def get_tree_style(**kwargs):
style = TreeStyle()
style.layout_fn = layout
style.allow_face_overlap = True
style.branch_vertical_margin = 5
style.complete_branch_lines_when_necessary = False
style.draw_aligned_faces_as_table = False
style.scale = 1500
style.scale_length = 0.05
style.show_branch_support = False
style.show_leaf_name = False
for key, value in kwargs.items():
if value == "True":
value = True
else:
try:
value = float(value)
except ValueError:
pass
setattr(style, key, value)
return style
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)
for idx, family in enumerate(families):
for cidx, genus_tree in enumerate(genera_trees[idx]):
tf_genera = TreeFace(genus_tree, ts_genera)
tf_genera.border.width = 2
genus = genera[idx][cidx]
color = colors[str(genus)]
tf_genera.border.color = color
(t & family).add_face(tf_genera, column=0, position='aligned')
for n in genus_tree.iter_search_nodes():
if n.dist == 1:
n.img_style = ns_genera
ts_genera.show_leaf_name = False
ts_genera.show_scale = False
ts_genera.layout_fn = my_layout
ts.branch_vertical_margin = 10
ts.show_leaf_name = False
ts.branch_vertical_margin = 15
ts.layout_fn = my_layout
ts.draw_guiding_lines = True
ts.guiding_lines_type = 1
ts.show_scale = False
ts.allow_face_overlap = False
# ts.mode = "c"
# ts.arc_start = 180 # 0 degrees = 3 o'clock
# ts.arc_span = 270
t.show(tree_style=ts)
t.render("mytree.png", w=183, units="mm", tree_style=ts)
def format_tree(tree,
alignment,
al_len_dict,
edpos,
codontable={},
colors=None,
codon_col={},
text="C-to-U RNA editing",
ic_contents=[]):
"""Format the rendering of tree data for alignment"""
t = tree.copy()
# alignment is ordered dict
# flip alignment dict from gene ==> species ==> seq
# to species ==> gene ==> seq
specSeq = ddict(str)
edposSeq = ddict(list)
cur_len = 0
limits = []
for gname, specdict in alignment.items():
for node in t:
# fill missing with gap
specSeq[node.name] += specdict.get(node.name,
al_len_dict[gname] * '-')
edposSeq[node.name] += [
x + cur_len for x in edpos[gname].get(node.name, [])
]
# if node.name == 'Y08501':
# print(gname)
# print( edposSeq[node.name])
cur_len += al_len_dict.get(gname, 0)
limits.append((gname, cur_len))
for node in t:
node.add_feature("sequence", specSeq[node.name])
node.add_feature('edlist', edposSeq[node.name])
ts = TreeStyle()
ts.branch_vertical_margin = 15
ts.scale = 15
ts.allow_face_overlap = False
ts.show_scale = False
ts.show_leaf_name = False
ns = NodeStyle()
ns['shape'] = 'square'
ns['fgcolor'] = 'black'
ns['size'] = 0
def layout(node):
node.img_style = ns
if node.is_leaf():
faces.add_face_to_node(AttrFace(
'fullname',
fsize=14,
fgcolor=(MARKED_NODE_COLOR if
(node.name in colors
or node.fullname in colors) else 'black')),
node,
0,
position="aligned")
if hasattr(node, "sequence") and node.sequence:
seqface = SequenceFace(node.sequence,
"codon",
fsize=13,
codontable=codontable,
col_w=RES_COL_WIDTH,
bg_colors=codon_col,
black_out=node.edlist)
faces.add_face_to_node(seqface, node, 1, position="aligned")
ts.layout_fn = layout
# ts.title.add_face(TextFace('(%s) - SP score : %.0f | IC = %.2f' % (codon, sum(SP_score), sum(ic_contents)),
# fsize=14, fgcolor='red'), 0)
# ts.aligned_header.add_face(
# faces.RectFace(14, 14, 'white', 'white'), 1)
# ts.aligned_foot.add_face(
# faces.RectFace(14, 14, 'white', 'white'), 1)
# for (cod, col) in codon_col.items():
# ts.legend.add_face(faces.RectFace(50, 25, col, col), column=0)
# ts.legend.add_face(TextFace(" %s " % cod, fsize=8), column=1)
ts.legend.add_face(TextFace(text, fsize=14), column=1)
ts.legend_position = 1
ind = 1
prev_gend = 0
for (gname, gend) in limits:
ts.aligned_foot.add_face(
List90Face(list(range(0, gend - prev_gend, 3)),
fsize=13,
ftype='Monospace',
col_w=RES_COL_WIDTH * 3), ind)
ts.aligned_foot.add_face(
faces.RectFace(RES_COL_WIDTH * (gend - prev_gend), 13, '#BBBBBB',
'#EEEEEE'), ind)
ts.aligned_foot.add_face(TextFace(gname, fsize=13), ind)
ts.aligned_foot.add_face(
faces.RectFace(RES_COL_WIDTH * (gend - prev_gend), 5, 'white',
'white'), ind)
prev_gend += gend
ind += 1
#t.dist = 0
ts.margin_left = 5
ts.margin_right = 5
ts.margin_bottom = 5
return t, ts
def generateFigure(PF, sample, rank, input_file, output_base_name, file_type, plot_l1, scaling, output_dpi):
# Make the ETE3 tree
try:
tree = ncbi.get_topology(PF.get_all_tax_ids(sample), rank_limit=rank)
except:
logging.getLogger('Tampa').critical("Input format not compatible.")
exit(1)
ts = TreeStyle()
ts.layout_fn = PF.layout
ts.mode = "c"
ts.show_leaf_name = False
ts.show_branch_length = False
ts.show_branch_support = False
ts.min_leaf_separation = 10
ts.arc_span = 360
#ts.legend.add_face(CircleFace(100, "#1b9e77", label="Predicted"), column=0)
#ts.legend.add_face(CircleFace(100, '#d95f02', label="True"), column=1)
# add white space to move the legend closer
ts.legend.add_face(CircleFace(65, "#FFFFFF"), column=2)
ts.legend.add_face(CircleFace(65, "#FFFFFF"), column=1)
ts.legend.add_face(CircleFace(65, "#FFFFFF"), column=0)
ts.legend.add_face(CircleFace(65, "#FFFFFF"), column=2)
ts.legend.add_face(CircleFace(65, "#FFFFFF"), column=1)
ts.legend.add_face(CircleFace(65, "#FFFFFF"), column=0)
# add the legend
legend_fs = 50
C1 = CircleFace(100, "#1b9e77")
C1.hz_align = True
ts.legend.add_face(C1, column=0)
T1 = TextFace("Predicted", fsize=legend_fs)
T1.hz_align = True
ts.legend.add_face(T1, column=0)
if len(PF.ground_truth_dict) > 0:
C2 = CircleFace(100, "#d95f02")
C2.hz_align = True
ts.legend.add_face(C2, column=1)
T2 = TextFace("True", fsize=legend_fs)
T2.hz_align = True
ts.legend.add_face(T2, column=1)
T3 = TextFace(f"Tool: {os.path.basename(input_file).split('.')[0]}", fsize=legend_fs)
T3.hz_align = True
ts.legend.add_face(T3, column=0)
ts.allow_face_overlap = False # this lets me mess a bit with font size and face size without the interaction of the two
ts.min_leaf_separation = 10
tree_output_file = f"{output_base_name}_tree_{rank}_{sample}.{file_type}"
tree.render(tree_output_file, h=5.2, w=5, tree_style=ts, units="in", dpi=output_dpi)
if plot_l1:
# if you asked for L1 too, then plot that
true_abundance_at_rank = []
predicted_abundance_at_rank = []
for node in tree.get_leaves():
if node.rank == rank:
tax_id = str(node.taxid)
if tax_id in PF.ground_truth_tax_id_to_percentage:
true_abundance_at_rank.append(PF.ground_truth_tax_id_to_percentage[str(node.taxid)] / 100.)
else:
true_abundance_at_rank.append(0)
if tax_id in PF.profile_tax_id_to_percentage:
predicted_abundance_at_rank.append(PF.profile_tax_id_to_percentage[str(node.taxid)] / 100.)
else:
predicted_abundance_at_rank.append(0)
data = np.zeros((len(true_abundance_at_rank), 2))
data[:, 0] = np.array(true_abundance_at_rank)
data[:, 1] = np.array(predicted_abundance_at_rank)
df = pd.DataFrame(data, columns=['True', 'Predicted'])
# g = seaborn.FacetGrid(df, height=6)
ax = seaborn.scatterplot(x='True', y='Predicted', data=df, color='b', s=55)
eps = 1
ax.set_aspect('equal')
max_val = np.max(data) + eps
ax.set_xlim(-.5, max_val)
ax.set_ylim(-.5, max_val)
ax.set_xbound(-.5, max_val)
ax.set_ybound(-.5, max_val)
#plt.figure(figsize=(6,6))
plt.plot(np.linspace(0, max_val, 100), np.linspace(0, max_val, 100), color='k')
for (x, y) in zip(true_abundance_at_rank, predicted_abundance_at_rank):
if x > y:
ax.vlines(x, y, x, colors='r')
if y > x:
ax.vlines(x, x, y, colors='r')
plt.title(f"Tool: {os.path.basename(input_file).split('.')[0]}")
plt.tight_layout()
l1_out_file = f"{output_base_name}_L1_{rank}.{file_type}"
plt.savefig(l1_out_file, dpi=output_dpi)
def scan_internals_pearR(tree,
size,
threshold,
sources="none",
simpson_threhold=0.4):
global t
#sources is defaulted to be "none"
import math, seaborn
import numpy as np
from scipy.stats import pearsonr, spearmanr
R_list = []
R2_list = []
S_index_list = []
tree_path = os.path.join(filepath, tree)
t = Tree(tree_path, format=0)
internals_dict = {}
internal_nodes = []
avoid_sources = ["Unknown"] # sources to be omitted
i = 0
path_trees = 'time_signal_trees'
if not os.path.exists(path_trees):
os.mkdir(path_trees)
for node in t.traverse():
if len(node) >= size:
internal_nodes.append(node)
dist_list = []
year_list = []
if sources != "none":
source_list = []
node.add_features(nodetype='internal')
conf = node.support
for leaf in node: #may change with different label format
###time
internal_dist = node.get_distance(leaf)
year_list.append(leaf.name.split('_')[1])
dist_list.append(internal_dist)
###end of time
###sources
if sources != "none":
z = leaf.name
s = z.split("_")[4]
if s not in avoid_sources:
source_list.append(s)
####end of sources
len_leaves = len(year_list)
x_years = np.asarray(year_list).astype(np.int)
y_dists = np.asarray(dist_list)
R, P = spearmanr(x_years, y_dists)
Rpear, Ppear = pearsonr(x_years, y_dists)
###for sources
if sources != "none":
source_names, source_fre = Uniq(source_list)
s_index = simpson(source_fre)
S_index_list.append(s_index)
###end of sources
if math.isnan(R) != True:
if R * R >= threshold:
i += 1
nodetree = str(i) + '_R2_' + str(round(R * R, 2)) + '.tree'
node.write(outfile=filepath + '/' + path_trees + '/' +
nodetree,
format=0)
nt = Tree(filepath + '/' + path_trees + '/' + nodetree)
leaves = [leaf.name.replace("'", "") for leaf in nt]
leaves_num = len(leaves)
leave_first = leaves[0].split('_')[0]
leavesfile = open(
filepath + '/' + path_trees + '/' + nodetree + '.' +
str(leaves_num) + '.' + leave_first + '.leaves.txt',
'w')
leavesfile.write("\n".join(leaves))
internals_dict[node] = R, P
node.add_features(Rsize=int(R * R * 50))
R2_text = TextFace('R2=' + str(round(R * R, 2)))
#node.add_face(R2_text,column=0,position='branch-top')
leaves_text = TextFace('Leaves=' + str(len_leaves))
#node.add_face(leaves_text,column=0,position='branch-bottom')
R2_list.append(R * R)
for leaf in node:
leaf.add_features(showname=True)
elif Rpear * Rpear >= threshold:
i += 1
nodetree = str(i) + '_R2_' + str(round(Rpear * Rpear,
2)) + '.tree'
node.write(outfile=filepath + '/' + path_trees + '/' +
nodetree,
format=0)
nt = Tree(filepath + '/' + path_trees + '/' + nodetree)
leaves = [leaf.name.replace("'", "") for leaf in nt]
leaves_num = len(leaves)
leave_first = leaves[0].split('_')[0]
leavesfile = open(
filepath + '/' + path_trees + '/' + nodetree + '.' +
str(leaves_num) + '.' + leave_first + '.leaves.txt',
'w')
leavesfile.write("\n".join(leaves))
internals_dict[node] = Rpear, Ppear
node.add_features(Rpearsize=int(Rpear * Rpear * 50))
R2_text = TextFace('R2=' + str(round(Rpear * Rpear, 2)))
#node.add_face(R2_text,column=0,position='branch-top')
leaves_text = TextFace('Leaves=' + str(len_leaves))
#node.add_face(leaves_text,column=0,position='branch-bottom')
R2_list.append(Rpear * Rpear)
for leaf in node:
leaf.add_features(showname=True)
else:
internals_dict[node] = R, P
R2_list.append(R * R)
###for sources
if sources != "none":
if s_index <= simpson_threhold: #more clonal, low diversity
nstyle["hz_line_color"] = "blue"
node.set_style(nstyle)
source_text = TextFace('S=' + str(round(s_index, 2)),
fgcolor="blue",
fsize=15)
node.add_face(source_text,
column=1,
position='branch-bottom')
else:
nstyle["hz_line_color"] = "green"
node.set_style(nstyle)
source_text = TextFace('S=' + str(round(s_index, 2)),
fgcolor="green",
fsize=15)
node.add_face(source_text,
column=1,
position='branch-bottom')
###end of sources
###for time
## seaborn.set(style="white", palette="muted", color_codes=True)
## sns_plot=seaborn.distplot(np.array(R2_list),rug=True)
## fig = sns_plot.get_figure()
## fig.savefig(os.path.join(filepath, tree.rsplit('.')[0]+"_R2_distribution.png"))
## sns_plot.clear()
###end of time
###for source
if sources != "none":
seaborn.set(style="white", palette="muted", color_codes=True)
sns_plot2 = seaborn.distplot(np.array(S_index_list), rug=True)
fig2 = sns_plot2.get_figure()
fig2.savefig(filepath + tree.rsplit('.')[0] +
"_simpson_index_distribution.png")
sns_plot2.clear()
###end of source
ns = NodeStyle()
ns["vt_line_width"] = 2
ns["hz_line_width"] = 2
ns["size"] = 0
for node in t.traverse():
node.set_style(ns)
ts = TreeStyle()
ts.layout_fn = layout
ts.mode = "c"
ts.scale = 180
ts.show_leaf_name = False
ts.force_topology = True
ts.allow_face_overlap = True
#ts.branch_vertical_margin=2
#t.render(filepath+tree.rsplit('.')[0]+"_time_signals.png",dpi=300,tree_style=ts)
outpath = os.path.join(filepath, tree.rsplit('.')[0] + "_time_signals.pdf")
t.render(outpath, tree_style=ts)
def render(self,
outfile,
idlabel=False,
isolabel=False,
colormap=None,
chain_split=None):
'''Render to image file, filetype inferred from suffix, svg for color images'''
def my_layout(node):
circle_color = 'lightgray' if colormap is None or node.name not in colormap else colormap[
node.name]
text_color = 'black'
if isinstance(circle_color, str):
if isolabel and hasattr(node, 'isotype'):
nl = ''.join(
sorted(set([ISO_SHORT[iss] for iss in node.isotype]),
key=lambda x: ISO_TYPE_charORDER[x]))
else:
nl = str(node.frequency)
C = CircleFace(radius=max(3, 10 * scipy.sqrt(node.frequency)),
color=circle_color,
label={
'text': nl,
'color': text_color
} if node.frequency > 0 else None)
C.rotation = -90
C.hz_align = 1
faces.add_face_to_node(C, node, 0)
else:
P = PieChartFace(
[100 * x / node.frequency for x in circle_color.values()],
2 * 10 * scipy.sqrt(node.frequency),
2 * 10 * scipy.sqrt(node.frequency),
colors=[(color if color != 'None' else 'lightgray')
for color in list(circle_color.keys())],
line_color=None)
T = TextFace(' '.join(
[str(x) for x in list(circle_color.values())]),
tight_text=True)
T.hz_align = 1
T.rotation = -90
faces.add_face_to_node(P, node, 0, position='branch-right')
faces.add_face_to_node(T, node, 1, position='branch-right')
if idlabel:
T = TextFace(node.name, tight_text=True, fsize=6)
T.rotation = -90
T.hz_align = 1
faces.add_face_to_node(
T,
node,
1 if isinstance(circle_color, str) else 2,
position='branch-right')
elif isolabel and hasattr(node, 'isotype') and False:
iso_name = ''.join(
sorted(set([ISO_SHORT[iss] for iss in node.isotype]),
key=lambda x: ISO_TYPE_charORDER[x]))
#T = TextFace(iso_name, tight_text=True, fsize=6)
#T.rotation = -90
#T.hz_align = 1
#faces.add_face_to_node(T, node, 1 if isinstance(circle_color, str) else 2, position='branch-right')
C = CircleFace(radius=max(3, 10 * scipy.sqrt(node.frequency)),
color=circle_color,
label={
'text': iso_name,
'color': text_color
} if node.frequency > 0 else None)
C.rotation = -90
C.hz_align = 1
faces.add_face_to_node(C, node, 0)
for node in self.tree.traverse():
nstyle = NodeStyle()
nstyle['size'] = 0
if node.up is not None:
if set(node.sequence.upper()) == set(
'ACGT'): # Don't know what this do, try and delete
aa = translate(node.sequence)
aa_parent = translate(node.up.sequence)
nonsyn = hamming_distance(aa, aa_parent)
if '*' in aa:
nstyle['bgcolor'] = 'red'
if nonsyn > 0:
nstyle['hz_line_color'] = 'black'
nstyle['hz_line_width'] = nonsyn
else:
nstyle['hz_line_type'] = 1
node.set_style(nstyle)
ts = TreeStyle()
ts.show_leaf_name = False
ts.rotation = 90
ts.draw_aligned_faces_as_table = False
ts.allow_face_overlap = True
ts.layout_fn = my_layout
ts.show_scale = False
self.tree.render(outfile, tree_style=ts)
# If we labelled seqs, let's also write the alignment out so we have the sequences (including of internal nodes):
if idlabel:
aln = MultipleSeqAlignment([])
for node in self.tree.traverse():
aln.append(
SeqRecord(Seq(str(node.sequence), generic_dna),
id=node.name,
description='abundance={}'.format(
node.frequency)))
AlignIO.write(aln,
open(os.path.splitext(outfile)[0] + '.fasta', 'w'),
'fasta')
def renderingTreeImage(self):
path = os.path.join('Input', 'ProteinInput')
seq_records = SeqIO.parse(path, 'fasta')
for record in seq_records:
self.input_protein_accession_number.append(record.id)
self.input_protein_sequence.append(record.seq)
with open(os.path.join('execs', 'tmp',
"rooted_tree.nwk")) as nwk_tree_handle:
nwk_tree = nwk_tree_handle.read()
t = Tree(nwk_tree)
print(t)
print '\n'
ts = TreeStyle()
ts.title.add_face(TextFace(
'PhyloEpsilon - Protein Ortholog Finding Tool by Bryan Dighera',
fsize=16,
),
column=0)
ts.allow_face_overlap = True
ts.show_leaf_name = True
ts.show_branch_support = True
leaf_names = []
for leaf in t.get_leaf_names():
np_xp_pattern = re.compile('N[P]|X[P]')
digits_pattern = re.compile('\d+.\d')
np_xp_search_obj = re.search(np_xp_pattern, leaf)
digits_search_obj = re.search(digits_pattern, leaf)
np_xp_name = np_xp_search_obj.group()
digits_name = digits_search_obj.group()
final_accession = str(np_xp_name + '_' + digits_name)
print final_accession
leaf_names.append(final_accession)
#print 'leaf names: ' + '%s' % leaf_names
P = Protein()
protein_domains, domain_colors, unrepeated_domains = P.Domains()
print domain_colors
#Creates a dictionary that corresponds the protein accession number to its corresponding introns
for i in range(len(leaf_names)):
self.accession_dict_with_introns[
self.input_protein_accession_number[i]] = self.exon_lengths[i]
i = 0
print 'protein accession number: ' + '%s' % self.input_protein_accession_number
print 'Accession dict: ' + '%s' % self.accession_dict_with_introns + '\n'
#Iterates through the accession numbers that correspond the the order of the leaves of the phylogenetic tree to retrieve introns and build fig
for accession_number in leaf_names:
intron_motifs = [[0, 0, "[]", None, 12, "White", "White", None]]
#Checks the accession number against the dictionary and retrieves the corresponding introns, if no introns then doesn't append any
if accession_number in self.accession_dict_with_introns:
print accession_number, self.accession_dict_with_introns[
accession_number]
exon_list = self.accession_dict_with_introns[accession_number]
print exon_list
for exon_length in exon_list:
if str(exon_length) != 'NONE':
for location in exon_length:
split_exon_location = str(location).split('-')
protein_seq_exon_location = int(
math.floor(int(split_exon_location[1]) / 3))
#Calculates the intron phase and then checks the phase to append appropriate color indicating phase on diagram
intron_phase = (int(split_exon_location[1]) -
int(split_exon_location[0])) % 3
if intron_phase == 0:
intron_motifs.append([
protein_seq_exon_location - 2,
protein_seq_exon_location + 2, "[]", None,
5, "Grey", "Grey", None
])
elif intron_phase == 1:
intron_motifs.append([
protein_seq_exon_location - 2,
protein_seq_exon_location + 2, "[]", None,
5, "Black", "Black", None
])
elif intron_phase == 2:
intron_motifs.append([
protein_seq_exon_location - 2,
protein_seq_exon_location + 2, "[]", None,
5, "Blue", "Blue", None
])
else:
print 'NO INTRONS FOUND FOR RECORD'
print str(intron_motifs) + '\n'
msa_protein_seq = self.msa_aligned_protein[i].strip('-')
#ete3 module that adds the introns(motifs) to the phylogenetic tree
seqFace = SeqMotifFace(str(msa_protein_seq),
gapcolor="black",
seq_format='line',
scale_factor=1,
motifs=intron_motifs)
(t & t.get_leaf_names()[i]).add_face(seqFace, 0, "aligned")
i += 1
n = 0
# Iterates through the accession numbers that correspond to the order of the leaves of the phylogenetic tree and compare to domain dict values
# TODO: Add the legend and possibly give a number to each of the domains so they can be easily identified in the legend
for accession_number in leaf_names:
domain_motifs = [[0, 0, "[]", None, 12, "White", "White", None]]
for domain in protein_domains:
if accession_number in domain:
print 'leaf accession #: ' + '%s' % accession_number
print 'domains accession: ' + '%s' % domain.keys()[0]
print domain.values()[0]
for each_domain in domain.values()[0]:
try:
domain_motif_color = domain_colors[each_domain[0]]
start_domain_loc = int(
each_domain[1].split(':')[0])
end_domain_loc = int(each_domain[1].split(':')[1])
domain_name = str(each_domain[0])
domain_motifs.append([
start_domain_loc, end_domain_loc, "<>", 20, 20,
'Black', domain_motif_color, 'arial|8|black|'
])
except ValueError:
domain_motif_color = domain_colors[each_domain[0]]
start_pattern = re.compile('(?<!=\W)\d+')
start_pattern_search = re.search(
start_pattern,
str(each_domain[1].split(':')[0]))
start_domain_loc = int(
start_pattern_search.group())
end_pattern = re.compile('(?<!=\W)\d+')
end_pattern_search = re.search(
end_pattern, str(each_domain[1].split(':')[1]))
end_domain_loc = int(end_pattern_search.group())
domain_motifs.append([
start_domain_loc, end_domain_loc, "<>", 20, 20,
'Black', domain_motif_color, 'arial|8|black|'
])
print domain_motifs
msa_protein_seq = self.msa_aligned_protein[n].strip('-')
print msa_protein_seq
print len(msa_protein_seq)
print '*' * 100
domainFace = SeqMotifFace(str(msa_protein_seq),
gapcolor="black",
seq_format='line',
scale_factor=1,
motifs=domain_motifs)
(t & t.get_leaf_names()[n]).add_face(domainFace, 0, "aligned")
n += 1
#Creating the legend
print protein_domains
for single_unrepeat, colors in domain_colors.iteritems():
ts.legend.add_face(TextFace(single_unrepeat), column=0)
ts.legend.add_face(SeqMotifFace(
"A" * 45, [[0, 80, "[]", None, 8, "Black", colors, None]]),
column=1)
ts.legend_position = 1
#name_of_run = nameOfRun()
file_name = self.run_name
t.show(tree_style=ts)
t.render(os.path.join('CompletedTrees', file_name + '.pdf'),
tree_style=ts)
def render(self,
outfile,
idlabel=False,
colormap=None,
show_support=False,
chain_split=None):
'''render to image file, filetype inferred from suffix, svg for color images'''
def my_layout(node):
circle_color = 'lightgray' if colormap is None or node.name not in colormap else colormap[
node.name]
text_color = 'black'
if isinstance(circle_color, str):
C = CircleFace(radius=max(3, 10 * scipy.sqrt(node.frequency)),
color=circle_color,
label={
'text': str(node.frequency),
'color': text_color
} if node.frequency > 0 else None)
C.rotation = -90
C.hz_align = 1
faces.add_face_to_node(C, node, 0)
else:
P = PieChartFace(
[100 * x / node.frequency for x in circle_color.values()],
2 * 10 * scipy.sqrt(node.frequency),
2 * 10 * scipy.sqrt(node.frequency),
colors=[(color if color != 'None' else 'lightgray')
for color in list(circle_color.keys())],
line_color=None)
T = TextFace(' '.join(
[str(x) for x in list(circle_color.values())]),
tight_text=True)
T.hz_align = 1
T.rotation = -90
faces.add_face_to_node(P, node, 0, position='branch-right')
faces.add_face_to_node(T, node, 1, position='branch-right')
if idlabel:
T = TextFace(node.name, tight_text=True, fsize=6)
T.rotation = -90
T.hz_align = 1
faces.add_face_to_node(
T,
node,
1 if isinstance(circle_color, str) else 2,
position='branch-right')
for node in self.tree.traverse():
nstyle = NodeStyle()
nstyle['size'] = 0
if node.up is not None:
if set(node.sequence.upper()) == set('ACGT'):
if chain_split is not None:
if self.frame is not None:
raise NotImplementedError(
'frame not implemented with chain_split')
leftseq_mutated = hamming_distance(
node.sequence[:chain_split],
node.up.sequence[:chain_split]) > 0
rightseq_mutated = hamming_distance(
node.sequence[chain_split:],
node.up.sequence[chain_split:]) > 0
if leftseq_mutated and rightseq_mutated:
nstyle['hz_line_color'] = 'purple'
nstyle['hz_line_width'] = 3
elif leftseq_mutated:
nstyle['hz_line_color'] = 'red'
nstyle['hz_line_width'] = 2
elif rightseq_mutated:
nstyle['hz_line_color'] = 'blue'
nstyle['hz_line_width'] = 2
if self.frame is not None:
aa = Seq(
node.sequence[(self.frame -
1):(self.frame - 1 +
(3 *
(((len(node.sequence) -
(self.frame - 1)) // 3))))],
generic_dna).translate()
aa_parent = Seq(
node.up.sequence[(self.frame -
1):(self.frame - 1 + (3 * ((
(len(node.sequence) -
(self.frame - 1)) // 3))))],
generic_dna).translate()
nonsyn = hamming_distance(aa, aa_parent)
if '*' in aa:
nstyle['bgcolor'] = 'red'
if nonsyn > 0:
nstyle['hz_line_color'] = 'black'
nstyle['hz_line_width'] = nonsyn
else:
nstyle['hz_line_type'] = 1
node.set_style(nstyle)
ts = TreeStyle()
ts.show_leaf_name = False
ts.rotation = 90
ts.draw_aligned_faces_as_table = False
ts.allow_face_overlap = True
ts.layout_fn = my_layout
ts.show_scale = False
ts.show_branch_support = show_support
self.tree.render(outfile, tree_style=ts)
# if we labelled seqs, let's also write the alignment out so we have the sequences (including of internal nodes)
if idlabel:
aln = MultipleSeqAlignment([])
for node in self.tree.traverse():
aln.append(
SeqRecord(Seq(str(node.sequence), generic_dna),
id=str(node.name),
description='abundance={}'.format(
node.frequency)))
AlignIO.write(aln,
open(os.path.splitext(outfile)[0] + '.fasta', 'w'),
'fasta')
