def label_groups(t, ngroups):
nstyles = {}
for i in range(1, ngroups + 1):
nstyle = NodeStyle()
nstyle["size"] = 10
color = random_color()
nstyle["fgcolor"] = color
# assign a node style for each group from 1...N
nstyles["Group{}".format(i)] = nstyle
for n in t.traverse():
if n.name.startswith('Group'):
n.set_style(nstyles[n.name])
return nstyles
def draw_tree(tree, conf, outfile):
try:
from ete3 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 range(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)
def draw_tree(tree, conf, outfile):
try:
from ete3 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 range(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)
def double_or_triple():
seqdict = {}
ny_aligned = list(SeqIO.parse('final_ny_aligned_name_fixed.txt', "fasta"))
for record in ny_aligned:
# new = str(record.description).split(' ')
# # print(record.description)
# new = new[0] + '_' + new[1]
# new = new.replace('|', '_')
seqdict[record.id] = str(record.seq)
t = Tree("beast_tree_newick.nh")
place_list = [
'albany', 'manhattan', 'richmond', 'bronx', 'montgomery', 'rockland',
'brooklyn', 'broome', 'new_rochelle', 'schenecktady', 'cayuga',
'new_york', 'staten_island', 'chenango', 'ononda', 'suffolk',
'clinton', 'orange', 'ulster', 'dutchess', 'passiac', 'washington',
'franklin', 'putnam', 'westchester', 'jefferson', 'queens'
]
# color_list = ["red", "yellow", "blue", "green", "cyan", "magenta", "orange", "light blue", "grey", "purple", "brown",
# "pink", "light green"]
color_list = random_color(num=len(place_list))
color_list_mut = random_color(num=9)
mut_list = ['AAC', 'AAG', 'AGC', 'AGG', 'GAG', 'GGC', 'GAC', 'GGG']
for n in t.traverse():
if n.name == '' or n.name == None:
nstyle = NodeStyle()
nstyle["fgcolor"] = "light grey"
nstyle["size"] = 0
n.set_style(nstyle)
else:
n.name = n.name.replace("\'", "")
nstyle = NodeStyle()
sequence = seqdict[n.name][28432:28435]
color = color_list_mut[mut_list.index(sequence)]
nstyle["hz_line_color"] = color
nstyle["fgcolor"] = color
n.set_style(nstyle)
## --------------------------------------------------------------------------
## FOR TRIPLE MUTANT
##
## Author: Arjun
## --------------------------------------------------------------------------
for n in t.traverse():
new_name = n.name.replace("\'", "")
if n.name == None or '|' not in n.name:
nstyle_blank = NodeStyle()
nstyle_blank["fgcolor"] = color_list_mut[8]
nstyle_blank["size"] = 0
n.set_style(nstyle_blank)
# print(n.name)
# names = n.name.split('_', 1)
elif seqdict[new_name][28432] == 'A' and seqdict[new_name][
28433] == 'A' and seqdict[new_name][28434] == 'C':
nstyle_AAC = NodeStyle()
nstyle_AAC["fgcolor"] = "red"
nstyle_AAC["hz_line_color"] = "red"
# nstyle["fgcolor"] = color_list_mut[0]
# nstyle["hz_line_color"] = color_list_mut[0]
nstyle_AAC["size"] = 15
n.set_style(nstyle_AAC)
n.name = 'AAC'
elif seqdict[new_name][28432] == 'A' and seqdict[new_name][
28433] == 'A' and seqdict[new_name][28434] == 'G':
nstyle_AAG = NodeStyle()
# nstyle_AAG["fgcolor"] = "yellow"
nstyle_AAG["fgcolor"] = color_list_mut[1]
nstyle_AAG["hz_line_color"] = color_list_mut[1]
nstyle_AAG["size"] = 15
n.set_style(nstyle_AAG)
n.name = 'AAG'
elif seqdict[new_name][28432] == 'A' and seqdict[new_name][
28433] == 'G' and seqdict[new_name][28434] == 'C':
nstyle_AGC = NodeStyle()
# nstyle_AGC["fgcolor"] = "purple"
nstyle_AGC["fgcolor"] = color_list_mut[2]
nstyle_AGC["hz_line_color"] = color_list_mut[2]
nstyle_AGC["size"] = 15
n.set_style(nstyle_AGC)
n.name = 'AGC'
elif seqdict[new_name][28432] == 'A' and seqdict[new_name][
28433] == 'G' and seqdict[new_name][28434] == 'G':
nstyle_AGG = NodeStyle()
# nstyle_AGG["fgcolor"] = "green"
nstyle_AGG["fgcolor"] = color_list_mut[3]
nstyle_AGG["hz_line_color"] = color_list_mut[3]
nstyle_AGG["size"] = 15
n.set_style(nstyle_AGG)
n.name = 'AGG'
elif seqdict[new_name][28432] == 'G' and seqdict[new_name][
28433] == 'A' and seqdict[new_name][28434] == 'G':
nstyle_GAG = NodeStyle()
# nstyle_GAG["fgcolor"] = "grey"
nstyle_GAG["fgcolor"] = color_list_mut[4]
nstyle_GAG["hz_line_color"] = color_list_mut[4]
nstyle_GAG["size"] = 15
n.set_style(nstyle_GAG)
n.name = 'GAG'
elif seqdict[new_name][28432] == 'G' and seqdict[new_name][
28433] == 'G' and seqdict[new_name][28434] == 'C':
nstyle_GGC = NodeStyle()
# nstyle_GGC["fgcolor"] = "light blue"
nstyle_GGC["fgcolor"] = color_list_mut[5]
nstyle_GGC["hz_line_color"] = color_list_mut[5]
nstyle_GGC["size"] = 15
n.set_style(nstyle_GGC)
n.name = 'GGC'
elif seqdict[new_name][28432] == 'G' and seqdict[new_name][
28433] == 'A' and seqdict[new_name][28434] == 'C':
nstyle_GAC = NodeStyle()
# nstyle_GAC["fgcolor"] = "orange"
nstyle_GAC["fgcolor"] = color_list_mut[6]
nstyle_GAC["hz_line_color"] = color_list_mut[6]
nstyle_GAC["size"] = 15
n.set_style(nstyle_GAC)
n.name = 'GAC'
elif seqdict[new_name][28432] == 'G' and seqdict[new_name][
28433] == 'G' and seqdict[new_name][28434] == 'G':
n.name = 'GGG'
nstyle_GGG = NodeStyle()
nstyle_GGG["fgcolor"] = color_list_mut[7]
nstyle_GGG["hz_line_color"] = color_list_mut[7]
nstyle_GGG["size"] = 15
n.set_style(nstyle_GGG)
# namesplit = new_name.split('|')
# n.name = namesplit[-1]
## --------------------------------------------------------------------------
## FOR DOUBLE MUTANT
##
## Author: Arjun
## --------------------------------------------------------------------------
for n in t.traverse():
new_name = n.name.replace("\'", "")
if n.name == None or '|' not in n.name:
nstyle_blank = NodeStyle()
nstyle_blank["fgcolor"] = color_list_mut[8]
nstyle_blank["size"] = 0
n.set_style(nstyle_blank)
elif seqdict[new_name][18549] == 'C' and seqdict[new_name][
29091] == 'G':
n.name = 'CG'
nstyle_CG = NodeStyle()
nstyle_CG["fgcolor"] = color_list_mut[0]
nstyle_CG["hz_line_color"] = color_list_mut[0]
nstyle_CG["size"] = 15
n.set_style(nstyle_CG)
elif seqdict[new_name][18549] == 'C' and seqdict[new_name][
29091] == 'A':
n.name = 'CA'
nstyle_CA = NodeStyle()
nstyle_CA["fgcolor"] = color_list_mut[1]
nstyle_CA["hz_line_color"] = color_list_mut[1]
nstyle_CA["size"] = 15
n.set_style(nstyle_CA)
elif seqdict[new_name][18549] == 'T' and seqdict[new_name][
29091] == 'G':
n.name = 'TG'
nstyle_TG = NodeStyle()
nstyle_TG["fgcolor"] = color_list_mut[2]
nstyle_TG["hz_line_color"] = color_list_mut[2]
nstyle_TG["size"] = 15
n.set_style(nstyle_TG)
elif seqdict[new_name][18549] == 'T' and seqdict[new_name][
29091] == 'A':
n.name = 'TA'
nstyle_TA = NodeStyle()
nstyle_TA["fgcolor"] = color_list_mut[3]
nstyle_TA["hz_line_color"] = color_list_mut[3]
nstyle_TA["size"] = 15
n.set_style(nstyle_TA)
# for n in t.traverse():
# if n.name != '':
# sequence = seqdict[n.name][28432:28435]
# if n.nstyle["fgcolor"] != color_list_mut[mut_list.index(sequence)]:
# for bad in badlist:
# new_bad = bad.replace("\'", "")
# if new_bad in seqdict:
# print(seqdict[new_bad][28432:28435])
# else:
# print('Name: ' + str(bad))
make_long = TreeStyle()
make_long.scale = 1500
# t.show(tree_style=make_long)
# print(len(badlist))
# print(badlist)
# for n in t.traverse():
# if n.name == '':
# nstyle = NodeStyle()
# nstyle["fgcolor"] = "light grey"
# nstyle["size"] = 0
# n.set_style(nstyle)
# else:
# n.name = n.name.replace("\'", "")
# nstyle = NodeStyle()
# nstyle["hz_line_color"] = color_list[place_list.index(n.name.split('|')[-1])]
# nstyle["fgcolor"] = color_list[place_list.index(n.name.split('|')[-1])]
# n.set_style(nstyle)
# make_long = TreeStyle()
# make_long.scale = 15000
collapse(t)
collapse(t)
collapse(t)
collapse(t)
collapse(t)
# collapse(t)
# collapse(t)
t.show(tree_style=make_long)
def run(args):
if args.text_mode:
from ete3 import Tree
for tindex, tfile in enumerate(args.src_tree_iterator):
#print tfile
if args.raxml:
nw = re.sub(":(\d+\.\d+)\[(\d+)\]", ":\\1[&&NHX:support=\\2]",
open(tfile).read())
t = Tree(nw)
else:
t = Tree(tfile)
print(
t.get_ascii(show_internal=args.show_internal_names,
attributes=args.show_attributes))
return
import random
import re
import colorsys
from collections import defaultdict
from ete3 import (Tree, PhyloTree, TextFace, RectFace, faces, TreeStyle,
add_face_to_node, random_color)
global FACES
if args.face:
FACES = parse_faces(args.face)
else:
FACES = []
# VISUALIZATION
ts = TreeStyle()
ts.mode = args.mode
ts.show_leaf_name = True
ts.tree_width = args.tree_width
for f in FACES:
if f["value"] == "@name":
ts.show_leaf_name = False
break
if args.as_ncbi:
ts.show_leaf_name = False
FACES.extend(
parse_faces([
'value:@sci_name, size:10, fstyle:italic',
'value:@taxid, color:grey, size:6, format:" - %s"',
'value:@sci_name, color:steelblue, size:7, pos:b-top, nodetype:internal',
'value:@rank, color:indianred, size:6, pos:b-bottom, nodetype:internal',
]))
if args.alg:
FACES.extend(
parse_faces([
'value:@sequence, size:10, pos:aligned, ftype:%s' %
args.alg_type
]))
if args.heatmap:
FACES.extend(
parse_faces(['value:@name, size:10, pos:aligned, ftype:heatmap']))
if args.bubbles:
for bubble in args.bubbles:
FACES.extend(
parse_faces([
'value:@%s, pos:float, ftype:bubble, opacity:0.4' % bubble,
]))
ts.branch_vertical_margin = args.branch_separation
if args.show_support:
ts.show_branch_support = True
if args.show_branch_length:
ts.show_branch_length = True
if args.force_topology:
ts.force_topology = True
ts.layout_fn = lambda x: None
for tindex, tfile in enumerate(args.src_tree_iterator):
#print tfile
if args.raxml:
nw = re.sub(":(\d+\.\d+)\[(\d+)\]", ":\\1[&&NHX:support=\\2]",
open(tfile).read())
t = PhyloTree(nw)
else:
t = PhyloTree(tfile)
if args.alg:
t.link_to_alignment(args.alg, alg_format=args.alg_format)
if args.heatmap:
DEFAULT_COLOR_SATURATION = 0.3
BASE_LIGHTNESS = 0.7
def gradient_color(value, max_value, saturation=0.5, hue=0.1):
def rgb2hex(rgb):
return '#%02x%02x%02x' % rgb
def hls2hex(h, l, s):
return rgb2hex(
tuple([
int(x * 255) for x in colorsys.hls_to_rgb(h, l, s)
]))
lightness = 1 - (value * BASE_LIGHTNESS) / max_value
return hls2hex(hue, lightness, DEFAULT_COLOR_SATURATION)
heatmap_data = {}
max_value, min_value = None, None
for line in open(args.heatmap):
if line.startswith('#COLNAMES'):
pass
elif line.startswith('#') or not line.strip():
pass
else:
fields = line.split('\t')
name = fields[0].strip()
values = [float(x) if x else None for x in fields[1:]]
maxv = max(values)
minv = min(values)
if max_value is None or maxv > max_value:
max_value = maxv
if min_value is None or minv < min_value:
min_value = minv
heatmap_data[name] = values
heatmap_center_value = 0
heatmap_color_center = "white"
heatmap_color_up = 0.3
heatmap_color_down = 0.7
heatmap_color_missing = "black"
heatmap_max_value = abs(heatmap_center_value - max_value)
heatmap_min_value = abs(heatmap_center_value - min_value)
if heatmap_center_value <= min_value:
heatmap_max_value = heatmap_min_value + heatmap_max_value
else:
heatmap_max_value = max(heatmap_min_value, heatmap_max_value)
# scale the tree
if not args.height:
args.height = None
if not args.width:
args.width = None
f2color = {}
f2last_seed = {}
for node in t.traverse():
node.img_style['size'] = 0
if len(node.children) == 1:
node.img_style['size'] = 2
node.img_style['shape'] = "square"
node.img_style['fgcolor'] = "steelblue"
ftype_pos = defaultdict(int)
for findex, f in enumerate(FACES):
if (f['nodetype'] == 'any'
or (f['nodetype'] == 'leaf' and node.is_leaf()) or
(f['nodetype'] == 'internal' and not node.is_leaf())):
# if node passes face filters
if node_matcher(node, f["filters"]):
if f["value"].startswith("@"):
fvalue = getattr(node, f["value"][1:], None)
else:
fvalue = f["value"]
# if node's attribute has content, generate face
if fvalue is not None:
fsize = f["size"]
fbgcolor = f["bgcolor"]
fcolor = f['color']
if fcolor:
# Parse color options
auto_m = re.search("auto\(([^)]*)\)", fcolor)
if auto_m:
target_attr = auto_m.groups()[0].strip()
if not target_attr:
color_keyattr = f["value"]
else:
color_keyattr = target_attr
color_keyattr = color_keyattr.lstrip('@')
color_bin = getattr(
node, color_keyattr, None)
last_seed = f2last_seed.setdefault(
color_keyattr, random.random())
seed = last_seed + 0.10 + random.uniform(
0.1, 0.2)
f2last_seed[color_keyattr] = seed
fcolor = f2color.setdefault(
color_bin, random_color(h=seed))
if fbgcolor:
# Parse color options
auto_m = re.search("auto\(([^)]*)\)", fbgcolor)
if auto_m:
target_attr = auto_m.groups()[0].strip()
if not target_attr:
color_keyattr = f["value"]
else:
color_keyattr = target_attr
color_keyattr = color_keyattr.lstrip('@')
color_bin = getattr(
node, color_keyattr, None)
last_seed = f2last_seed.setdefault(
color_keyattr, random.random())
seed = last_seed + 0.10 + random.uniform(
0.1, 0.2)
f2last_seed[color_keyattr] = seed
fbgcolor = f2color.setdefault(
color_bin, random_color(h=seed))
if f["ftype"] == "text":
if f.get("format", None):
fvalue = f["format"] % fvalue
F = TextFace(fvalue,
fsize=fsize,
fgcolor=fcolor or "black",
fstyle=f.get('fstyle', None))
elif f["ftype"] == "fullseq":
F = faces.SeqMotifFace(seq=fvalue,
seq_format="seq",
seqtail_format="seq",
height=fsize)
elif f["ftype"] == "compactseq":
F = faces.SeqMotifFace(
seq=fvalue,
seq_format="compactseq",
seqtail_format="compactseq",
height=fsize)
elif f["ftype"] == "blockseq":
F = faces.SeqMotifFace(
seq=fvalue,
seq_format="blockseq",
seqtail_format="blockseq",
height=fsize,
fgcolor=fcolor or "slategrey",
bgcolor=fbgcolor or "slategrey",
scale_factor=1.0)
fbgcolor = None
elif f["ftype"] == "bubble":
try:
v = float(fvalue)
except ValueError:
rad = fsize
else:
rad = fsize * v
F = faces.CircleFace(radius=rad,
style="sphere",
color=fcolor
or "steelblue")
elif f["ftype"] == "heatmap":
if not f['column']:
col = ftype_pos[f["pos"]]
else:
col = f["column"]
for i, value in enumerate(
heatmap_data.get(node.name, [])):
ftype_pos[f["pos"]] += 1
if value is None:
color = heatmap_color_missing
elif value > heatmap_center_value:
color = gradient_color(
abs(heatmap_center_value - value),
heatmap_max_value,
hue=heatmap_color_up)
elif value < heatmap_center_value:
color = gradient_color(
abs(heatmap_center_value - value),
heatmap_max_value,
hue=heatmap_color_down)
else:
color = heatmap_color_center
node.add_face(RectFace(
20, 20, color, color),
position="aligned",
column=col + i)
# Add header
# for i, name in enumerate(header):
# nameF = TextFace(name, fsize=7)
# nameF.rotation = -90
# tree_style.aligned_header.add_face(nameF, column=i)
F = None
elif f["ftype"] == "profile":
# internal profiles?
F = None
elif f["ftype"] == "barchart":
F = None
elif f["ftype"] == "piechart":
F = None
# Add the Face
if F:
F.opacity = f['opacity'] or 1.0
# Set face general attributes
if fbgcolor:
F.background.color = fbgcolor
if not f['column']:
col = ftype_pos[f["pos"]]
ftype_pos[f["pos"]] += 1
else:
col = f["column"]
node.add_face(F, column=col, position=f["pos"])
if args.image:
t.render("t%d.%s" % (tindex, args.image),
tree_style=ts,
w=args.width,
h=args.height,
units=args.size_units)
else:
t.show(None, tree_style=ts)
# new = new[0] + '_' + new[1]
# new = new.replace('|', '_')
seqdict[record.id] = str(record.seq)
t = Tree("beast_tree_newick.nh")
place_list = [
'albany', 'manhattan', 'richmond', 'bronx', 'montgomery', 'rockland',
'brooklyn', 'broome', 'new_rochelle', 'schenecktady', 'cayuga', 'new_york',
'staten_island', 'chenango', 'ononda', 'suffolk', 'clinton', 'orange',
'ulster', 'dutchess', 'passiac', 'washington', 'franklin', 'putnam',
'westchester', 'jefferson', 'queens'
]
# color_list = ["red", "yellow", "blue", "green", "cyan", "magenta", "orange", "light blue", "grey", "purple", "brown",
# "pink", "light green"]
color_list = random_color(num=len(place_list))
color_list_mut = random_color(num=9)
mut_list = ['AAC', 'AAG', 'AGC', 'AGG', 'GAG', 'GGC', 'GAC', 'GGG']
# for n in t.traverse():
# if n.name == '' or n.name == None:
# nstyle = NodeStyle()
# nstyle["fgcolor"] = "light grey"
# nstyle["size"] = 0
# n.set_style(nstyle)
# else:
# n.name = n.name.replace("\'", "")
# nstyle = NodeStyle()
# sequence = seqdict[n.name][28432:28435]
# color = color_list_mut[mut_list.index(sequence)]
def run(args):
if args.text_mode:
from ete3 import Tree
for tindex, tfile in enumerate(args.src_tree_iterator):
#print tfile
if args.raxml:
nw = re.sub(":(\d+\.\d+)\[(\d+)\]", ":\\1[&&NHX:support=\\2]", open(tfile).read())
t = Tree(nw)
else:
t = Tree(tfile)
print(t.get_ascii(show_internal=args.show_internal_names,
attributes=args.show_attributes))
return
import random
import re
import colorsys
from collections import defaultdict
from ete3 import (Tree, PhyloTree, TextFace, RectFace, faces, TreeStyle,
add_face_to_node, random_color)
global FACES
if args.face:
FACES = parse_faces(args.face)
else:
FACES = []
# VISUALIZATION
ts = TreeStyle()
ts.mode = args.mode
ts.show_leaf_name = True
ts.tree_width = args.tree_width
for f in FACES:
if f["value"] == "@name":
ts.show_leaf_name = False
break
if args.as_ncbi:
ts.show_leaf_name = False
FACES.extend(parse_faces(
['value:@sci_name, size:10, fstyle:italic',
'value:@taxid, color:grey, size:6, format:" - %s"',
'value:@sci_name, color:steelblue, size:7, pos:b-top, nodetype:internal',
'value:@rank, color:indianred, size:6, pos:b-bottom, nodetype:internal',
]))
if args.alg:
FACES.extend(parse_faces(
['value:@sequence, size:10, pos:aligned, ftype:%s' %args.alg_type]
))
if args.heatmap:
FACES.extend(parse_faces(
['value:@name, size:10, pos:aligned, ftype:heatmap']
))
if args.bubbles:
for bubble in args.bubbles:
FACES.extend(parse_faces(
['value:@%s, pos:float, ftype:bubble, opacity:0.4' %bubble,
]))
ts.branch_vertical_margin = args.branch_separation
if args.show_support:
ts.show_branch_support = True
if args.show_branch_length:
ts.show_branch_length = True
if args.force_topology:
ts.force_topology = True
ts.layout_fn = lambda x: None
for tindex, tfile in enumerate(args.src_tree_iterator):
#print tfile
if args.raxml:
nw = re.sub(":(\d+\.\d+)\[(\d+)\]", ":\\1[&&NHX:support=\\2]", open(tfile).read())
t = PhyloTree(nw)
else:
t = PhyloTree(tfile)
if args.alg:
t.link_to_alignment(args.alg, alg_format=args.alg_format)
if args.heatmap:
DEFAULT_COLOR_SATURATION = 0.3
BASE_LIGHTNESS = 0.7
def gradient_color(value, max_value, saturation=0.5, hue=0.1):
def rgb2hex(rgb):
return '#%02x%02x%02x' % rgb
def hls2hex(h, l, s):
return rgb2hex( tuple([int(x*255) for x in colorsys.hls_to_rgb(h, l, s)]))
lightness = 1 - (value * BASE_LIGHTNESS) / max_value
return hls2hex(hue, lightness, DEFAULT_COLOR_SATURATION)
heatmap_data = {}
max_value, min_value = None, None
for line in open(args.heatmap):
if line.startswith('#COLNAMES'):
pass
elif line.startswith('#') or not line.strip():
pass
else:
fields = line.split('\t')
name = fields[0].strip()
values = [float(x) if x else None for x in fields[1:]]
maxv = max(values)
minv = min(values)
if max_value is None or maxv > max_value:
max_value = maxv
if min_value is None or minv < min_value:
min_value = minv
heatmap_data[name] = values
heatmap_center_value = 0
heatmap_color_center = "white"
heatmap_color_up = 0.3
heatmap_color_down = 0.7
heatmap_color_missing = "black"
heatmap_max_value = abs(heatmap_center_value - max_value)
heatmap_min_value = abs(heatmap_center_value - min_value)
if heatmap_center_value <= min_value:
heatmap_max_value = heatmap_min_value + heatmap_max_value
else:
heatmap_max_value = max(heatmap_min_value, heatmap_max_value)
# scale the tree
if not args.height:
args.height = None
if not args.width:
args.width = None
f2color = {}
f2last_seed = {}
for node in t.traverse():
node.img_style['size'] = 0
if len(node.children) == 1:
node.img_style['size'] = 2
node.img_style['shape'] = "square"
node.img_style['fgcolor'] = "steelblue"
ftype_pos = defaultdict(int)
for findex, f in enumerate(FACES):
if (f['nodetype'] == 'any' or
(f['nodetype'] == 'leaf' and node.is_leaf()) or
(f['nodetype'] == 'internal' and not node.is_leaf())):
# if node passes face filters
if node_matcher(node, f["filters"]):
if f["value"].startswith("@"):
fvalue = getattr(node, f["value"][1:], None)
else:
fvalue = f["value"]
# if node's attribute has content, generate face
if fvalue is not None:
fsize = f["size"]
fbgcolor = f["bgcolor"]
fcolor = f['color']
if fcolor:
# Parse color options
auto_m = re.search("auto\(([^)]*)\)", fcolor)
if auto_m:
target_attr = auto_m.groups()[0].strip()
if not target_attr :
color_keyattr = f["value"]
else:
color_keyattr = target_attr
color_keyattr = color_keyattr.lstrip('@')
color_bin = getattr(node, color_keyattr, None)
last_seed = f2last_seed.setdefault(color_keyattr, random.random())
seed = last_seed + 0.10 + random.uniform(0.1, 0.2)
f2last_seed[color_keyattr] = seed
fcolor = f2color.setdefault(color_bin, random_color(h=seed))
if fbgcolor:
# Parse color options
auto_m = re.search("auto\(([^)]*)\)", fbgcolor)
if auto_m:
target_attr = auto_m.groups()[0].strip()
if not target_attr :
color_keyattr = f["value"]
else:
color_keyattr = target_attr
color_keyattr = color_keyattr.lstrip('@')
color_bin = getattr(node, color_keyattr, None)
last_seed = f2last_seed.setdefault(color_keyattr, random.random())
seed = last_seed + 0.10 + random.uniform(0.1, 0.2)
f2last_seed[color_keyattr] = seed
fbgcolor = f2color.setdefault(color_bin, random_color(h=seed))
if f["ftype"] == "text":
if f.get("format", None):
fvalue = f["format"] % fvalue
F = TextFace(fvalue,
fsize = fsize,
fgcolor = fcolor or "black",
fstyle = f.get('fstyle', None))
elif f["ftype"] == "fullseq":
F = faces.SeqMotifFace(seq=fvalue, seq_format="seq",
seqtail_format="seq",
height=fsize)
elif f["ftype"] == "compactseq":
F = faces.SeqMotifFace(seq=fvalue, seq_format="compactseq",
seqtail_format="compactseq",
height=fsize)
elif f["ftype"] == "blockseq":
F = faces.SeqMotifFace(seq=fvalue, seq_format="blockseq",
seqtail_format="blockseq",
height=fsize,
fgcolor=fcolor or "slategrey",
bgcolor=fbgcolor or "slategrey",
scale_factor = 1.0)
fbgcolor = None
elif f["ftype"] == "bubble":
try:
v = float(fvalue)
except ValueError:
rad = fsize
else:
rad = fsize * v
F = faces.CircleFace(radius=rad, style="sphere",
color=fcolor or "steelblue")
elif f["ftype"] == "heatmap":
if not f['column']:
col = ftype_pos[f["pos"]]
else:
col = f["column"]
for i, value in enumerate(heatmap_data.get(node.name, [])):
ftype_pos[f["pos"]] += 1
if value is None:
color = heatmap_color_missing
elif value > heatmap_center_value:
color = gradient_color(abs(heatmap_center_value - value), heatmap_max_value, hue=heatmap_color_up)
elif value < heatmap_center_value:
color = gradient_color(abs(heatmap_center_value - value), heatmap_max_value, hue=heatmap_color_down)
else:
color = heatmap_color_center
node.add_face(RectFace(20, 20, color, color), position="aligned", column=col + i)
# Add header
# for i, name in enumerate(header):
# nameF = TextFace(name, fsize=7)
# nameF.rotation = -90
# tree_style.aligned_header.add_face(nameF, column=i)
F = None
elif f["ftype"] == "profile":
# internal profiles?
F = None
elif f["ftype"] == "barchart":
F = None
elif f["ftype"] == "piechart":
F = None
# Add the Face
if F:
F.opacity = f['opacity'] or 1.0
# Set face general attributes
if fbgcolor:
F.background.color = fbgcolor
if not f['column']:
col = ftype_pos[f["pos"]]
ftype_pos[f["pos"]] += 1
else:
col = f["column"]
node.add_face(F, column=col, position=f["pos"])
if args.image:
t.render("t%d.%s" %(tindex, args.image),
tree_style=ts, w=args.width, h=args.height, units=args.size_units)
else:
t.show(None, tree_style=ts)
