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()
t.populate(10)
ts = TreeStyle()
ts.rotation = 45
ts.show_leaf_name = False
ts.layout_fn = rotation_layout
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()
t.populate(8)
# 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)
# Set bold red branch to the root node
style = NodeStyle()
style["fgcolor"] = "#0f0f0f"
style["size"] = 0
style["vt_line_color"] = "#ff0000"
style["hz_line_color"] = "#ff0000"
style["vt_line_width"] = 8
style["hz_line_width"] = 8
style["vt_line_type"] = 0 # 0 solid, 1 dashed, 2 dotted
style["hz_line_type"] = 0
t.set_style(style)
#Set dotted red lines to the first two branches
style1 = NodeStyle()
style1["fgcolor"] = "#0f0f0f"
style1["size"] = 0
style1["vt_line_color"] = "#ff0000"
style1["hz_line_color"] = "#ff0000"
style1["vt_line_width"] = 2
style1["hz_line_width"] = 2
style1["vt_line_type"] = 2 # 0 solid, 1 dashed, 2 dotted
style1["hz_line_type"] = 2
t.children[0].img_style = style1
t.children[1].img_style = style1
# Set dashed blue lines in all leaves
style2 = NodeStyle()
style2["fgcolor"] = "#000000"
style2["shape"] = "circle"
style2["vt_line_color"] = "#0000aa"
style2["hz_line_color"] = "#0000aa"
style2["vt_line_width"] = 2
style2["hz_line_width"] = 2
style2["vt_line_type"] = 1 # 0 solid, 1 dashed, 2 dotted
style2["hz_line_type"] = 1
for l in t.iter_leaves():
l.img_style = style2
ts = TreeStyle()
ts.layout_fn = layout
ts.show_leaf_name = False
return t, ts
def get_example_tree():
t = Tree()
t.populate(8)
# 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)
# Set bold red branch to the root node
style = NodeStyle()
style["fgcolor"] = "#0f0f0f"
style["size"] = 0
style["vt_line_color"] = "#ff0000"
style["hz_line_color"] = "#ff0000"
style["vt_line_width"] = 8
style["hz_line_width"] = 8
style["vt_line_type"] = 0 # 0 solid, 1 dashed, 2 dotted
style["hz_line_type"] = 0
t.set_style(style)
#Set dotted red lines to the first two branches
style1 = NodeStyle()
style1["fgcolor"] = "#0f0f0f"
style1["size"] = 0
style1["vt_line_color"] = "#ff0000"
style1["hz_line_color"] = "#ff0000"
style1["vt_line_width"] = 2
style1["hz_line_width"] = 2
style1["vt_line_type"] = 2 # 0 solid, 1 dashed, 2 dotted
style1["hz_line_type"] = 2
t.children[0].img_style = style1
t.children[1].img_style = style1
# Set dashed blue lines in all leaves
style2 = NodeStyle()
style2["fgcolor"] = "#000000"
style2["shape"] = "circle"
style2["vt_line_color"] = "#0000aa"
style2["hz_line_color"] = "#0000aa"
style2["vt_line_width"] = 2
style2["hz_line_width"] = 2
style2["vt_line_type"] = 1 # 0 solid, 1 dashed, 2 dotted
style2["hz_line_type"] = 1
for l in t.iter_leaves():
l.img_style = style2
ts = TreeStyle()
ts.layout_fn = layout
ts.show_leaf_name = False
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)
import random
from ete_dev import Tree, TreeStyle, NodeStyle, faces, AttrFace, TreeFace
# Tree Style used to render small trees used as leaf faces
small_ts = TreeStyle()
small_ts.show_leaf_name = True
small_ts.scale = 10
def layout(node):
if node.is_leaf():
# Add node name to laef nodes
N = AttrFace("name", fsize=14, fgcolor="black")
faces.add_face_to_node(N, node, 0)
t = Tree()
t.populate(10)
T = TreeFace(t, small_ts)
# Let's make the sphere transparent
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))
# and www-data (usally the apache user) cannot access display, try # modifiying DISPLAY permisions by executing "xhost +" 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
# and www-data (usally the apache user) cannot access display, try
# modifiying DISPLAY permisions by executing "xhost +"
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
#
def draw(t, draw_alg=True):
def ly(node):
node.img_style['vt_line_width'] = 1
node.img_style['hz_line_width'] = 1
if node.is_leaf():
add_face_to_node(TextFace(' (%s)' %node.name.split()[0].replace("/exon2", ""), fsize=10, fgcolor='slategrey', tight_text=False), node, 1, position='branch-right')
add_face_to_node(TextFace(node.species, fsize=12, fgcolor='black', fstyle='italic', tight_text=False), node, 0, position='branch-right')
c = 1
for tname in tracked_clades:
if tname in node.named_lineage:
linF = TextFace(tname, fsize=10, fgcolor='white')
linF.margin_left = 3
linF.background.color = lin2color[tname]
add_face_to_node(linF, node, c, position='aligned')
c += 1
for n in xrange(1, 20-(c-1)):
add_face_to_node(TextFace('', fsize=10, fgcolor='slategrey'), node, c, position='aligned')
c+=1
if draw_alg and 'sequence' in node.features:
#seqFace = SequenceFace(node.sequence,"aa",13)
seqFace = SeqMotifFace(node.sequence, [])
# [10, 100, "[]", None, 10, "black", "rgradient:blue", "arial|8|white|domain Name"],
motifs = []
last_lt = None
for c, lt in enumerate(node.sequence):
if lt != '-':
if last_lt is None:
last_lt = c
if c+1 == len(node.sequence):
start, end = last_lt, c
w = end-start
motifs.append([start, end, "[]", w, 13, "slategrey", "slategrey", None])
last_lt = None
elif lt == '-':
if last_lt is not None:
start, end = last_lt, c-1
w = end-start
motifs.append([start, end, "[]", w, 13, "slategrey", "slategrey", None])
last_lt = None
if not motifs:
print node, node.sequence
seqFace = SeqMotifFace(node.sequence, motifs,
intermotif_format="line",
seqtail_format="line", scale_factor=1)
add_face_to_node(seqFace, node, 20, aligned=True)
else:
if node.up:
add_face_to_node(TextFace('% 3g' %node.support, fsize=11, fgcolor='indianred'), node, 0, position='branch-top')
if hasattr(node, "support2") and node.up:
add_face_to_node(TextFace('% 3g' %float(node.support2), fsize=11, fgcolor='steelblue'), node, 0, position='branch-bottom')
node.img_style['size'] = 0
node.img_style['hz_line_color'] = 'black'
node.img_style['vt_line_color'] = 'black'
colors = random_color(num=len(tracked_clades))
lin2color = dict([(ln, colors[i]) for i, ln in enumerate(tracked_clades)])
ts = TreeStyle()
ts.draw_aligned_faces_as_table = False
ts.draw_guiding_lines = False
ts.show_leaf_name = False
ts.show_branch_support = False
ts.layout_fn = ly
print 'Rendering tree.pdf'
t.render('tree.svg', tree_style=ts)
t.render('tree.png', tree_style=ts)
def draw_tree(tree, conf, outfile):
try:
from ete_dev import (add_face_to_node, AttrFace, TextFace, TreeStyle, RectFace, CircleFace,
SequenceFace, random_color, SeqMotifFace)
except ImportError as e:
print e
return
def ly_basic(node):
if node.is_leaf():
node.img_style['size'] = 0
else:
node.img_style['size'] = 0
node.img_style['shape'] = 'square'
if len(MIXED_RES) > 1 and hasattr(node, "tree_seqtype"):
if node.tree_seqtype == "nt":
node.img_style["bgcolor"] = "#CFE6CA"
ntF = TextFace("nt", fsize=6, fgcolor='#444', ftype='Helvetica')
add_face_to_node(ntF, node, 10, position="branch-bottom")
if len(NPR_TREES) > 1 and hasattr(node, "tree_type"):
node.img_style['size'] = 4
node.img_style['fgcolor'] = "steelblue"
node.img_style['hz_line_width'] = 1
node.img_style['vt_line_width'] = 1
def ly_leaf_names(node):
if node.is_leaf():
spF = TextFace(node.species, fsize=10, fgcolor='#444444', fstyle='italic', ftype='Helvetica')
add_face_to_node(spF, node, column=0, position='branch-right')
if hasattr(node, 'genename'):
geneF = TextFace(" (%s)" %node.genename, fsize=8, fgcolor='#777777', ftype='Helvetica')
add_face_to_node(geneF, node, column=1, position='branch-right')
def ly_supports(node):
if not node.is_leaf() and node.up:
supFace = TextFace("%0.2g" %(node.support), fsize=7, fgcolor='indianred')
add_face_to_node(supFace, node, column=0, position='branch-top')
def ly_tax_labels(node):
if node.is_leaf():
c = LABEL_START_COL
largest = 0
for tname in TRACKED_CLADES:
if hasattr(node, "named_lineage") and tname in node.named_lineage:
linF = TextFace(tname, fsize=10, fgcolor='white')
linF.margin_left = 3
linF.margin_right = 2
linF.background.color = lin2color[tname]
add_face_to_node(linF, node, c, position='aligned')
c += 1
for n in xrange(c, len(TRACKED_CLADES)):
add_face_to_node(TextFace('', fsize=10, fgcolor='slategrey'), node, c, position='aligned')
c+=1
def ly_full_alg(node):
pass
def ly_block_alg(node):
if node.is_leaf():
if 'sequence' in node.features:
seqFace = SeqMotifFace(node.sequence, [])
# [10, 100, "[]", None, 10, "black", "rgradient:blue", "arial|8|white|domain Name"],
motifs = []
last_lt = None
for c, lt in enumerate(node.sequence):
if lt != '-':
if last_lt is None:
last_lt = c
if c+1 == len(node.sequence):
start, end = last_lt, c
motifs.append([start, end, "()", 0, 12, "slategrey", "slategrey", None])
last_lt = None
elif lt == '-':
if last_lt is not None:
start, end = last_lt, c-1
motifs.append([start, end, "()", 0, 12, "grey", "slategrey", None])
last_lt = None
seqFace = SeqMotifFace(node.sequence, motifs,
intermotif_format="line",
seqtail_format="line", scale_factor=ALG_SCALE)
add_face_to_node(seqFace, node, ALG_START_COL, aligned=True)
TRACKED_CLADES = ["Eukaryota", "Viridiplantae", "Fungi",
"Alveolata", "Metazoa", "Stramenopiles", "Rhodophyta",
"Amoebozoa", "Crypthophyta", "Bacteria",
"Alphaproteobacteria", "Betaproteobacteria", "Cyanobacteria",
"Gammaproteobacteria",]
# ["Opisthokonta", "Apicomplexa"]
colors = random_color(num=len(TRACKED_CLADES), s=0.45)
lin2color = dict([(ln, colors[i]) for i, ln in enumerate(TRACKED_CLADES)])
NAME_FACE = AttrFace('name', fsize=10, fgcolor='#444444')
LABEL_START_COL = 10
ALG_START_COL = 40
ts = TreeStyle()
ts.draw_aligned_faces_as_table = False
ts.draw_guiding_lines = False
ts.show_leaf_name = False
ts.show_branch_support = False
ts.scale = 160
ts.layout_fn = [ly_basic, ly_leaf_names, ly_supports, ly_tax_labels]
MIXED_RES = set()
MAX_SEQ_LEN = 0
NPR_TREES = []
for n in tree.traverse():
if hasattr(n, "tree_seqtype"):
MIXED_RES.add(n.tree_seqtype)
if hasattr(n, "tree_type"):
NPR_TREES.append(n.tree_type)
seq = getattr(n, "sequence", "")
MAX_SEQ_LEN = max(len(seq), MAX_SEQ_LEN)
if MAX_SEQ_LEN:
ALG_SCALE = min(1, 1000./MAX_SEQ_LEN)
ts.layout_fn.append(ly_block_alg)
if len(NPR_TREES) > 1:
rF = RectFace(4, 4, "steelblue", "steelblue")
rF.margin_right = 10
rF.margin_left = 10
ts.legend.add_face(rF, 0)
ts.legend.add_face(TextFace(" NPR node"), 1)
ts.legend_position = 3
if len(MIXED_RES) > 1:
rF = RectFace(20, 20, "#CFE6CA", "#CFE6CA")
rF.margin_right = 10
rF.margin_left = 10
ts.legend.add_face(rF, 0)
ts.legend.add_face(TextFace(" Nucleotide based alignment"), 1)
ts.legend_position = 3
try:
tree.set_species_naming_function(spname)
annotate_tree_with_ncbi(tree)
a = tree.search_nodes(species='Dictyostelium discoideum')[0]
b = tree.search_nodes(species='Chondrus crispus')[0]
#out = tree.get_common_ancestor([a, b])
#out = tree.search_nodes(species='Haemophilus parahaemolyticus')[0].up
tree.set_outgroup(out)
tree.swap_children()
except Exception:
pass
tree.render(outfile, tree_style=ts, w=170, units='mm', dpi=150)
tree.render(outfile+'.svg', tree_style=ts, w=170, units='mm', dpi=150)
tree.render(outfile+'.pdf', tree_style=ts, w=170, units='mm', dpi=150)
import random
from ete_dev import Tree, TreeStyle, NodeStyle, faces, AttrFace, TreeFace
# Tree Style used to render small trees used as leaf faces
small_ts = TreeStyle()
small_ts.show_leaf_name = True
small_ts.scale = 10
def layout(node):
if node.is_leaf():
# Add node name to laef nodes
N = AttrFace("name", fsize=14, fgcolor="black")
faces.add_face_to_node(N, node, 0)
t = Tree()
t.populate(10)
T = TreeFace(t, small_ts)
# Let's make the sphere transparent
T.opacity = 0.8
# And place as a float face over the tree
faces.add_face_to_node(T, node, 1, position="aligned")
def get_example_tree():
# Random tree
t = Tree()
t.populate(20, random_branches=True)
# Some random features in all nodes
if "improve" in node.features:
color = "orange" if float(node.improve) < 0 else "green"
if float(node.improve) == 0:
color = "blue"
support_face = faces.CircleFace(200, color)
faces.add_face_to_node(support_face, node, 0, position="float-behind")
try:
from ete_dev import TreeStyle, NodeStyle, faces
from ete_dev.treeview import random_color
NPR_TREE_STYLE = TreeStyle()
NPR_TREE_STYLE.layout_fn = npr_layout
NPR_TREE_STYLE.show_leaf_name = False
except ImportError:
TreeStyle, NodeStyle, faces, random_color = [None] * 4
NPR_TREE_STYLE = None
# CONVERT shell colors to the same curses palette
COLORS = {
"wr": "\033[1;37;41m", # white on red
"wo": "\033[1;37;43m", # white on orange
"wm": "\033[1;37;45m", # white on magenta
"wb": "\033[1;37;46m", # white on blue
"bw": "\033[1;37;40m", # black on white
"lblue": "\033[1;34m", # light blue
"lred": "\033[1;31m", # light red
"lgreen": "\033[1;32m", # light green
def draw(t, draw_alg=True):
def ly(node):
node.img_style['vt_line_width'] = 1
node.img_style['hz_line_width'] = 1
if node.is_leaf():
add_face_to_node(TextFace(
' (%s)' % node.name.split()[0].replace("/exon2", ""),
fsize=10,
fgcolor='slategrey',
tight_text=False),
node,
1,
position='branch-right')
add_face_to_node(TextFace(node.species,
fsize=12,
fgcolor='black',
fstyle='italic',
tight_text=False),
node,
0,
position='branch-right')
c = 1
for tname in tracked_clades:
if tname in node.named_lineage:
linF = TextFace(tname, fsize=10, fgcolor='white')
linF.margin_left = 3
linF.background.color = lin2color[tname]
add_face_to_node(linF, node, c, position='aligned')
c += 1
for n in xrange(1, 20 - (c - 1)):
add_face_to_node(TextFace('', fsize=10, fgcolor='slategrey'),
node,
c,
position='aligned')
c += 1
if draw_alg and 'sequence' in node.features:
#seqFace = SequenceFace(node.sequence,"aa",13)
seqFace = SeqMotifFace(node.sequence, [])
# [10, 100, "[]", None, 10, "black", "rgradient:blue", "arial|8|white|domain Name"],
motifs = []
last_lt = None
for c, lt in enumerate(node.sequence):
if lt != '-':
if last_lt is None:
last_lt = c
if c + 1 == len(node.sequence):
start, end = last_lt, c
w = end - start
motifs.append([
start, end, "[]", w, 13, "slategrey",
"slategrey", None
])
last_lt = None
elif lt == '-':
if last_lt is not None:
start, end = last_lt, c - 1
w = end - start
motifs.append([
start, end, "[]", w, 13, "slategrey",
"slategrey", None
])
last_lt = None
if not motifs:
print node, node.sequence
seqFace = SeqMotifFace(node.sequence,
motifs,
intermotif_format="line",
seqtail_format="line",
scale_factor=1)
add_face_to_node(seqFace, node, 20, aligned=True)
else:
if node.up:
add_face_to_node(TextFace('% 3g' % node.support,
fsize=11,
fgcolor='indianred'),
node,
0,
position='branch-top')
if hasattr(node, "support2") and node.up:
add_face_to_node(TextFace('% 3g' % float(node.support2),
fsize=11,
fgcolor='steelblue'),
node,
0,
position='branch-bottom')
node.img_style['size'] = 0
node.img_style['hz_line_color'] = 'black'
node.img_style['vt_line_color'] = 'black'
colors = random_color(num=len(tracked_clades))
lin2color = dict([(ln, colors[i]) for i, ln in enumerate(tracked_clades)])
ts = TreeStyle()
ts.draw_aligned_faces_as_table = False
ts.draw_guiding_lines = False
ts.show_leaf_name = False
ts.show_branch_support = False
ts.layout_fn = ly
print 'Rendering tree.pdf'
t.render('tree.svg', tree_style=ts)
t.render('tree.png', tree_style=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()
import random
from ete_dev import Tree, TreeStyle, NodeStyle, faces, AttrFace, TreeFace
# Tree Style used to render small trees used as leaf faces
small_ts = TreeStyle()
small_ts.show_leaf_name = True
small_ts.scale = 10
def layout(node):
if node.is_leaf():
# Add node name to laef nodes
N = AttrFace("name", fsize=14, fgcolor="black")
faces.add_face_to_node(N, node, 0)
t = Tree()
t.populate(10)
T = TreeFace(t, small_ts)
# Let's make the sphere transparent
T.opacity = 0.8
# And place as a float face over the tree
faces.add_face_to_node(T, node, 1, position="aligned")
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))
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 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"))
import random
from ete_dev import Tree, TreeStyle, NodeStyle, faces, AttrFace, TreeFace
# Tree Style used to render small trees used as leaf faces
small_ts = TreeStyle()
small_ts.show_leaf_name = True
small_ts.scale = 10
def layout(node):
if node.is_leaf():
# Add node name to laef nodes
N = AttrFace("name", fsize=14, fgcolor="black")
faces.add_face_to_node(N, node, 0)
t = Tree()
t.populate(10)
T = TreeFace(t, small_ts)
# Let's make the sphere transparent
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():
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"))
