009000")
rs1.margin_top = 10
rs1.margin_bottom = 50
rs1.margin_left = 40
rs1.margin_right = 40
rs1.border.width = 1
rs1.background.color = "lightgreen"
rs1.inner_border.width = 0
rs1.inner_border.line_style = 1
rs1.inner_border.color = "red"
rs1.opacity = 0.6
rs1.hz_align = 2 # 0 left, 1 center, 2 right
rs1.vt_align = 1 # 0 left, 1 center, 2 right
br1 = faces.TextFace("branch-right1", fsize=12, fgcolor="#009000")
br2 = faces.TextFace("branch-right3", fsize=12, fgcolor="#009000")
# New face positions (branch-top and branch-bottom)
bb = faces.TextFace("branch-bottom 1", fsize=8, fgcolor="#909000")
bb2 = faces.TextFace("branch-bottom 2", fsize=8, fgcolor="#909000")
bt = faces.TextFace("branch-top 1", fsize=6, fgcolor="#099000")
# And faces can also be used as headers or foot notes of aligned
# columns
t1 = faces.TextFace("Header Face", fsize=12, fgcolor="#aa0000")
t2 = faces.TextFace("Footer Face", fsize=12, fgcolor="#0000aa")
# Attribute faces can now contain prefix and suffix fixed text
aligned = faces.AttrFace("name",
fsize=12,
fgcolor="RoyalBlue",
text_prefix="Aligned (",
text_suffix=")")
# horizontal and vertical alignment per face
aligned.hz_align = 1 # 0 left, 1 center, 2 right
aligned.vt_align = 1
# Node style handling is no longer limited to layout functions. You
# can now create fixed node styles and use them many times, save them
# or even add them to nodes before drawing (this allows to save and
# reproduce an tree image design)
style = NodeStyle()
style["fgcolor"] = "Gold"
style["shape"] = "square"
style["size"] = 15
style["vt_line_color"] = "#ff0000"
t.set_style(style)
# add a face to the style. This face will be render in any node
# associated to the style.
fixed = faces.TextFace("FIXED branch-right", fsize=11, fgcolor="blue")
t.add_face(fixed, column=1, position="branch-right")
# Bind the precomputed style to the root node
# ETE 2.1 has now support for general image properties
ts = TreeStyle()
# You can add faces to the tree image (without any node
# associated). They will be used as headers and foot notes of the
# aligned columns (aligned faces)
ts.aligned_header.add_face(t1, column=0)
ts.aligned_header.add_face(t1, 1)
ts.aligned_header.add_face(t1, 2)
ts.aligned_header.add_face(t1, 3)
t1.hz_align = 1 # 0 left, 1 center, 2 right
t1.border.width = 1
ts.aligned_foot.add_face(t2, column=0)
ts.aligned_foot.add_face(t2, 1)
ts.aligned_foot.add_face(t2, 2)
ts.aligned_foot.add_face(t2, 3)
t2.hz_align = 1
# Set tree image style. Note that aligned header and foot is only
# visible in "rect" mode.
ts.mode = "r"
ts.scale = 10
for node in t.traverse():
# If node is a leaf, add the nodes name and a its scientific
# name
if node.is_leaf():
node.add_face(aligned, column=0, position="aligned")
node.add_face(aligned, column=1, position="aligned")
node.add_face(aligned, column=3, position="aligned")
else:
node.add_face(bt, column=0, position="branch-top")
node.add_face(bb, column=0, position="branch-bottom")
node.add_face(bb2, column=0, position="branch-bottom")
node.add_face(br1, column=0, position="branch-right")
node.add_face(rs1, column=0, position="branch-right")
node.add_face(br2, column=0, position="branch-right")
return t, ts
application.CONFIG["DISPLAY"] = ":0" # This is the most common
# configuration
# We extend the minimum WebTreeApplication with our own WSGI
# application
application.set_external_app_handler(example_app)
# Lets now apply our custom tree loader function to the main
# application
application.set_tree_loader(my_tree_loader)
# And our layout as the default one to render trees
ts = TreeStyle()
ts.show_leaf_name = False
ts.layout_fn.append(main_layout)
ts.mode = "r"
ts.branch_vertical_margin = 5
#ts.scale = 20
application.set_tree_style(ts)
#application.set_default_layout_fn(main_layout)
application.set_tree_size(None, None)
# I want to make up how tree image in shown using a custrom tree
# renderer that adds much more HTML code
application.set_external_tree_renderer(tree_renderer)
# ==============================================================================
# ADD CUSTOM ACTIONS TO THE APPLICATION
#
# The function "register_action" allows to attach functionality to
# nodes in the image. All registered accions will be shown in the
def get_example_tree():
t = Tree()
t.populate(10)
# Margins, alignment, border, background and opacity can now be set for any face
rs1 = faces.TextFace("branch-right\nmargins&borders",
fsize=12, fgcolor="#009000")
rs1.margin_top = 10
rs1.margin_bottom = 50
rs1.margin_left = 40
rs1.margin_right = 40
rs1.border.width = 1
rs1.background.color = "lightgreen"
rs1.inner_border.width = 0
rs1.inner_border.line_style = 1
rs1.inner_border.color= "red"
rs1.opacity = 0.6
rs1.hz_align = 2 # 0 left, 1 center, 2 right
rs1.vt_align = 1 # 0 left, 1 center, 2 right
br1 = faces.TextFace("branch-right1", fsize=12, fgcolor="#009000")
br2 = faces.TextFace("branch-right3", fsize=12, fgcolor="#009000")
# New face positions (branch-top and branch-bottom)
bb = faces.TextFace("branch-bottom 1", fsize=8, fgcolor="#909000")
bb2 = faces.TextFace("branch-bottom 2", fsize=8, fgcolor="#909000")
bt = faces.TextFace("branch-top 1", fsize=6, fgcolor="#099000")
# And faces can also be used as headers or foot notes of aligned
# columns
t1 = faces.TextFace("Header Face", fsize=12, fgcolor="#aa0000")
t2 = faces.TextFace("Footer Face", fsize=12, fgcolor="#0000aa")
# Attribute faces can now contain prefix and suffix fixed text
aligned = faces.AttrFace("name", fsize=12, fgcolor="RoyalBlue",
text_prefix="Aligned (", text_suffix=")")
# horizontal and vertical alignment per face
aligned.hz_align = 1 # 0 left, 1 center, 2 right
aligned.vt_align = 1
# Node style handling is no longer limited to layout functions. You
# can now create fixed node styles and use them many times, save them
# or even add them to nodes before drawing (this allows to save and
# reproduce an tree image design)
style = NodeStyle()
style["fgcolor"] = "Gold"
style["shape"] = "square"
style["size"] = 15
style["vt_line_color"] = "#ff0000"
t.set_style(style)
# add a face to the style. This face will be render in any node
# associated to the style.
fixed = faces.TextFace("FIXED branch-right", fsize=11, fgcolor="blue")
t.add_face(fixed, column=1, position="branch-right")
# Bind the precomputed style to the root node
# ETE 2.1 has now support for general image properties
ts = TreeStyle()
# You can add faces to the tree image (without any node
# associated). They will be used as headers and foot notes of the
# aligned columns (aligned faces)
ts.aligned_header.add_face(t1, column = 0)
ts.aligned_header.add_face(t1, 1)
ts.aligned_header.add_face(t1, 2)
ts.aligned_header.add_face(t1, 3)
t1.hz_align = 1 # 0 left, 1 center, 2 right
t1.border.width = 1
ts.aligned_foot.add_face(t2, column = 0)
ts.aligned_foot.add_face(t2, 1)
ts.aligned_foot.add_face(t2, 2)
ts.aligned_foot.add_face(t2, 3)
t2.hz_align = 1
# Set tree image style. Note that aligned header and foot is only
# visible in "rect" mode.
ts.mode = "r"
ts.scale = 10
for node in t.traverse():
# If node is a leaf, add the nodes name and a its scientific
# name
if node.is_leaf():
node.add_face(aligned, column=0, position="aligned")
node.add_face(aligned, column=1, position="aligned")
node.add_face(aligned, column=3, position="aligned")
else:
node.add_face(bt, column=0, position="branch-top")
node.add_face(bb, column=0, position="branch-bottom")
node.add_face(bb2, column=0, position="branch-bottom")
node.add_face(br1, column=0, position="branch-right")
node.add_face(rs1, column=0, position="branch-right")
node.add_face(br2, column=0, position="branch-right")
return t, ts
def main(argv):
parser = argparse.ArgumentParser(description=__DESCRIPTION__,
formatter_class=argparse.RawDescriptionHelpFormatter)
# name or flags - Either a name or a list of option strings, e.g. foo or -f, --foo.
# action - The basic type of action to be taken when this argument is encountered at the command line. (store, store_const, store_true, store_false, append, append_const, version)
# nargs - The number of command-line arguments that should be consumed. (N, ? (one or default), * (all 1 or more), + (more than 1) )
# const - A constant value required by some action and nargs selections.
# default - The value produced if the argument is absent from the command line.
# type - The type to which the command-line argument should be converted.
# choices - A container of the allowable values for the argument.
# required - Whether or not the command-line option may be omitted (optionals only).
# help - A brief description of what the argument does.
# metavar - A name for the argument in usage messages.
# dest - The name of the attribute to be added to the object returned by parse_args().
parser.add_argument("--show", dest="show_tree",
action="store_true",
help="""Display tree after the analysis.""")
parser.add_argument("--render", dest="render",
action="store_true",
help="""Render tree.""")
parser.add_argument("--dump", dest="dump",
action="store_true",
help="""Dump analysis""")
parser.add_argument("--explore", dest="explore",
type=str,
help="""Reads a previously analyzed tree and visualize it""")
input_args = parser.add_mutually_exclusive_group()
input_args.required=True
input_args.add_argument("-t", "--tree", dest="target_tree", nargs="+",
type=str,
help="""Tree file in newick format""")
input_args.add_argument("-tf", dest="tree_list_file",
type=str,
help="File with the list of tree files")
parser.add_argument("--tax", dest="tax_info", type=str,
help="If the taxid attribute is not set in the"
" newick file for all leaf nodes, a tab file file"
" with the translation of name and taxid can be"
" provided with this option.")
parser.add_argument("--sp_delimiter", dest="sp_delimiter", type=str,
help="If taxid is part of the leaf name, delimiter used to split the string")
parser.add_argument("--sp_field", dest="sp_field", type=int, default=0,
help="field position for taxid after splitting leaf names")
parser.add_argument("--ref", dest="ref_tree", type=str,
help="Uses ref tree to compute robinson foulds"
" distances of the different subtrees")
parser.add_argument("--rf-only", dest="rf_only",
action = "store_true",
help="Skip ncbi consensus analysis")
parser.add_argument("--outgroup", dest="outgroup",
type=str, nargs="+",
help="A list of node names defining the trees outgroup")
parser.add_argument("--is_sptree", dest="is_sptree",
action = "store_true",
help="Assumes no duplication nodes in the tree")
parser.add_argument("-o", dest="output", type=str,
help="Writes result into a file")
parser.add_argument("--tax2name", dest="tax2name", type=str,
help="")
parser.add_argument("--tax2track", dest="tax2track", type=str,
help="")
parser.add_argument("--dump_tax_info", dest="dump_tax_info", action="store_true",
help="")
args = parser.parse_args(argv)
if args.sp_delimiter:
GET_TAXID = lambda x: x.split(args.sp_delimiter)[args.sp_field]
else:
GET_TAXID = None
reftree_name = os.path.basename(args.ref_tree) if args.ref_tree else ""
if args.explore:
print >>sys.stderr, "Reading tree from file:", args.explore
t = cPickle.load(open(args.explore))
ts = TreeStyle()
ts.force_topology = True
ts.show_leaf_name = False
ts.layout_fn = ncbi_layout
ts.mode = "r"
t.show(tree_style=ts)
print >>sys.stderr, "dumping color config"
cPickle.dump(name2color, open("ncbi_colors.pkl", "w"))
sys.exit()
if args.output:
OUT = open(args.output, "w")
else:
OUT = sys.stdout
print >>sys.stderr, "Dumping results into", OUT
target_trees = []
if args.tree_list_file:
target_trees = [line.strip() for line in open(args.tree_list_file)]
if args.target_tree:
target_trees += args.target_tree
prev_tree = None
if args.tax2name:
tax2name = cPickle.load(open(args.tax2name))
else:
tax2name = {}
if args.tax2track:
tax2track = cPickle.load(open(args.tax2track))
else:
tax2track = {}
print len(tax2track), len(tax2name)
header = ("TargetTree", "Subtrees", "Ndups", "Broken subtrees", "Broken clades", "Clade sizes", "RF (avg)", "RF (med)", "RF (std)", "RF (max)", "Shared tips")
print >>OUT, '|'.join([h.ljust(15) for h in header])
if args.ref_tree:
print >>sys.stderr, "Reading ref tree from", args.ref_tree
reft = Tree(args.ref_tree, format=1)
else:
reft = None
SHOW_TREE = False
if args.show_tree or args.render:
SHOW_TREE = True
prev_broken = set()
ENTRIES = []
ncbi.connect_database()
for tfile in target_trees:
#print tfile
t = PhyloTree(tfile, sp_naming_function=None)
if GET_TAXID:
for n in t.iter_leaves():
n.name = GET_TAXID(n.name)
if args.outgroup:
if len(args.outgroup) == 1:
out = t & args.outgroup[0]
else:
out = t.get_common_ancestor(args.outgroup)
if set(out.get_leaf_names()) ^ set(args.outgroup):
raise ValueError("Outgroup is not monophyletic")
t.set_outgroup(out)
t.ladderize()
if prev_tree:
tree_compare(t, prev_tree)
prev_tree = t
if args.tax_info:
tax2name, tax2track = annotate_tree_with_taxa(t, args.tax_info, tax2name, tax2track)
if args.dump_tax_info:
cPickle.dump(tax2track, open("tax2track.pkl", "w"))
cPickle.dump(tax2name, open("tax2name.pkl", "w"))
print "Tax info written into pickle files"
else:
for n in t.iter_leaves():
spcode = n.name
n.add_features(taxid=spcode)
n.add_features(species=spcode)
tax2name, tax2track = annotate_tree_with_taxa(t, None, tax2name, tax2track)
# Split tree into species trees
#subtrees = t.get_speciation_trees()
if not args.rf_only:
#print "Calculating tree subparts..."
t1 = time.time()
if not args.is_sptree:
subtrees = t.split_by_dups()
#print "Subparts:", len(subtrees), time.time()-t1
else:
subtrees = [t]
valid_subtrees, broken_subtrees, ncbi_mistakes, broken_branches, total_rf, broken_clades, broken_sizes = analyze_subtrees(t, subtrees, show_tree=SHOW_TREE)
#print valid_subtrees, broken_subtrees, ncbi_mistakes, total_rf
else:
subtrees = []
valid_subtrees, broken_subtrees, ncbi_mistakes, broken_branches, total_rf, broken_clades, broken_sizes = 0, 0, 0, 0, 0, 0
ndups = 0
nsubtrees = len(subtrees)
rf = 0
rf_max = 0
rf_std = 0
rf_med = 0
common_names = 0
max_size = 0
if reft and len(subtrees) == 1:
rf = t.robinson_foulds(reft, attr_t1="realname")
rf_max = rf[1]
rf = rf[0]
rf_med = rf
elif reft:
#print "Calculating avg RF..."
nsubtrees, ndups, subtrees = t.get_speciation_trees(map_features=["taxid"])
#print len(subtrees), "Sub-Species-trees found"
avg_rf = []
rf_max = 0.0 # reft.robinson_foulds(reft)[1]
sum_size = 0.0
print nsubtrees, "subtrees", ndups, "duplications"
for ii, subt in enumerate(subtrees):
print "\r%d" %ii,
sys.stdout.flush()
try:
partial_rf = subt.robinson_foulds(reft, attr_t1="taxid")
except ValueError:
pass
else:
sptree_size = len(set([n.taxid for n in subt.iter_leaves()]))
sum_size += sptree_size
avg_rf.append((partial_rf[0]/float(partial_rf[1])) * sptree_size)
common_names = len(partial_rf[3])
max_size = max(max_size, sptree_size)
rf_max = max(rf_max, partial_rf[1])
#print partial_rf[:2]
rf = numpy.sum(avg_rf) / float(sum_size) # Treeko dist
rf_std = numpy.std(avg_rf)
rf_med = numpy.median(avg_rf)
sizes_info = "%0.1f/%0.1f +- %0.1f" %( numpy.mean(broken_sizes), numpy.median(broken_sizes), numpy.std(broken_sizes))
iter_values = [os.path.basename(tfile), nsubtrees, ndups,
broken_subtrees, ncbi_mistakes, broken_branches, sizes_info, rf, rf_med,
rf_std, rf_max, common_names]
print >>OUT, '|'.join(map(lambda x: str(x).strip().ljust(15), iter_values))
fixed = sorted([n for n in prev_broken if n not in broken_clades])
new_problems = sorted(broken_clades - prev_broken)
fixed_string = color(', '.join(fixed), "green") if fixed else ""
problems_string = color(', '.join(new_problems), "red") if new_problems else ""
OUT.write(" Fixed clades: %s\n" %fixed_string) if fixed else None
OUT.write(" New broken: %s\n" %problems_string) if new_problems else None
prev_broken = broken_clades
ENTRIES.append([os.path.basename(tfile), nsubtrees, ndups,
broken_subtrees, ncbi_mistakes, broken_branches, sizes_info, fixed_string, problems_string])
OUT.flush()
if args.show_tree or args.render:
ts = TreeStyle()
ts.force_topology = True
#ts.tree_width = 500
ts.show_leaf_name = False
ts.layout_fn = ncbi_layout
ts.mode = "r"
t.dist = 0
if args.show_tree:
#if args.hide_monophyletic:
# tax2monophyletic = {}
# n2content = t.get_node2content()
# for node in t.traverse():
# term2count = defaultdict(int)
# for leaf in n2content[node]:
# if leaf.lineage:
# for term in leaf.lineage:
# term2count[term] += 1
# expected_size = len(n2content)
# for term, count in term2count.iteritems():
# if count > 1
print "Showing tree..."
t.show(tree_style=ts)
else:
t.render("img.svg", tree_style=ts, dpi=300)
print "dumping color config"
cPickle.dump(name2color, open("ncbi_colors.pkl", "w"))
if args.dump:
cPickle.dump(t, open("ncbi_analysis.pkl", "w"))
print
print
HEADER = ("TargetTree", "Subtrees", "Ndups", "Broken subtrees", "Broken clades", "Broken branches", "Clade sizes", "Fixed Groups", "New Broken Clades")
print_table(ENTRIES, max_col_width = 50, row_line=True, header=HEADER)
if args.output:
OUT.close()
application.CONFIG["DISPLAY"] = ":0" # This is the most common # configuration # We extend the minimum WebTreeApplication with our own WSGI # application application.set_external_app_handler(example_app) # Lets now apply our custom tree loader function to the main # application application.set_tree_loader(my_tree_loader) # And our layout as the default one to render trees ts = TreeStyle() ts.show_leaf_name = False ts.layout_fn.append(main_layout) ts.mode = "r" ts.branch_vertical_margin = 5 #ts.scale = 20 application.set_tree_style(ts) #application.set_default_layout_fn(main_layout) application.set_tree_size(None, None) # I want to make up how tree image in shown using a custrom tree # renderer that adds much more HTML code application.set_external_tree_renderer(tree_renderer) # ============================================================================== # ADD CUSTOM ACTIONS TO THE APPLICATION # # The function "register_action" allows to attach functionality to # nodes in the image. All registered accions will be shown in the # popup menu bound to the nodes and faces in the web image.
T.opacity = 0.8
# And place as a float face over the tree
faces.add_face_to_node(T, node, 1, position="aligned")
# Random tree
t = Tree()
t.populate(20, random_branches=True)
# Some random features in all nodes
for n in t.traverse():
n.add_features(weight=random.randint(0, 50))
# Create an empty TreeStyle
ts = TreeStyle()
# Set our custom layout function
ts.layout_fn = layout
# Draw a tree
ts.mode = "c"
# We will add node names manually
ts.show_leaf_name = False
# Show branch data
ts.show_branch_length = True
ts.show_branch_support = True
t.render("tree_faces.png", w=600, dpi=300, tree_style=ts)
t.show(tree_style=ts)
if node.is_leaf():
#node.img_style["size"] = random.randint(50, 50)
f = faces.TextFace("alignedFace", fsize=8, fgcolor="blue")
#f = faces.AttrFace("name", fsize=random.randint(20,20))
faces.add_face_to_node(f, node, 0, position="aligned")
f.border.width = 0
#f = faces.CircleFace(20, "red")
#f = faces.AttrFace("name", fsize=20)
f = faces.TextFace("NAME", fsize=10)
#faces.add_face_to_node(f, node, 0, position="branch-right")
f.border.width = 0
#node.img_style["bgcolor"] = random_color()
#Tree().show()
ts = TreeStyle()
ts.mode = "c"
ts.layout_fn = layout
ts.show_leaf_name = False
ts.arc_span = 340
ts.arc_start = -70
#ts.allow_face_overlap = True
#ts.show_branch_length = True
ts.draw_guiding_lines = False
ts.optimal_scale_level = "mid"
ts.extra_branch_line_color = "red"
ts.root_opening_factor = 0.50
ts.show_border = True
ts.scale = None
t = Tree()
t.populate(200, random_branches=True, branch_range=(0, 0))
t.dist = 0.0
def ncbi_consensus(self, ):
nsubtrees, ndups, subtrees = self.get_speciation_trees(
map_features=["taxid"])
valid_subtrees, broken_subtrees, ncbi_mistakes, broken_branches, total_rf, broken_clades, broken_sizes = analyze_subtrees(
t, subtrees, show_tree=SHOW_TREE)
avg_rf = []
rf_max = 0.0 # reft.robinson_foulds(reft)[1]
sum_size = 0.0
#reftree =
for tn, subt in enumerate(subtrees):
partial_rf = subt.robinson_foulds(reft, attr_t1="taxid")
sptree_size = len(set([n.taxid for n in subt.iter_leaves()]))
sum_size += sptree_size
avg_rf.append((partial_rf[0] / float(partial_rf[1])) * sptree_size)
common_names = len(partial_rf[3])
max_size = max(max_size, sptree_size)
rf_max = max(rf_max, partial_rf[1])
rf = numpy.sum(avg_rf) / float(sum_size) # Treeko dist
rf_std = numpy.std(avg_rf)
rf_med = numpy.median(avg_rf)
sizes_info = "%0.1f/%0.1f +- %0.1f" % (numpy.mean(broken_sizes),
numpy.median(broken_sizes),
numpy.std(broken_sizes))
iter_values = [
os.path.basename(tfile), nsubtrees, ndups, broken_subtrees,
ncbi_mistakes, broken_branches, sizes_info, rf, rf_med, rf_std,
rf_max, common_names
]
print >> OUT, '|'.join(
map(lambda x: str(x).strip().ljust(15), iter_values))
fixed = sorted([n for n in prev_broken if n not in broken_clades])
new_problems = sorted(broken_clades - prev_broken)
fixed_string = color(', '.join(fixed), "green") if fixed else ""
problems_string = color(', '.join(new_problems),
"red") if new_problems else ""
OUT.write(" Fixed clades: %s\n" % fixed_string) if fixed else None
OUT.write(" New broken: %s\n" %
problems_string) if new_problems else None
prev_broken = broken_clades
ENTRIES.append([
os.path.basename(tfile), nsubtrees, ndups, broken_subtrees,
ncbi_mistakes, broken_branches, sizes_info, fixed_string,
problems_string
])
OUT.flush()
if args.show_tree or args.render:
ts = TreeStyle()
ts.force_topology = True
#ts.tree_width = 500
ts.show_leaf_name = False
ts.layout_fn = ncbi_layout
ts.mode = "r"
t.dist = 0
if args.show_tree:
#if args.hide_monophyletic:
# tax2monophyletic = {}
# n2content = t.get_node2content()
# for node in t.traverse():
# term2count = defaultdict(int)
# for leaf in n2content[node]:
# if leaf.lineage:
# for term in leaf.lineage:
# term2count[term] += 1
# expected_size = len(n2content)
# for term, count in term2count.iteritems():
# if count > 1
print "Showing tree..."
t.show(tree_style=ts)
else:
t.render("img.svg", tree_style=ts, dpi=300)
print "dumping color config"
cPickle.dump(name2color, open("ncbi_colors.pkl", "w"))
if args.dump:
cPickle.dump(t, open("ncbi_analysis.pkl", "w"))