def mylayout(node): # Internal Node formatting nstyle = NodeStyle() nstyle["size"] = 0 nstyle["vt_line_width"] = 2 nstyle["hz_line_width"] = 2 if node.is_leaf(): circle_face = CircleFace(10, color_dict[node.name], style='circle') faces.add_face_to_node(circle_face, node, position="branch-right", column=0) text_face = TextFace(meta_dict[node.name][0]) text_face.fgcolor = color_dict[node.name] faces.add_face_to_node(text_face, node, position="branch-right", column=1) node.set_style(nstyle) elif node.is_root(): node.dist = 0 node.set_style(nstyle) else: node.set_style(nstyle)
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
time_begin = trackResult[r, 1] time_end = trackResult[r, 2] parent_id = trackResult[r, 3] time_duration = np.abs(time_begin-time_end) # for root if parent_id == 0: # Add name to root for the first iteration root.add_feature("name", str(cell_id)) # change the branch length root.add_feature("dist", time_duration) #change node style root.set_style(ns_root) # set node name to face nameFace = TextFace(root.name) nameFace.fgcolor = "white" nameFace.fsize = 15 # nameFace.border.width = 1 nameFace.background.color = "green" node_cur.add_face(nameFace, column=1, position="branch-bottom") else: # for child #### search the parent node by parent_id node_cur = root.search_nodes(name=str(parent_id)) # there should be only one parent node if len(node_cur) == 1: #### set child with its id node_cur = node_cur[0].add_child(name=str(cell_id)) #### set duration node_cur.add_feature("dist", time_duration)
def plot_ete_tree(tree_file, ordered_queries, leaf_id2protein_id2identity, leaf_id2mlst, leaf_id2spa, leaf_id2meca, show_identity_values=True, leaf_id2description=False): mlst_list = list(set(leaf_id2mlst.values())) mlst2color = dict(zip(mlst_list, get_spaced_colors(len(mlst_list)))) mlst2color['-'] = 'white' t1 = Tree(tree_file) tss = TreeStyle() R = t1.get_midpoint_outgroup() t1.set_outgroup(R) t1.ladderize() head = True column_add = 4 for lf in t1.iter_leaves(): lf.branch_vertical_margin = 0 # add MLST if head: n = TextFace(' MLST ') n.margin_top = 2 n.margin_right = 2 n.margin_left = 2 n.margin_bottom = 2 n.rotation = 270 n.vt_align = 2 n.hz_align = 2 n.inner_background.color = "white" n.opacity = 1. tss.aligned_header.add_face(n, 1) if lf.name in leaf2mlst: n = TextFace(' %s ' % leaf_id2mlst[lf.name]) n.inner_background.color = 'white' m = TextFace(' ') m.inner_background.color = mlst2color[leaf_id2mlst[lf.name]] else: 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 = 20 m.margin_bottom = 2 lf.add_face(m, 0, position="aligned") lf.add_face(n, 1, position="aligned") # add spa typing if head: n = TextFace(' spa ') n.margin_top = 2 n.margin_right = 2 n.margin_left = 2 n.margin_bottom = 2 n.rotation = 270 n.vt_align = 2 n.hz_align = 2 n.inner_background.color = "white" n.opacity = 1. tss.aligned_header.add_face(n, column_add-2) if lf.name in leaf_id2spa: n = TextFace(' %s ' % leaf_id2spa[lf.name]) n.inner_background.color = "white" else: n = TextFace(' na ') n.inner_background.color = "grey" n.opacity = 1. n.margin_top = 2 n.margin_right = 2 n.margin_left = 2 n.margin_bottom = 2 lf.add_face(n, column_add-2, position="aligned") # add mecA typing if head: n = TextFace(' mecA ') n.margin_top = 2 n.margin_right = 2 n.margin_left = 2 n.margin_bottom = 2 n.rotation = 270 n.vt_align = 2 n.hz_align = 2 n.inner_background.color = "white" n.opacity = 1. tss.aligned_header.add_face(n, column_add-1) if lf.name in leaf_id2meca: n = TextFace(' %s ' % leaf_id2meca[lf.name]) if leaf_id2meca[lf.name] == 'Perfect': n.inner_background.color = "red" elif leaf_id2meca[lf.name] == 'Strict': n.inner_background.color = "orange" else: n.inner_background.color = "white" else: n = TextFace(' na ') n.inner_background.color = "grey" n.opacity = 1. n.margin_top = 2 n.margin_right = 2 n.margin_left = 2 n.margin_bottom = 2 lf.add_face(n, column_add-1, position="aligned") # loop to add virulence gene hits for column, protein_id in enumerate(ordered_queries): # draw labels at the top of each column if head: if show_identity_values: n = TextFace(' %s ' % str(protein_id)) n.margin_top = 2 n.margin_right = 2 n.margin_left = 2 n.margin_bottom = 2 n.rotation = 270 n.vt_align = 2 n.hz_align = 2 n.inner_background.color = "white" n.opacity = 1. tss.aligned_header.add_face(n, column+column_add) else: n = TextFace(' %s ' % str(protein_id), fsize=6) n.margin_top = 0 n.margin_right = 0 n.margin_left = 0 n.margin_bottom = 0 n.rotation = 270 n.vt_align = 2 n.hz_align = 2 n.inner_background.color = "white" n.opacity = 1. # lf.add_face(n, col, position="aligned") tss.aligned_header.add_face(n, column+column_add) # draw column content if lf.name not in leaf_id2protein_id2identity: n = TextFace(' %s ' % str(' na ')) n.opacity = 1. n.margin_top = 2 n.margin_right = 2 n.margin_left = 2 n.margin_bottom = 2 n.inner_background.color = "grey" lf.add_face(n, column+column_add, position="aligned") else: if protein_id in leaf_id2protein_id2identity[lf.name]: identity_value = float(leaf_id2protein_id2identity[lf.name][protein_id]) color = rgb2hex(m_blue.to_rgba(identity_value)) if show_identity_values: # report identity values in coloured boxes # adapt box size depending the digit width if str(identity_value) == '100.00' or str(identity_value) == '100.0': identity_value = '100' n = TextFace(" %s " % identity_value) else: n = TextFace("%.2f" % round(float(identity_value), 2)) # color text to white for dark cells if float(identity_value) > 95: n.fgcolor = "white" n.opacity = 1. n.margin_top = 2 n.margin_right = 2 n.margin_left = 2 n.margin_bottom = 2 n.inner_background.color = color lf.add_face(n, column+column_add, position="aligned") else: # draw coloured boxes without text n = TextFace(' ') n.margin_top = 0 n.margin_right = 0 n.margin_left = 0 n.margin_bottom = 0 # n.color = color n.inner_background.color = color lf.add_face(n, column+column_add, position="aligned") else: n = TextFace(' %s ' % str(' - ')) n.opacity = 1. n.margin_top = 2 n.margin_right = 2 n.margin_left = 2 n.margin_bottom = 2 n.inner_background.color = "white" lf.add_face(n, column+column_add, position="aligned") # end of first leaf: turn off header head = False # add boostrap supports for n in t1.traverse(): nstyle = NodeStyle() if n.support < 0.9: nstyle["fgcolor"] = "blue" nstyle["size"] = 6 n.set_style(nstyle) else: nstyle["fgcolor"] = "red" nstyle["size"] = 0 n.set_style(nstyle) return t1, tss
time_begin = trackResult[r, 1] time_end = trackResult[r, 2] parent_id = trackResult[r, 3] time_duration = np.abs(time_begin - time_end) # for root if parent_id == 0: # Add name to root for the first iteration root.add_feature("name", str(cell_id)) # change the branch length root.add_feature("dist", time_duration) #change node style root.set_style(ns_root) # set node name to face nameFace = TextFace(root.name) nameFace.fgcolor = "white" nameFace.fsize = 15 # nameFace.border.width = 1 nameFace.background.color = "green" node_cur.add_face(nameFace, column=1, position="branch-bottom") else: # for child #### search the parent node by parent_id node_cur = root.search_nodes(name=str(parent_id)) # there should be only one parent node if len(node_cur) == 1: #### set child with its id node_cur = node_cur[0].add_child(name=str(cell_id)) #### set duration node_cur.add_feature("dist", time_duration)
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_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