def get_tree_style(self):
ts = TreeStyle()
ts.layout_fn = self.custom_layout
ts.show_leaf_name = False
ts.draw_guiding_lines = True
ts.guiding_lines_type = 0
ts.guiding_lines_color = "#000000"
self._treestyle = ts
return ts
def run_action_change_style(self, tree, a_data):
#print "action change style called.."
if tree.tree_style == self._treestyle:
ts2 = TreeStyle()
ts2.layout_fn = self.custom_layout
ts2.show_leaf_name = False
ts2.draw_guiding_lines = True
ts2.guiding_lines_type = 0 #solid line
ts2.guiding_lines_color = a_data
tree.tree_style = ts2
self._treestyle = ts2
else:
tree.tree_style = self._treestyle
default="pies.svg")
args = parser.parse_args()
plot_tree, subtrees_dict, subtrees_topids = get_phyparts_nodes(
args.species_tree, args.phyparts_root)
concord_dict, conflict_dict = get_concord_and_conflict(args.phyparts_root,
subtrees_dict,
subtrees_topids)
phyparts_dist, phyparts_pies = get_pie_chart_data(args.phyparts_root,
args.num_genes, concord_dict,
conflict_dict)
#Plot Pie Chart
ts = TreeStyle()
ts.show_leaf_name = False
ts.layout_fn = phyparts_pie_layout
nstyle = NodeStyle()
nstyle["size"] = 0
for n in plot_tree.traverse():
n.set_style(nstyle)
n.img_style["vt_line_width"] = 0
ts.draw_guiding_lines = True
ts.guiding_lines_color = "black"
ts.guiding_lines_type = 0
ts.scale = 30
ts.branch_vertical_margin = 10
plot_tree.convert_to_ultrametric()
my_svg = plot_tree.render(args.svg_name, tree_style=ts, w=595)
def plot_tree_barplot(tree_file,
taxon2value_list_barplot,
header_list,
taxon2set2value_heatmap=False,
header_list2=False,
column_scale=True,
general_max=False,
barplot2percentage=False,
taxon2mlst=False):
'''
display one or more barplot
:param tree_file:
:param taxon2value_list:
:param exclude_outgroup:
:param bw_scale:
:param barplot2percentage: list of bool to indicates if the number are percentages and the range should be set to 0-100
:return:
'''
import matplotlib.cm as cm
from matplotlib.colors import rgb2hex
import matplotlib as mpl
if taxon2mlst:
mlst_list = list(set(taxon2mlst.values()))
mlst2color = dict(zip(mlst_list, get_spaced_colors(len(mlst_list))))
mlst2color['-'] = 'white'
if isinstance(tree_file, Tree):
t1 = tree_file
else:
t1 = Tree(tree_file)
# Calculate the midpoint node
R = t1.get_midpoint_outgroup()
# and set it as tree outgroup
t1.set_outgroup(R)
tss = TreeStyle()
value = 1
tss.draw_guiding_lines = True
tss.guiding_lines_color = "gray"
tss.show_leaf_name = False
if column_scale and header_list2:
import matplotlib.cm as cm
from matplotlib.colors import rgb2hex
import matplotlib as mpl
column2scale = {}
for column in header_list2:
values = taxon2set2value_heatmap[column].values()
norm = mpl.colors.Normalize(vmin=min(values), vmax=max(values))
cmap = cm.OrRd
m = cm.ScalarMappable(norm=norm, cmap=cmap)
column2scale[column] = m
cmap = cm.YlGnBu #YlOrRd#OrRd
values_lists = taxon2value_list_barplot.values()
scale_list = []
max_value_list = []
for n, header in enumerate(header_list):
#print 'scale', n, header
data = [float(i[n]) for i in values_lists]
if barplot2percentage is False:
max_value = max(data) #3424182#
min_value = min(data) #48.23
else:
if barplot2percentage[n] is True:
max_value = 100
min_value = 0
else:
max_value = max(data) #3424182#
min_value = min(data) #48.23
norm = mpl.colors.Normalize(vmin=min_value, vmax=max_value)
m1 = cm.ScalarMappable(norm=norm, cmap=cmap)
scale_list.append(m1)
if not general_max:
max_value_list.append(float(max_value))
else:
max_value_list.append(general_max)
for i, lf in enumerate(t1.iter_leaves()):
#if taxon2description[lf.name] == 'Pirellula staleyi DSM 6068':
# lf.name = 'Pirellula staleyi DSM 6068'
# continue
if i == 0:
col_add = 0
if taxon2mlst:
header_list = ['MLST'] + header_list
for col, header in enumerate(header_list):
#lf.add_face(n, column, position="aligned")
n = TextFace(' ')
n.margin_top = 1
n.margin_right = 2
n.margin_left = 2
n.margin_bottom = 1
n.rotation = 90
n.inner_background.color = "white"
n.opacity = 1.
n.hz_align = 2
n.vt_align = 2
tss.aligned_header.add_face(n, col_add + 1)
n = TextFace('%s' % header)
n.margin_top = 1
n.margin_right = 2
n.margin_left = 2
n.margin_bottom = 2
n.rotation = 270
n.inner_background.color = "white"
n.opacity = 1.
n.hz_align = 2
n.vt_align = 1
tss.aligned_header.add_face(n, col_add)
col_add += 2
if header_list2:
for col, header in enumerate(header_list2):
n = TextFace('%s' % header)
n.margin_top = 1
n.margin_right = 20
n.margin_left = 2
n.margin_bottom = 1
n.rotation = 270
n.hz_align = 2
n.vt_align = 2
n.inner_background.color = "white"
n.opacity = 1.
tss.aligned_header.add_face(n, col + col_add)
if taxon2mlst:
try:
#if lf.name in leaf2mlst or int(lf.name) in leaf2mlst:
n = TextFace(' %s ' % taxon2mlst[int(lf.name)])
n.inner_background.color = 'white'
m = TextFace(' ')
m.inner_background.color = mlst2color[taxon2mlst[int(lf.name)]]
except:
n = TextFace(' na ')
n.inner_background.color = "grey"
m = TextFace(' ')
m.inner_background.color = "white"
n.opacity = 1.
n.margin_top = 2
n.margin_right = 2
n.margin_left = 0
n.margin_bottom = 2
m.margin_top = 2
m.margin_right = 0
m.margin_left = 2
m.margin_bottom = 2
lf.add_face(m, 0, position="aligned")
lf.add_face(n, 1, position="aligned")
col_add = 2
else:
col_add = 0
try:
val_list = taxon2value_list_barplot[lf.name]
except:
if not taxon2mlst:
val_list = ['na'] * len(header_list)
else:
val_list = ['na'] * (len(header_list) - 1)
for col, value in enumerate(val_list):
# show value itself
try:
n = TextFace(' %s ' % str(value))
except:
n = TextFace(' %s ' % str(value))
n.margin_top = 1
n.margin_right = 5
n.margin_left = 10
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, col_add, position="aligned")
# show bar
try:
color = rgb2hex(scale_list[col].to_rgba(float(value)))
except:
color = 'white'
try:
percentage = (value / max_value_list[col]) * 100
#percentage = value
except:
percentage = 0
try:
maximum_bar = (
(max_value_list[col] - value) / max_value_list[col]) * 100
except:
maximum_bar = 0
#maximum_bar = 100-percentage
b = StackedBarFace([percentage, maximum_bar],
width=100,
height=10,
colors=[color, "white"])
b.rotation = 0
b.inner_border.color = "grey"
b.inner_border.width = 0
b.margin_right = 15
b.margin_left = 0
lf.add_face(b, col_add + 1, position="aligned")
col_add += 2
if taxon2set2value_heatmap:
shift = col + col_add + 1
i = 0
for col, col_name in enumerate(header_list2):
try:
value = taxon2set2value_heatmap[col_name][lf.name]
except:
try:
value = taxon2set2value_heatmap[col_name][int(lf.name)]
except:
value = 0
if int(value) > 0:
if int(value) > 9:
n = TextFace(' %i ' % int(value))
else:
n = TextFace(' %i ' % int(value))
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.fgcolor = "white"
n.inner_background.color = rgb2hex(
column2scale[col_name].to_rgba(
float(value))) #"orange"
n.opacity = 1.
lf.add_face(n, col + col_add, position="aligned")
i += 1
else:
n = TextFace(' ') #% str(value))
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, col + col_add, position="aligned")
n = TextFace(lf.name, fgcolor="black", fsize=12, fstyle='italic')
lf.add_face(n, 0)
for n in t1.traverse():
nstyle = NodeStyle()
if n.support < 1:
nstyle["fgcolor"] = "black"
nstyle["size"] = 6
n.set_style(nstyle)
else:
nstyle["fgcolor"] = "red"
nstyle["size"] = 0
n.set_style(nstyle)
return t1, tss
def bub_tree(tree, fasta, outfile1, root, types, c_dict, show, size, colours,
field1, field2, scale, multiplier, dna):
"""
:param tree: tree object from ete
:param fasta: the fasta file used to make the tree
:param outfile1: outfile suffix
:param root: sequence name to use as root
:param types: tree type: circular (c) or rectangle (r)
:param c_dict: dictionary mapping colour to time point (from col_map)
:param show: show the tree in a gui (y/n)
:param size: scale the terminal nodes by frequency information (y/n)
:param colours: if using a matched fasta file, colour the sequence by charge/IUPAC
:param field1: the field that contains the size/frequency value
:param field2: the field that contains the size/frequency value
:param scale: how much to scale the x axis
:param multiplier
:param dna true/false, is sequence a DNA sequence?
:param t_list list of time points
:return: None, outputs svg/pdf image of the tree
"""
if multiplier is None:
mult = 500
else:
mult = multiplier
if dna:
dna_prot = 'dna'
bg_c = {
'A': 'green',
'C': 'blue',
'G': 'black',
'T': 'red',
'-': 'grey',
'X': 'white'
}
fg_c = {
'A': 'black',
'C': 'black',
'G': 'black',
'T': 'black',
'-': 'black',
'X': 'white'
}
else:
dna_prot = 'aa'
bg_c = {
'K': '#145AFF',
'R': '#145AFF',
'H': '#8282D2',
'E': '#E60A0A',
'D': '#E60A0A',
'N': '#00DCDC',
'Q': '#00DCDC',
'S': '#FA9600',
'T': '#FA9600',
'L': '#0F820F',
'I': '#0F820F',
'V': '#0F820F',
'Y': '#3232AA',
'F': '#3232AA',
'W': '#B45AB4',
'C': '#E6E600',
'M': '#E6E600',
'A': '#C8C8C8',
'G': '#EBEBEB',
'P': '#DC9682',
'-': 'grey',
'X': 'white'
}
fg_c = {
'K': 'black',
'R': 'black',
'H': 'black',
'E': 'black',
'D': 'black',
'N': 'black',
'Q': 'black',
'S': 'black',
'T': 'black',
'L': 'black',
'I': 'black',
'V': 'black',
'Y': 'black',
'F': 'black',
'W': 'black',
'C': 'black',
'M': 'black',
'A': 'black',
'G': 'black',
'P': 'black',
'-': 'grey',
'X': 'white'
}
if colours == 3:
bg_c = None
fg_c = None
# outfile3 = str(outfile1.replace(".svg", ".nwk"))
tstyle = TreeStyle()
tstyle.force_topology = False
tstyle.mode = types
tstyle.scale = scale
tstyle.min_leaf_separation = 0
tstyle.optimal_scale_level = 'full' # 'mid'
# tstyle.complete_branch_lines_when_necessary = False
if types == 'c':
tstyle.root_opening_factor = 0.25
tstyle.draw_guiding_lines = False
tstyle.guiding_lines_color = 'slateblue'
tstyle.show_leaf_name = False
tstyle.allow_face_overlap = True
tstyle.show_branch_length = False
tstyle.show_branch_support = False
TreeNode(format=0, support=True)
# tnode = TreeNode()
if root is not None:
tree.set_outgroup(root)
# else:
# r = tnode.get_midpoint_outgroup()
# print("r", r)
# tree.set_outgroup(r)
time_col = []
for node in tree.traverse():
# node.ladderize()
if node.is_leaf() is True:
try:
name = node.name.split("_")
time = name[field2]
kind = name[3]
# print(name)
except:
time = 'zero'
name = node.name
print("Incorrect name format for ", node.name)
if size is True:
try:
s = 20 + float(name[field1]) * mult
except:
s = 20
print("No frequency information for ", node.name)
else:
s = 20
colour = c_dict[time]
time_col.append((time, colour))
nstyle = NodeStyle()
nstyle["fgcolor"] = colour
nstyle["size"] = s
nstyle["hz_line_width"] = 10
nstyle["vt_line_width"] = 10
nstyle["hz_line_color"] = colour
nstyle["vt_line_color"] = 'black'
nstyle["hz_line_type"] = 0
nstyle["vt_line_type"] = 0
node.set_style(nstyle)
if root is not None and node.name == root: # place holder in case you want to do something with the root leaf
print('root is ', node.name)
# nstyle["shape"] = "square"
# nstyle["fgcolor"] = "black"
# nstyle["size"] = s
# nstyle["shape"] = "circle"
# node.set_style(nstyle)
else:
nstyle["shape"] = "circle"
node.set_style(nstyle)
if fasta is not None:
seq = fasta[str(node.name)]
seqFace = SequenceFace(seq,
seqtype=dna_prot,
fsize=10,
fg_colors=fg_c,
bg_colors=bg_c,
codon=None,
col_w=40,
alt_col_w=3,
special_col=None,
interactive=True)
# seqFace = SeqMotifFace(seq=seq, motifs=None, seqtype=dna_prot, gap_format=' ', seq_format='()', scale_factor=20,
# height=20, width=50, fgcolor='white', bgcolor='grey', gapcolor='white', )
# seqFace = SeqMotifFace(seq, seq_format="seq", fgcolor=fg_c, bgcolor=bg_c) #interactive=True
(tree & node.name).add_face(seqFace, 0, "aligned")
else:
nstyle = NodeStyle()
nstyle["size"] = 0.1
nstyle["hz_line_width"] = 10
nstyle["vt_line_width"] = 10
node.set_style(nstyle)
continue
tree.ladderize()
# tnode.ladderize()
legendkey = sorted(set(time_col))
legendkey = [(tp, col) for tp, col in legendkey]
# legendkey.insert(0, ('Root', 'black'))
legendkey.append(('', 'white'))
for tm, clr in legendkey:
tstyle.legend.add_face(faces.CircleFace(30, clr), column=0)
tstyle.legend.add_face(faces.TextFace('\t' + tm,
ftype='Arial',
fsize=60,
fgcolor='black',
tight_text=True),
column=1)
if show is True:
tree.show(tree_style=tstyle)
tree.render(outfile1, dpi=600, tree_style=tstyle)
def plot_tree_barplot(tree_file, taxon2mlst, header_list):
'''
display one or more barplot
:param tree_file:
:param taxon2value_list:
:param exclude_outgroup:
:param bw_scale:
:param barplot2percentage: list of bool to indicates if the number are percentages and the range should be set to 0-100
:return:
'''
import matplotlib.cm as cm
from matplotlib.colors import rgb2hex
import matplotlib as mpl
mlst_list = list(set(taxon2mlst.values()))
mlst2color = dict(zip(mlst_list, get_spaced_colors(len(mlst_list))))
mlst2color['-'] = 'white'
if isinstance(tree_file, Tree):
t1 = tree_file
else:
t1 = Tree(tree_file)
# Calculate the midpoint node
R = t1.get_midpoint_outgroup()
# and set it as tree outgroup
t1.set_outgroup(R)
tss = TreeStyle()
value = 1
tss.draw_guiding_lines = True
tss.guiding_lines_color = "gray"
tss.show_leaf_name = False
cmap = cm.YlGnBu #YlOrRd#OrRd
scale_list = []
max_value_list = []
for i, lf in enumerate(t1.iter_leaves()):
#if taxon2description[lf.name] == 'Pirellula staleyi DSM 6068':
# lf.name = 'Pirellula staleyi DSM 6068'
# continue
if i == 0:
# header
col_add = 0
#lf.add_face(n, column, position="aligned")
n = TextFace('MLST')
n.margin_top = 1
n.margin_right = 2
n.margin_left = 2
n.margin_bottom = 1
n.rotation = 90
n.inner_background.color = "white"
n.opacity = 1.
n.hz_align = 2
n.vt_align = 2
tss.aligned_header.add_face(n, col_add + 1)
try:
#if lf.name in leaf2mlst or int(lf.name) in leaf2mlst:
n = TextFace(' %s ' % taxon2mlst[int(lf.name)])
n.inner_background.color = 'white'
m = TextFace(' ')
m.inner_background.color = mlst2color[taxon2mlst[int(lf.name)]]
except:
n = TextFace(' na ')
n.inner_background.color = "grey"
m = TextFace(' ')
m.inner_background.color = "white"
n.opacity = 1.
n.margin_top = 2
n.margin_right = 2
n.margin_left = 0
n.margin_bottom = 2
m.margin_top = 2
m.margin_right = 0
m.margin_left = 2
m.margin_bottom = 2
lf.add_face(m, 0, position="aligned")
lf.add_face(n, 1, position="aligned")
n = TextFace(lf.name, fgcolor="black", fsize=12, fstyle='italic')
lf.add_face(n, 0)
for n in t1.traverse():
nstyle = NodeStyle()
if n.support < 1:
nstyle["fgcolor"] = "black"
nstyle["size"] = 6
n.set_style(nstyle)
else:
nstyle["fgcolor"] = "red"
nstyle["size"] = 0
n.set_style(nstyle)
return t1, tss
from ete3 import TreeStyle
from ete3 import EvolTree
from ete3 import faces
tree = EvolTree("data/S_example/measuring_S_tree.nw")
tree.link_to_alignment('data/S_example/alignment_S_measuring_evol.fasta')
print(tree)
print('\n Running free-ratio model with calculation of ancestral sequences...')
tree.run_model('fb_anc')
#tree.link_to_evol_model('/tmp/ete3-codeml/fb_anc/out', 'fb_anc')
I = TreeStyle()
I.force_topology = False
I.draw_aligned_faces_as_table = True
I.draw_guiding_lines = True
I.guiding_lines_type = 2
I.guiding_lines_color = "#CCCCCC"
for n in sorted(tree.get_descendants() + [tree], key=lambda x: x.node_id):
if n.is_leaf(): continue
anc_face = faces.SequenceFace(n.sequence, 'aa', fsize=10, bg_colors={})
I.aligned_foot.add_face(anc_face, 1)
I.aligned_foot.add_face(
faces.TextFace('node_id: #%d ' % (n.node_id), fsize=8), 0)
print('display result of bs_anc model, with ancestral amino acid sequences.')
tree.show(tree_style=I)
print('\nThe End.')
F = faces.TextFace(mynode.name,fsize=20)
faces.add_face_to_node(F,mynode,0,position="aligned")
#Plot Pie Chart
ts = TreeStyle()
ts.show_leaf_name = False
ts.layout_fn = phyparts_pie_layout
nstyle = NodeStyle()
nstyle["size"] = 0
for n in plot_tree.traverse():
n.set_style(nstyle)
n.img_style["vt_line_width"] = 0
ts.draw_guiding_lines = True
ts.guiding_lines_color = "black"
ts.guiding_lines_type = 0
ts.scale = 30
ts.branch_vertical_margin = 10
plot_tree.convert_to_ultrametric()
plot_tree.ladderize(direction=1)
my_svg = plot_tree.render(args.svg_name,tree_style=ts,w=595,dpi=300)
if args.show_nodes:
node_style = TreeStyle()
node_style.show_leaf_name=False
node_style.layout_fn = node_text_layout
plot_tree.show(tree_style=node_style)
def plot_phylum_counts(NOG_id,
rank='phylum',
colapse_low_species_counts=4,
remove_unlassified=True):
'''
1. get phylum tree
2. foreach species => get phylum
3. build phylum2count dictionnary
3. plot barchart
# merge eukaryotes into 5 main clades
# merge virus as a single clade
ATTENTION: no-rank groups and no-rank species...
'''
import MySQLdb
import os
from chlamdb.biosqldb import manipulate_biosqldb
from ete3 import NCBITaxa, Tree, TextFace, TreeStyle, StackedBarFace
ncbi = NCBITaxa()
sqlpsw = os.environ['SQLPSW']
conn = MySQLdb.connect(
host="localhost", # your host, usually localhost
user="******", # your username
passwd=sqlpsw, # your password
db="eggnog") # name of the data base
cursor = conn.cursor()
sql = 'select * from eggnog.leaf2n_genomes_%s' % rank
cursor.execute(sql, )
leaf_taxon2n_species = manipulate_biosqldb.to_dict(cursor.fetchall())
leaf_taxon2n_species_with_domain = get_NOG_taxonomy(NOG_id, rank)
sql = 'select phylogeny from eggnog.phylogeny where rank="%s"' % (rank)
cursor.execute(sql, )
tree = Tree(cursor.fetchall()[0][0], format=1)
sql = 'select * from eggnog.taxid2label_%s' % rank
cursor.execute(sql, )
taxon_id2scientific_name_and_rank = manipulate_biosqldb.to_dict(
cursor.fetchall())
taxon_id2scientific_name_and_rank = {
str(k): v
for k, v in taxon_id2scientific_name_and_rank.items()
}
tss = TreeStyle()
tss.draw_guiding_lines = True
tss.guiding_lines_color = "blue"
keep = []
for lf in tree.iter_leaves():
# n genomes
if remove_unlassified:
label = taxon_id2scientific_name_and_rank[str(lf.name)][0]
if 'unclassified' in label:
continue
n_genomes = int(leaf_taxon2n_species[lf.name])
if n_genomes > colapse_low_species_counts:
keep.append(lf.name)
print('number of leaaves:', len(keep))
tree.prune(keep)
header_list = ['Rank', 'N genomes', 'N with %s' % NOG_id, 'Percentage']
for col, header in enumerate(header_list):
n = TextFace('%s' % (header))
n.margin_top = 0
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.rotation = 270
n.hz_align = 2
n.vt_align = 2
n.inner_background.color = "white"
n.opacity = 1.
tss.aligned_header.add_face(n, col)
for lf in tree.iter_leaves():
# n genomes
n_genomes = int(leaf_taxon2n_species[lf.name])
if n_genomes <= colapse_low_species_counts:
continue
n = TextFace(' %s ' % str(leaf_taxon2n_species[lf.name]))
n.margin_top = 1
n.margin_right = 1
n.margin_left = 0
n.margin_bottom = 1
n.fsize = 7
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, 2, position="aligned")
# n genomes with domain
try:
m = TextFace(' %s ' %
str(leaf_taxon2n_species_with_domain[lf.name]))
except:
m = TextFace(' 0 ')
m.margin_top = 1
m.margin_right = 1
m.margin_left = 0
m.margin_bottom = 1
m.fsize = 7
m.inner_background.color = "white"
m.opacity = 1.
lf.add_face(m, 3, position="aligned")
# rank
ranks = ncbi.get_rank([lf.name])
try:
r = ranks[max(ranks.keys())]
except:
r = '-'
n = TextFace(' %s ' % r, fsize=14, fgcolor='red')
n.margin_top = 1
n.margin_right = 1
n.margin_left = 0
n.margin_bottom = 1
n.fsize = 7
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, 1, position="aligned")
# percent with target domain
try:
percentage = (float(leaf_taxon2n_species_with_domain[lf.name]) /
float(leaf_taxon2n_species[lf.name])) * 100
except:
percentage = 0
m = TextFace(' %s ' % str(round(percentage, 2)))
m.fsize = 1
m.margin_top = 1
m.margin_right = 1
m.margin_left = 0
m.margin_bottom = 1
m.fsize = 7
m.inner_background.color = "white"
m.opacity = 1.
lf.add_face(m, 4, position="aligned")
b = StackedBarFace([percentage, 100 - percentage],
width=100,
height=10,
colors=["#7fc97f", "white"])
b.rotation = 0
b.inner_border.color = "grey"
b.inner_border.width = 0
b.margin_right = 15
b.margin_left = 0
lf.add_face(b, 5, position="aligned")
n = TextFace('%s' % taxon_id2scientific_name_and_rank[str(lf.name)][0],
fgcolor="black",
fsize=9) # , fstyle = 'italic'
lf.name = " %s (%s)" % (taxon_id2scientific_name_and_rank[str(
lf.name)][0], str(lf.name))
n.margin_right = 10
lf.add_face(n, 0)
tss.show_leaf_name = False
for node in tree.traverse("postorder"):
try:
r = taxon_id2scientific_name_and_rank[str(node.name)][1]
except:
pass
try:
if r in ['phylum', 'superkingdom', 'class', 'subphylum'
] or taxon_id2scientific_name_and_rank[str(
node.name)][0] in ['FCB group']:
hola = TextFace(
"%s" %
(taxon_id2scientific_name_and_rank[str(node.name)][0]))
node.add_face(hola, column=0, position="branch-top")
except:
pass
return tree, tss
circular_style = TreeStyle()
circular_style.show_leaf_name = False
circular_style.show_branch_length = True
circular_style.show_branch_support = True
circular_style.scale = 75
circular_style.tree_width = 50
#circular_style.rotation = 90
#circular_style.extra_branch_line_type=(0)
#circular_style.guiding_lines_type= (0)
#circular_style.title.add_face(TextFace(File, fsize=25), column=0)
circular_style.layout_fn = layout
t.render(adres+"/out/"+File[:-4]+".png", tree_style=circular_style)
circular_style.mode = "r" # draw tree in circular mode
circular_style.extra_branch_line_type=(2)
circular_style.guiding_lines_type= (2)
circular_style.guiding_lines_color =("red")
for n in t.traverse():
nstyle = NodeStyle()
nstyle["fgcolor"] = "red"
nstyle["size"] = 15
n.set_style(nstyle)
#N = AttrFace("name", fsize=30)
#faces.add_face_to_node(N, node, 0, position="aligned")
circular_style.legend.add_face =(TextFace("0.5 support"), column=1)
circular_style.legend.add_face =(CircleFace(10, "red"), column=0)
circular_style.layout_fn = layout
circular_style.legend.add_face(TextFace("0.5 support"), column=1)
circular_style.title.add_face(TextFace(File, fsize=40), column=0)
TextFace_=TextFace(text, ftype='Verdana', fsize=10, fgcolor='black', penwidth=0, fstyle='normal', tight_text=False, bold=False)
t.render(adres+"/out/"+File[:-4]+"_c.png", tree_style=circular_style)
from ete3 import faces
tree = EvolTree ("data/S_example/measuring_S_tree.nw")
tree.link_to_alignment ('data/S_example/alignment_S_measuring_evol.fasta')
print tree
print '\n Running free-ratio model with calculation of ancestral sequences...'
tree.run_model ('fb_anc')
#tree.link_to_evol_model('/tmp/ete3-codeml/fb_anc/out', 'fb_anc')
I = TreeStyle()
I.force_topology = False
I.draw_aligned_faces_as_table = True
I.draw_guiding_lines = True
I.guiding_lines_type = 2
I.guiding_lines_color = "#CCCCCC"
for n in sorted (tree.get_descendants()+[tree],
key=lambda x: x.node_id):
if n.is_leaf(): continue
anc_face = faces.SequenceFace (n.sequence, 'aa', fsize=10, bg_colors={})
I.aligned_foot.add_face(anc_face, 1)
I.aligned_foot.add_face(faces.TextFace('node_id: #%d '%(n.node_id),
fsize=8), 0)
print 'display result of bs_anc model, with ancestral amino acid sequences.'
tree.show(tree_style=I)
print '\nThe End.'
def plot_tree_stacked_barplot(
tree_file,
taxon2value_list_barplot=False,
header_list=False, # header stackedbarplots
taxon2set2value_heatmap=False,
taxon2label=False,
header_list2=False, # header counts columns
biodb=False,
column_scale=True,
general_max=False,
header_list3=False,
set2taxon2value_list_simple_barplot=False,
set2taxon2value_list_simple_barplot_counts=True,
rotate=False,
taxon2description=False):
'''
taxon2value_list_barplot list of lists:
[[bar1_part1, bar1_part2,...],[bar2_part1, bar2_part2]]
valeures de chaque liste transformes en pourcentages
:param tree_file:
:param taxon2value_list:
:param biodb:
:param exclude_outgroup:
:param bw_scale:
:return:
'''
if biodb:
from chlamdb.biosqldb import manipulate_biosqldb
server, db = manipulate_biosqldb.load_db(biodb)
taxon2description = manipulate_biosqldb.taxon_id2genome_description(
server, biodb, filter_names=True)
t1 = Tree(tree_file)
# Calculate the midpoint node
R = t1.get_midpoint_outgroup()
# and set it as tree outgroup
t1.set_outgroup(R)
colors2 = [
"red", "#FFFF00", "#58FA58", "#819FF7", "#F781F3", "#2E2E2E",
"#F7F8E0", 'black'
]
colors = [
"#7fc97f", "#386cb0", "#fdc086", "#ffffb3", "#fdb462", "#f0027f",
"#F7F8E0", 'black'
] # fdc086ff 386cb0ff f0027fff
tss = TreeStyle()
tss.draw_guiding_lines = True
tss.guiding_lines_color = "gray"
tss.show_leaf_name = False
if column_scale and header_list2:
import matplotlib.cm as cm
from matplotlib.colors import rgb2hex
import matplotlib as mpl
column2scale = {}
col_n = 0
for column in header_list2:
values = taxon2set2value_heatmap[column].values()
#print values
if min(values) == max(values):
min_val = 0
max_val = 1.5 * max(values)
else:
min_val = min(values)
max_val = max(values)
#print 'min-max', min_val, max_val
norm = mpl.colors.Normalize(vmin=min_val, vmax=max_val) # *1.1
if col_n < 4:
cmap = cm.OrRd #
else:
cmap = cm.YlGnBu #PuBu#OrRd
m = cm.ScalarMappable(norm=norm, cmap=cmap)
column2scale[column] = [m, float(max_val)] # *0.7
col_n += 1
for i, lf in enumerate(t1.iter_leaves()):
#if taxon2description[lf.name] == 'Pirellula staleyi DSM 6068':
# lf.name = 'Pirellula staleyi DSM 6068'
# continue
if i == 0:
if taxon2label:
n = TextFace(' ')
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.hz_align = 2
n.vt_align = 2
n.rotation = 270
n.inner_background.color = "white"
n.opacity = 1.
tss.aligned_header.add_face(n, 0)
col_add = 1
else:
col_add = 1
if header_list:
for col, header in enumerate(header_list):
n = TextFace('%s' % (header))
n.margin_top = 0
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.rotation = 270
n.hz_align = 2
n.vt_align = 2
n.inner_background.color = "white"
n.opacity = 1.
tss.aligned_header.add_face(n, col + col_add)
col_add += col + 1
if header_list3:
#print 'header_list 3!'
col_tmp = 0
for header in header_list3:
n = TextFace('%s' % (header))
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.rotation = 270
n.hz_align = 2
n.vt_align = 2
n.inner_background.color = "white"
n.opacity = 1.
if set2taxon2value_list_simple_barplot_counts:
if col_tmp == 0:
col_tmp += 1
tss.aligned_header.add_face(n, col_tmp + 1 + col_add)
n = TextFace(' ')
tss.aligned_header.add_face(n, col_tmp + col_add)
col_tmp += 2
else:
tss.aligned_header.add_face(n, col_tmp + col_add)
col_tmp += 1
if set2taxon2value_list_simple_barplot_counts:
col_add += col_tmp
else:
col_add += col_tmp
if header_list2:
for col, header in enumerate(header_list2):
n = TextFace('%s' % (header))
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.rotation = 270
n.hz_align = 2
n.vt_align = 2
n.inner_background.color = "white"
n.opacity = 1.
tss.aligned_header.add_face(n, col + col_add)
col_add += col + 1
if taxon2label:
try:
n = TextFace('%s' % taxon2label[lf.name])
except:
try:
n = TextFace('%s' % taxon2label[int(lf.name)])
except:
n = TextFace('-')
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
if rotate:
n.rotation = 270
lf.add_face(n, 1, position="aligned")
col_add = 2
else:
col_add = 2
if taxon2value_list_barplot:
try:
val_list_of_lists = taxon2value_list_barplot[lf.name]
except:
val_list_of_lists = taxon2value_list_barplot[int(lf.name)]
#col_count = 0
for col, value_list in enumerate(val_list_of_lists):
total = float(sum(value_list))
percentages = [(i / total) * 100 for i in value_list]
if col % 3 == 0:
col_list = colors2
else:
col_list = colors
b = StackedBarFace(percentages,
width=150,
height=18,
colors=col_list[0:len(percentages)])
b.rotation = 0
b.inner_border.color = "white"
b.inner_border.width = 0
b.margin_right = 5
b.margin_left = 5
if rotate:
b.rotation = 270
lf.add_face(b, col + col_add, position="aligned")
#col_count+=1
col_add += col + 1
if set2taxon2value_list_simple_barplot:
col_list = [
'#fc8d59', '#91bfdb', '#99d594', '#c51b7d', '#f1a340',
'#999999'
]
color_i = 0
col = 0
for one_set in header_list3:
if color_i > 5:
color_i = 0
color = col_list[color_i]
color_i += 1
# values for all taxons
values_lists = [
float(i) for i in
set2taxon2value_list_simple_barplot[one_set].values()
]
#print values_lists
#print one_set
value = set2taxon2value_list_simple_barplot[one_set][lf.name]
if set2taxon2value_list_simple_barplot_counts:
if isinstance(value, float):
a = TextFace(" %s " % str(round(value, 2)))
else:
a = TextFace(" %s " % str(value))
a.margin_top = 1
a.margin_right = 2
a.margin_left = 5
a.margin_bottom = 1
if rotate:
a.rotation = 270
lf.add_face(a, col + col_add, position="aligned")
#print 'value and max', value, max(values_lists)
fraction_biggest = (float(value) / max(values_lists)) * 100
fraction_rest = 100 - fraction_biggest
#print 'fractions', fraction_biggest, fraction_rest
b = StackedBarFace([fraction_biggest, fraction_rest],
width=100,
height=15,
colors=[color, 'white'])
b.rotation = 0
b.inner_border.color = "grey"
b.inner_border.width = 0
b.margin_right = 15
b.margin_left = 0
if rotate:
b.rotation = 270
if set2taxon2value_list_simple_barplot_counts:
if col == 0:
col += 1
lf.add_face(b, col + 1 + col_add, position="aligned")
col += 2
else:
lf.add_face(b, col + col_add, position="aligned")
col += 1
if set2taxon2value_list_simple_barplot_counts:
col_add += col
else:
col_add += col
if taxon2set2value_heatmap:
i = 0
#if not taxon2label:
# col_add-=1
for col2, head in enumerate(header_list2):
col_name = header_list2[i]
try:
value = taxon2set2value_heatmap[col_name][str(lf.name)]
except:
try:
value = taxon2set2value_heatmap[col_name][round(
float(lf.name), 2)]
except:
value = 0
if header_list2[i] == 'duplicates':
print('dupli', lf.name, value)
#print 'val----------------', value
if int(value) > 0:
if int(value) >= 10 and int(value) < 100:
n = TextFace('%4i' % value)
elif int(value) >= 100:
n = TextFace('%3i' % value)
else:
n = TextFace('%5i' % value)
n.margin_top = 1
n.margin_right = 2
n.margin_left = 5
n.margin_bottom = 1
n.hz_align = 1
n.vt_align = 1
if rotate:
n.rotation = 270
n.inner_background.color = rgb2hex(
column2scale[col_name][0].to_rgba(
float(value))) #"orange"
#print 'xaxaxaxaxa', value,
if float(value) > column2scale[col_name][1]:
n.fgcolor = 'white'
n.opacity = 1.
n.hz_align = 1
n.vt_align = 1
lf.add_face(n, col2 + col_add, position="aligned")
i += 1
else:
n = TextFace('')
n.margin_top = 1
n.margin_right = 1
n.margin_left = 5
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
if rotate:
n.rotation = 270
lf.add_face(n, col2 + col_add, position="aligned")
i += 1
#lf.name = taxon2description[lf.name]
n = TextFace(taxon2description[lf.name],
fgcolor="black",
fsize=12,
fstyle='italic')
lf.add_face(n, 0)
for n in t1.traverse():
nstyle = NodeStyle()
if n.support < 1:
nstyle["fgcolor"] = "black"
nstyle["size"] = 6
n.set_style(nstyle)
else:
nstyle["fgcolor"] = "red"
nstyle["size"] = 0
n.set_style(nstyle)
return t1, tss
def plot_heat_tree(tree_file,
biodb="chlamydia_04_16",
exclude_outgroup=False,
bw_scale=True):
from chlamdb.biosqldb import manipulate_biosqldb
import matplotlib.cm as cm
from matplotlib.colors import rgb2hex
import matplotlib as mpl
server, db = manipulate_biosqldb.load_db(biodb)
sql_biodatabase_id = 'select biodatabase_id from biodatabase where name="%s"' % biodb
db_id = server.adaptor.execute_and_fetchall(sql_biodatabase_id, )[0][0]
if type(tree_file) == str:
t1 = Tree(tree_file)
try:
R = t1.get_midpoint_outgroup()
#print 'root', R
# and set it as tree outgroup
t1.set_outgroup(R)
except:
pass
elif isinstance(tree_file, Tree):
t1 = tree_file
else:
IOError('Unkown tree format')
tss = TreeStyle()
tss.draw_guiding_lines = True
tss.guiding_lines_color = "gray"
tss.show_leaf_name = False
#print "tree", t1
sql1 = 'select taxon_id, description from bioentry where biodatabase_id=%s and description not like "%%%%plasmid%%%%"' % db_id
sql2 = 'select t2.taxon_id, t1.GC from genomes_info_%s as t1 inner join bioentry as t2 ' \
' on t1.accession=t2.accession where t2.biodatabase_id=%s and t1.description not like "%%%%plasmid%%%%";' % (biodb, db_id)
sql3 = 'select t2.taxon_id, t1.genome_size from genomes_info_%s as t1 ' \
' inner join bioentry as t2 on t1.accession=t2.accession ' \
' where t2.biodatabase_id=%s and t1.description not like "%%%%plasmid%%%%";' % (biodb, db_id)
sql4 = 'select t2.taxon_id,percent_non_coding from genomes_info_%s as t1 ' \
' inner join bioentry as t2 on t1.accession=t2.accession ' \
' where t2.biodatabase_id=%s and t1.description not like "%%%%plasmid%%%%";' % (biodb, db_id)
sql_checkm_completeness = 'select taxon_id, completeness from custom_tables.checkm_%s;' % biodb
sql_checkm_contamination = 'select taxon_id,contamination from custom_tables.checkm_%s;' % biodb
try:
taxon_id2completeness = manipulate_biosqldb.to_dict(
server.adaptor.execute_and_fetchall(sql_checkm_completeness))
taxon_id2contamination = manipulate_biosqldb.to_dict(
server.adaptor.execute_and_fetchall(sql_checkm_contamination))
except:
taxon_id2completeness = False
#taxon2description = manipulate_biosqldb.to_dict(server.adaptor.execute_and_fetchall(sql1,))
taxon2description = manipulate_biosqldb.taxon_id2genome_description(
server, biodb, filter_names=True)
taxon2gc = manipulate_biosqldb.to_dict(
server.adaptor.execute_and_fetchall(sql2, ))
taxon2genome_size = manipulate_biosqldb.to_dict(
server.adaptor.execute_and_fetchall(sql3, ))
taxon2coding_density = manipulate_biosqldb.to_dict(
server.adaptor.execute_and_fetchall(sql4, ))
my_taxons = [lf.name for lf in t1.iter_leaves()]
# Calculate the midpoint node
if exclude_outgroup:
excluded = str(list(t1.iter_leaves())[0].name)
my_taxons.pop(my_taxons.index(excluded))
genome_sizes = [float(taxon2genome_size[i]) for i in my_taxons]
gc_list = [float(taxon2gc[i]) for i in my_taxons]
fraction_list = [float(taxon2coding_density[i]) for i in my_taxons]
value = 1
max_genome_size = max(genome_sizes) #3424182#
max_gc = max(gc_list) #48.23
cmap = cm.YlGnBu #YlOrRd#OrRd
norm = mpl.colors.Normalize(vmin=min(genome_sizes) - 100000,
vmax=max(genome_sizes))
m1 = cm.ScalarMappable(norm=norm, cmap=cmap)
norm = mpl.colors.Normalize(vmin=min(gc_list), vmax=max(gc_list))
m2 = cm.ScalarMappable(norm=norm, cmap=cmap)
norm = mpl.colors.Normalize(vmin=min(fraction_list),
vmax=max(fraction_list))
m3 = cm.ScalarMappable(norm=norm, cmap=cmap)
for i, lf in enumerate(t1.iter_leaves()):
#if taxon2description[lf.name] == 'Pirellula staleyi DSM 6068':
# lf.name = 'Pirellula staleyi DSM 6068'
# continue
if i == 0:
n = TextFace('Size (Mbp)')
n.rotation = -25
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
#lf.add_face(n, 3, position="aligned")
tss.aligned_header.add_face(n, 3)
n = TextFace('GC (%)')
n.rotation = -25
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
#lf.add_face(n, 5, position="aligned")
tss.aligned_header.add_face(n, 5)
n = TextFace('')
#lf.add_face(n, 2, position="aligned")
tss.aligned_header.add_face(n, 2)
#lf.add_face(n, 4, position="aligned")
tss.aligned_header.add_face(n, 4)
n = TextFace('Non coding (%)')
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
n.rotation = -25
#lf.add_face(n, 7, position="aligned")
tss.aligned_header.add_face(n, 7)
n = TextFace('')
#lf.add_face(n, 6, position="aligned")
tss.aligned_header.add_face(n, 6)
if taxon_id2completeness:
n = TextFace('Completeness (%)')
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
n.rotation = -25
#lf.add_face(n, 7, position="aligned")
tss.aligned_header.add_face(n, 9)
n = TextFace('')
#lf.add_face(n, 6, position="aligned")
tss.aligned_header.add_face(n, 8)
n = TextFace('Contamination (%)')
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
n.rotation = -25
#lf.add_face(n, 7, position="aligned")
tss.aligned_header.add_face(n, 11)
n = TextFace('')
#lf.add_face(n, 6, position="aligned")
tss.aligned_header.add_face(n, 10)
value += 1
#print '------ %s' % lf.name
if exclude_outgroup and i == 0:
lf.name = taxon2description[lf.name]
#print '#######################'
continue
n = TextFace(
' %s ' %
str(round(taxon2genome_size[lf.name] / float(1000000), 2)))
n.margin_top = 1
n.margin_right = 1
n.margin_left = 0
n.margin_bottom = 1
n.fsize = 7
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, 2, position="aligned")
#if max_genome_size > 3424182:
# max_genome_size = 3424182
fraction_biggest = (float(taxon2genome_size[lf.name]) /
max_genome_size) * 100
fraction_rest = 100 - fraction_biggest
if taxon2description[lf.name] == 'Rhabdochlamydia helveticae T3358':
col = '#fc8d59'
else:
if not bw_scale:
col = rgb2hex(m1.to_rgba(float(
taxon2genome_size[lf.name]))) # 'grey'
else:
col = '#fc8d59'
b = StackedBarFace([fraction_biggest, fraction_rest],
width=100,
height=9,
colors=[col, 'white'])
b.rotation = 0
b.inner_border.color = "black"
b.inner_border.width = 0
b.margin_right = 15
b.margin_left = 0
lf.add_face(b, 3, position="aligned")
fraction_biggest = (float(taxon2gc[lf.name]) / max_gc) * 100
fraction_rest = 100 - fraction_biggest
if taxon2description[lf.name] == 'Rhabdochlamydia helveticae T3358':
col = '#91bfdb'
else:
if not bw_scale:
col = rgb2hex(m2.to_rgba(float(taxon2gc[lf.name])))
else:
col = '#91bfdb'
b = StackedBarFace([fraction_biggest, fraction_rest],
width=100,
height=9,
colors=[col, 'white'])
b.rotation = 0
b.inner_border.color = "black"
b.inner_border.width = 0
b.margin_left = 0
b.margin_right = 15
lf.add_face(b, 5, position="aligned")
n = TextFace(' %s ' % str(round(float(taxon2gc[lf.name]), 2)))
n.margin_top = 1
n.margin_right = 0
n.margin_left = 0
n.margin_bottom = 1
n.fsize = 7
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, 4, position="aligned")
if taxon2description[lf.name] == 'Rhabdochlamydia helveticae T3358':
col = '#99d594'
else:
if not bw_scale:
col = rgb2hex(m3.to_rgba(float(taxon2coding_density[lf.name])))
else:
col = '#99d594'
n = TextFace(' %s ' % str(float(taxon2coding_density[lf.name])))
n.margin_top = 1
n.margin_right = 0
n.margin_left = 0
n.margin_right = 0
n.margin_bottom = 1
n.fsize = 7
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, 6, position="aligned")
fraction = (float(taxon2coding_density[lf.name]) /
max(taxon2coding_density.values())) * 100
fraction_rest = ((max(taxon2coding_density.values()) -
taxon2coding_density[lf.name]) /
float(max(taxon2coding_density.values()))) * 100
#print 'fraction, rest', fraction, fraction_rest
b = StackedBarFace(
[fraction, fraction_rest],
width=100,
height=9,
colors=[col, 'white'
]) # 1-round(float(taxon2coding_density[lf.name]), 2)
b.rotation = 0
b.margin_right = 1
b.inner_border.color = "black"
b.inner_border.width = 0
b.margin_left = 5
lf.add_face(b, 7, position="aligned")
if taxon_id2completeness:
n = TextFace(' %s ' % str(float(taxon_id2completeness[lf.name])))
n.margin_top = 1
n.margin_right = 0
n.margin_left = 0
n.margin_right = 0
n.margin_bottom = 1
n.fsize = 7
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, 8, position="aligned")
fraction = float(taxon_id2completeness[lf.name])
fraction_rest = 100 - fraction
#print 'fraction, rest', fraction, fraction_rest
b = StackedBarFace(
[fraction, fraction_rest],
width=100,
height=9,
colors=["#d7191c", 'white'
]) # 1-round(float(taxon2coding_density[lf.name]), 2)
b.rotation = 0
b.margin_right = 1
b.inner_border.color = "black"
b.inner_border.width = 0
b.margin_left = 5
lf.add_face(b, 9, position="aligned")
n = TextFace(' %s ' % str(float(taxon_id2contamination[lf.name])))
n.margin_top = 1
n.margin_right = 0
n.margin_left = 0
n.margin_right = 0
n.margin_bottom = 1
n.fsize = 7
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, 10, position="aligned")
fraction = float(taxon_id2contamination[lf.name])
fraction_rest = 100 - fraction
#print 'fraction, rest', fraction, fraction_rest
b = StackedBarFace(
[fraction, fraction_rest],
width=100,
height=9,
colors=["black", 'white'
]) # 1-round(float(taxon2coding_density[lf.name]), 2)
b.rotation = 0
b.margin_right = 1
b.inner_border.color = "black"
b.inner_border.width = 0
b.margin_left = 5
lf.add_face(b, 11, position="aligned")
#lf.name = taxon2description[lf.name]
n = TextFace(taxon2description[lf.name],
fgcolor="black",
fsize=9,
fstyle='italic')
n.margin_right = 30
lf.add_face(n, 0)
for n in t1.traverse():
nstyle = NodeStyle()
if n.support < 1:
nstyle["fgcolor"] = "black"
nstyle["size"] = 6
n.set_style(nstyle)
else:
nstyle["fgcolor"] = "red"
nstyle["size"] = 0
n.set_style(nstyle)
return t1, tss
def plot_tree_text_metadata(tree_file, header2taxon2text, ordered_header_list,
biodb):
from chlamdb.biosqldb import manipulate_biosqldb
server, db = manipulate_biosqldb.load_db(biodb)
t1 = Tree(tree_file)
taxon2description = manipulate_biosqldb.taxon_id2genome_description(
server, biodb, filter_names=True)
# Calculate the midpoint node
R = t1.get_midpoint_outgroup()
# and set it as tree outgroup
t1.set_outgroup(R)
tss = TreeStyle()
tss.draw_guiding_lines = True
tss.guiding_lines_color = "gray"
tss.show_leaf_name = False
for i, leaf in enumerate(t1.iter_leaves()):
# first leaf, add headers
if i == 0:
for column, header in enumerate(ordered_header_list):
n = TextFace('%s' % (header))
n.margin_top = 0
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.rotation = 270
n.hz_align = 2
n.vt_align = 2
n.inner_background.color = "white"
n.opacity = 1.
tss.aligned_header.add_face(n, column)
for column, header in enumerate(ordered_header_list):
text = header2taxon2text[header][int(leaf.name)]
n = TextFace('%s' % text)
n.margin_top = 1
n.margin_right = 1
n.margin_left = 5
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
#n.rotation = 270
leaf.add_face(n, column + 1, position="aligned")
# rename leaf (taxon_id => description)
n = TextFace(taxon2description[leaf.name],
fgcolor="black",
fsize=12,
fstyle='italic')
leaf.add_face(n, 0)
for n in t1.traverse():
# rename leaf
nstyle = NodeStyle()
if n.support < 1:
nstyle["fgcolor"] = "black"
nstyle["size"] = 6
n.set_style(nstyle)
else:
nstyle["fgcolor"] = "red"
nstyle["size"] = 0
n.set_style(nstyle)
return t1, tss
def plot_heatmap_tree_locus(biodb,
tree_file,
taxid2count,
taxid2identity=False,
taxid2locus=False,
reference_taxon=False,
n_paralogs_barplot=False):
'''
plot tree and associated heatmap with count of homolgs
optional:
- add identity of closest homolog
- add locus tag of closest homolog
'''
from chlamdb.biosqldb import manipulate_biosqldb
server, db = manipulate_biosqldb.load_db(biodb)
taxid2organism = manipulate_biosqldb.taxon_id2genome_description(
server, biodb, True)
t1 = Tree(tree_file)
ts = TreeStyle()
ts.draw_guiding_lines = True
ts.guiding_lines_color = "gray"
# Calculate the midpoint node
R = t1.get_midpoint_outgroup()
# and set it as tree outgroup
t1.set_outgroup(R)
leaf_number = 0
for lf in t1.iter_leaves():
if str(lf.name) not in taxid2count:
taxid2count[str(lf.name)] = 0
max_count = max([taxid2count[str(lf.name)] for lf in t1.iter_leaves()])
for i, lf in enumerate(t1.iter_leaves()):
# top leaf, add header
if i == 0:
n = TextFace('Number of homologs')
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
n.rotation = -25
#lf.add_face(n, 7, position="aligned")
ts.aligned_header.add_face(n, 1)
if taxid2identity:
n = TextFace('Protein identity')
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
n.rotation = -25
#lf.add_face(n, 7, position="aligned")
ts.aligned_header.add_face(n, 2)
if taxid2locus:
n = TextFace('Locus tag')
n.margin_top = 1
n.margin_right = 1
n.margin_left = 20
n.margin_bottom = 1
n.inner_background.color = "white"
n.opacity = 1.
n.rotation = -25
#lf.add_face(n, 7, position="aligned")
ts.aligned_header.add_face(n, 3)
leaf_number += 1
lf.branch_vertical_margin = 0
data = [taxid2count[str(lf.name)]]
# possibility to add one or more columns
for col, value in enumerate(data):
col_index = col
if value > 0:
n = TextFace(' %s ' % str(value))
n.margin_top = 2
n.margin_right = 2
if col == 0:
n.margin_left = 20
else:
n.margin_left = 2
n.margin_bottom = 2
n.inner_background.color = "white" # #81BEF7
n.opacity = 1.
lf.add_face(n, col, position="aligned")
else:
n = TextFace(' %s ' % str(value))
n.margin_top = 2
n.margin_right = 2
if col == 0:
n.margin_left = 20
else:
n.margin_left = 2
n.margin_bottom = 2
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, col, position="aligned")
# optionally indicate number of paralogs as a barplot
if n_paralogs_barplot:
col_index += 1
percent = (float(value) / max_count) * 100
n = StackedBarFace([percent, 100 - percent],
width=150,
height=18,
colors=['#6699ff', 'white'],
line_color='white')
n.rotation = 0
n.inner_border.color = "white"
n.inner_border.width = 0
n.margin_right = 15
n.margin_left = 0
lf.add_face(n, col + 1, position="aligned")
# optionally add additionnal column with identity
if taxid2identity:
import matplotlib.cm as cm
from matplotlib.colors import rgb2hex
import matplotlib as mpl
norm = mpl.colors.Normalize(vmin=0, vmax=100)
cmap = cm.OrRd
m = cm.ScalarMappable(norm=norm, cmap=cmap)
try:
if round(taxid2identity[str(lf.name)], 2) != 100:
value = "%.2f" % round(taxid2identity[str(lf.name)], 2)
else:
value = "%.1f" % round(taxid2identity[str(lf.name)], 2)
except:
value = '-'
if str(lf.name) == str(reference_taxon):
value = ' '
n = TextFace(' %s ' % value)
n.margin_top = 2
n.margin_right = 2
n.margin_left = 20
n.margin_bottom = 2
if not value.isspace() and value is not '-':
n.inner_background.color = rgb2hex(m.to_rgba(float(value)))
if float(value) > 82:
n.fgcolor = 'white'
n.opacity = 1.
if str(lf.name) == str(reference_taxon):
n.inner_background.color = '#800000'
lf.add_face(n, col_index + 1, position="aligned")
# optionaly add column with locus name
if taxid2locus:
try:
value = str(taxid2locus[str(lf.name)])
except:
value = '-'
n = TextFace(' %s ' % value)
n.margin_top = 2
n.margin_right = 2
n.margin_left = 2
n.margin_bottom = 2
if str(lf.name) != str(reference_taxon):
n.inner_background.color = "white"
else:
n.fgcolor = '#ff0000'
n.inner_background.color = "white"
n.opacity = 1.
lf.add_face(n, col_index + 2, position="aligned")
lf.name = taxid2organism[str(lf.name)]
return t1, leaf_number, ts
