def get_colors(number_of_colors):
return get_list_of_colors(number_of_colors, colormap="Dark2")
def entropy_distribution_bar(alignment, entropy_values, output_file, quick = False, no_display = False, qual_stats_dict = None, weighted = False, verbose = False):
progress.verbose = verbose
progress.new('Entropy Distribution Figure')
progress.update('Computing ')
y_maximum = max(entropy_values) + (max(entropy_values) / 10.0)
y_maximum = 1 if y_maximum < 1 else y_maximum
number_of_uniques_to_show = int(y_maximum * 100)
if alignment == None:
quick = True
colors_dict = {}
if not quick:
unique_sequences = get_unique_sequences_from_FASTA(alignment, limit = number_of_uniques_to_show)
chars = []
for seq in unique_sequences:
chars += seq[0]
chars = set(chars)
colors_dict = NUCL_COLORS
missing_chars = [char for char in chars if char not in NUCL_COLORS.keys()]
if missing_chars:
colors_for_missing_chars = get_list_of_colors(len(missing_chars), colormap="RdYlGn")
for i in range(0, len(missing_chars)):
char = missing_chars[i]
colors_dict[char] = colors_for_missing_chars[i]
else:
unique_sequences = None
fig = plt.figure(figsize = (len(entropy_values) / 20, 10))
plt.rcParams.update({'axes.linewidth' : 0.1})
plt.rc('grid', color='0.70', linestyle='-', linewidth=0.1)
plt.grid(True)
plt.subplots_adjust(hspace = 0, wspace = 0, right = 0.995, left = 0.050, top = 0.92, bottom = 0.10)
ax = fig.add_subplot(111)
if not quick:
current = 0
for y in range(number_of_uniques_to_show - 1, 0, -3):
progress.append('.')
unique_sequence = unique_sequences[current][0].upper()
count = unique_sequences[current][1]
frequency = unique_sequences[current][2]
for i in range(0, len(unique_sequence)):
plt.text(i, y / 100.0, unique_sequence[i],\
fontsize = 5, color = colors_dict[unique_sequence[i]])
percent = int(round(frequency * len(unique_sequence))) or 1
plt.fill_between(range(0, percent), (y + 1.15) / 100.0, (y - 0.85) / 100.0, color="green", alpha = 0.2)
plt.text(percent + 0.8, (y - 1.2) / 100.0, count, fontsize = 5, color = 'gray')
current += 1
if current + 1 > len(unique_sequences):
break
if not quick and qual_stats_dict:
# add mean quality values in the background of the figure.
colors = get_list_of_colors(21, colormap="RdYlGn")
colors = [colors[0] for _ in range(0, 20)] + colors
max_count = max([qual_stats_dict[q]['count'] for q in qual_stats_dict if qual_stats_dict[q]])
for pos in range(0, len(entropy_values)):
if not qual_stats_dict[pos]:
continue
mean = int(round(qual_stats_dict[pos]['mean']))
count = qual_stats_dict[pos]['count']
plt.fill_between([pos, pos + 1], y1 = 0, y2 = y_maximum, color = colors[mean], alpha = (log(count) / log(max_count)) / 5)
ind = np.arange(len(entropy_values))
ax.bar(ind, entropy_values, color = 'black', lw = 0.5)
ax.set_xlim([0, len(entropy_values)])
ax.set_ylim([0, y_maximum])
plt.xlabel('Position in the Alignment')
if weighted:
plt.ylabel('Weighted Shannon Entropy')
else:
plt.ylabel('Shannon Entropy')
progress.update('Saving into "%s"' % output_file)
plt.savefig(output_file + '.png')
plt.savefig(output_file + '.pdf')
if verbose:
progress.reset()
run.info('Entropy figure output path', output_file + '.{png, pdf}')
if not no_display:
try:
progress.update('Entropy figure is being shown (you do not have display? you can avoid this step by using --no-display))')
plt.show()
except:
pass
progress.end()
def get_colors(number_of_colors):
return get_list_of_colors(number_of_colors, colormap="Dark2")
def generate_html_output(run_info_dict,
html_output_directory=None,
entropy_figure=None):
if not html_output_directory:
html_output_directory = os.path.join(run_info_dict['output_directory'],
'HTML-OUTPUT')
if not os.path.exists(html_output_directory):
os.makedirs(html_output_directory)
html_dict = copy.deepcopy(run_info_dict)
shutil.copy2(os.path.join(absolute, 'static/style.css'),
os.path.join(html_output_directory, 'style.css'))
shutil.copy2(os.path.join(absolute, 'static/header_1.png'),
os.path.join(html_output_directory, 'header.png'))
shutil.copy2(os.path.join(absolute, 'static/missing_image.png'),
os.path.join(html_output_directory, 'missing.png'))
shutil.copy2(os.path.join(absolute, 'static/colorbar.png'),
os.path.join(html_output_directory, 'colorbar.png'))
shutil.copy2(os.path.join(absolute, 'scripts/jquery-1.7.1.js'),
os.path.join(html_output_directory, 'jquery-1.7.1.js'))
shutil.copy2(os.path.join(absolute, 'scripts/popup.js'),
os.path.join(html_output_directory, 'popup.js'))
shutil.copy2(os.path.join(absolute, 'scripts/g.pie.js'),
os.path.join(html_output_directory, 'g.pie.js'))
shutil.copy2(os.path.join(absolute, 'scripts/g.raphael.js'),
os.path.join(html_output_directory, 'g.raphael.js'))
shutil.copy2(os.path.join(absolute, 'scripts/raphael.js'),
os.path.join(html_output_directory, 'raphael.js'))
shutil.copy2(os.path.join(absolute, 'scripts/morris.js'),
os.path.join(html_output_directory, 'morris.js'))
def copy_as(source, dest_name, essential=True):
dest = os.path.join(html_output_directory, dest_name)
if essential:
shutil.copy2(source, dest)
else:
# it is ok if you fail to copy files that are not
# essential..
try:
shutil.copy2(source, dest)
except:
sys.stderr.write(
'\n\n[HTML] Warning: Source file not found\n\tSource: "%s"\n\tDest: "%s\n\n"'
% (source, dest))
return os.path.basename(dest)
# embarrassingly ad-hoc:
if entropy_figure:
if entropy_figure.endswith('.pdf') or entropy_figure.endswith('.png'):
entropy_figure = entropy_figure[:-4]
CP = lambda e, o: copy_as(os.path.join(e + ('.%s' % ext)),
o,
essential=True if ext == 'png' else False)
for ext in ['png', 'pdf']:
output_file = 'entropy.%s' % ext
if entropy_figure:
html_dict['entropy_figure_%s' % ext] = CP(entropy_figure,
output_file)
else:
try:
html_dict['entropy_figure_%s' % ext] = CP(
run_info_dict['entropy'], output_file)
except:
html_dict['entropy_figure_%s' % ext] = CP(
run_info_dict['entropy'][:-4], output_file)
if run_info_dict['gexf_network_file_path']:
html_dict['gexf_network_file_path'] = copy_as(
run_info_dict['gexf_network_file_path'], 'network.gexf')
if run_info_dict['sample_mapping']:
html_dict['sample_mapping'] = copy_as(run_info_dict['sample_mapping'],
'sample_mapping.txt')
else:
html_dict['sample_mapping'] = None
html_dict['matrix_count_file_path'] = copy_as(
run_info_dict['matrix_count_file_path'], 'matrix_counts.txt')
html_dict['matrix_percent_file_path'] = copy_as(
run_info_dict['matrix_percent_file_path'], 'matrix_percents.txt')
html_dict['read_distribution_table_path'] = copy_as(
run_info_dict['read_distribution_table_path'], 'read_distribution.txt')
html_dict['environment_file_path'] = copy_as(
run_info_dict['environment_file_path'], 'environment.txt')
html_dict['oligos_fasta_file_path'] = copy_as(
run_info_dict['oligos_fasta_file_path'], 'oligos.fa.txt')
html_dict['oligos_nexus_file_path'] = copy_as(
run_info_dict['oligos_nexus_file_path'], 'oligos.nex.txt')
def get_figures_dict(html_dict_prefix):
html_dict_key = '%s_file_path' % html_dict_prefix
if html_dict.has_key(html_dict_key):
figures_dict = cPickle.load(open(html_dict[html_dict_key]))
for _map in figures_dict:
for _func in figures_dict[_map]:
for _op in figures_dict[_map][_func]:
if os.path.exists(figures_dict[_map][_func][_op] +
'.pdf') and os.path.exists(
figures_dict[_map][_func][_op] +
'.png'):
prefix = copy_as(
figures_dict[_map][_func][_op] + '.pdf',
'%s.pdf' % '-'.join([_map, _func, _op]))
prefix = copy_as(
figures_dict[_map][_func][_op] + '.png',
'%s.png' % '-'.join([_map, _func, _op]))
figures_dict[_map][_func][_op] = '.'.join(
prefix.split('.')[:-1])
else:
figures_dict[_map][_func][_op] = None
return figures_dict
else:
return None
html_dict['figures_dict'] = get_figures_dict('figures_dict')
html_dict['exclusive_figures_dict'] = get_figures_dict(
'exclusive_figures_dict')
if html_dict['generate_sets']:
html_dict['across_samples_MN_file_path'] = copy_as(
run_info_dict['across_samples_MN_file_path'],
'across_samples_max_normalized.txt')
html_dict['across_samples_SN_file_path'] = copy_as(
run_info_dict['across_samples_SN_file_path'],
'across_samples_sum_normalized.txt')
html_dict['oligo_sets_stackbar_figure'] = copy_as(
run_info_dict['stack_bar_with_agglomerated_oligos_file_path'],
'stackbar_with_oligo_sets.png')
html_dict['oligos_across_samples_figure'] = copy_as(
run_info_dict['oligos_across_samples_file_path'],
'oligos_across_samples.png')
html_dict['oligotype_sets_figure'] = copy_as(
run_info_dict['oligotype_sets_across_samples_figure_path'],
'oligotype_sets.png')
html_dict['matrix_count_oligo_sets_file_path'] = copy_as(
run_info_dict['matrix_count_oligo_sets_file_path'],
'matrix_counts_oligo_sets.txt')
html_dict['matrix_percent_oligo_sets_file_path'] = copy_as(
run_info_dict['matrix_percent_oligo_sets_file_path'],
'matrix_percents_oligo_sets.txt')
html_dict['oligotype_sets_file'] = copy_as(
run_info_dict['oligotype_sets_file_path'], 'oligotype_sets.txt')
html_dict['oligotype_sets'] = [
l.strip().split('\t')[1].split(',')
for l in open(run_info_dict['oligotype_sets_file_path'])
]
if html_dict.has_key('representative_seqs_fasta_file_path'):
html_dict['representative_seqs_fasta_file_path'] = copy_as(
run_info_dict['representative_seqs_fasta_file_path'],
'oligo-representatives.fa.txt')
else:
html_dict['representative_seqs_fasta_file_path'] = None
if run_info_dict.has_key('blast_ref_db') and os.path.exists(
run_info_dict['blast_ref_db']):
html_dict['blast_ref_db_path'] = copy_as(run_info_dict['blast_ref_db'],
'reference_db.fa')
html_dict['entropy_components'] = [
int(x) for x in html_dict['bases_of_interest_locs'].split(',')
]
html_dict['samples_dict'] = get_samples_dict_from_environment_file(
run_info_dict['environment_file_path'])
html_dict['samples'] = sorted(html_dict['samples_dict'].keys())
html_dict['blast_results_found'] = False
# get alignment length
html_dict['alignment_length'] = get_alignment_length(
run_info_dict['alignment'])
# include pretty names
html_dict['pretty_names'] = pretty_names
# get purity score colors dict
html_dict['score_color_dict'] = {}
gradient = get_list_of_colors(26, colormap='RdYlGn')
for oligo in run_info_dict['final_purity_score_dict']:
html_dict['score_color_dict'][oligo] = gradient[int(
run_info_dict['final_purity_score_dict'][oligo] * 25)]
# get total purity score color dict
html_dict['total_score_color'] = gradient[int(
float(run_info_dict['total_purity_score_dict']) * 25)]
# get colors dict
html_dict['color_dict'] = get_colors_dict(
run_info_dict['colors_file_path'])
# get abundant oligos list
html_dict['oligos'] = get_oligos_list(
run_info_dict['oligos_fasta_file_path'])
# get oligo frequencies
html_dict['frequency'] = {}
for oligo in html_dict['oligos']:
html_dict['frequency'][oligo] = pretty_print(
sum([
d[oligo] for d in html_dict['samples_dict'].values()
if d.has_key(oligo)
]))
# get purity score
html_dict['purity_score'] = run_info_dict['final_purity_score_dict']
# get total purity score
html_dict['total_purity_score'] = run_info_dict['total_purity_score_dict']
# get unique sequence dict (which will contain the most frequent unique sequence for given oligotype)
if html_dict.has_key('output_directory_for_reps'):
html_dict['rep_oligo_seqs_clean_dict'], html_dict[
'rep_oligo_seqs_fancy_dict'] = get_unique_sequences_dict(html_dict)
html_dict['oligo_reps_dict'] = get_oligo_reps_dict(
html_dict, html_output_directory)
html_dict['component_reference'] = ''.join([
'<a onmouseover="popup(\'\#%d\', 50)" href="">|</a>' % i
for i in range(0, html_dict['alignment_length'])
])
# get javascript code for sample pie-charts
html_dict['pie_charts_js'] = render_to_string('pie_charts_js.tmpl',
html_dict)
# FIXME: code below is very inefficient and causes a huge
# memory issue. fix it by not using deepcopy.
# generate individual oligotype pages
if html_dict.has_key('output_directory_for_reps'):
for i in range(0, len(html_dict['oligos'])):
oligo = html_dict['oligos'][i]
tmp_dict = copy.deepcopy(html_dict)
tmp_dict['oligo'] = oligo
tmp_dict['distribution'] = get_oligo_distribution_dict(
oligo, html_dict)
oligo_page = os.path.join(html_output_directory,
'oligo_%s.html' % oligo)
tmp_dict['index'] = i + 1
tmp_dict['total'] = len(html_dict['oligos'])
tmp_dict['prev'] = None
tmp_dict['next'] = None
if i > 0:
tmp_dict['prev'] = 'oligo_%s.html' % html_dict['oligos'][i - 1]
if i < (len(html_dict['oligos']) - 1):
tmp_dict['next'] = 'oligo_%s.html' % html_dict['oligos'][i + 1]
rendered = render_to_string('single_oligo.tmpl', tmp_dict)
open(oligo_page, 'w').write(rendered.encode("utf-8"))
# generate index
index_page = os.path.join(html_output_directory, 'index.html')
rendered = render_to_string('index_for_oligo.tmpl', html_dict)
open(index_page, 'w').write(rendered.encode("utf-8"))
return index_page
def entropy_distribution_bar(alignment, entropy_values, output_file, quick = False, no_display = False, qual_stats_dict = None, weighted = False, verbose = False):
progress.verbose = verbose
progress.new('Entropy Distribution Figure')
progress.update('Computing ')
y_maximum = max(entropy_values) + (max(entropy_values) / 10.0)
y_maximum = 1 if y_maximum < 1 else y_maximum
number_of_uniques_to_show = int(y_maximum * 100)
if alignment == None:
quick = True
colors_dict = {}
if not quick:
unique_sequences = get_unique_sequences_from_FASTA(alignment, limit = number_of_uniques_to_show)
chars = []
for seq in unique_sequences:
chars += seq[0]
chars = set(chars)
colors_dict = NUCL_COLORS
missing_chars = [char for char in chars if char not in NUCL_COLORS.keys()]
if missing_chars:
colors_for_missing_chars = get_list_of_colors(len(missing_chars), colormap="RdYlGn")
for i in range(0, len(missing_chars)):
char = missing_chars[i]
colors_dict[char] = colors_for_missing_chars[i]
else:
unique_sequences = None
fig = plt.figure(figsize = (len(entropy_values) / 20, 10))
plt.rcParams.update({'axes.linewidth' : 0.1})
plt.rc('grid', color='0.70', linestyle='-', linewidth=0.1)
plt.grid(True)
plt.subplots_adjust(hspace = 0, wspace = 0, right = 0.995, left = 0.050, top = 0.92, bottom = 0.10)
ax = fig.add_subplot(111)
if not quick:
current = 0
for y in range(number_of_uniques_to_show - 1, 0, -3):
progress.append('.')
unique_sequence = unique_sequences[current][0].upper()
count = unique_sequences[current][1]
frequency = unique_sequences[current][2]
for i in range(0, len(unique_sequence)):
plt.text(i, y / 100.0, unique_sequence[i],\
fontsize = 5, color = colors_dict[unique_sequence[i]])
percent = int(round(frequency * len(unique_sequence))) or 1
plt.fill_between(range(0, percent), (y + 1.15) / 100.0, (y - 0.85) / 100.0, color="green", alpha = 0.2)
plt.text(percent + 0.8, (y - 1.2) / 100.0, count, fontsize = 5, color = 'gray')
current += 1
if current + 1 > len(unique_sequences):
break
if not quick and qual_stats_dict:
# add mean quality values in the background of the figure.
colors = get_list_of_colors(21, colormap="RdYlGn")
colors = [colors[0] for _ in range(0, 20)] + colors
max_count = max([qual_stats_dict[q]['count'] for q in qual_stats_dict if qual_stats_dict[q]])
for pos in range(0, len(entropy_values)):
if not qual_stats_dict[pos]:
continue
mean = int(round(qual_stats_dict[pos]['mean']))
count = qual_stats_dict[pos]['count']
plt.fill_between([pos, pos + 1], y1 = 0, y2 = y_maximum, color = colors[mean], alpha = (log(count) / log(max_count)) / 5)
ind = np.arange(len(entropy_values))
ax.bar(ind, entropy_values, color = 'black', lw = 0.5)
ax.set_xlim([0, len(entropy_values)])
ax.set_ylim([0, y_maximum])
plt.xlabel('Position in the Alignment')
if weighted:
plt.ylabel('Weighted Shannon Entropy')
else:
plt.ylabel('Shannon Entropy')
progress.update('Saving into "%s"' % output_file)
plt.savefig(output_file + '.png')
plt.savefig(output_file + '.pdf')
if verbose:
progress.clear()
run.info('Entropy figure output path', output_file + '.{png, pdf}')
if not no_display:
try:
progress.update('Entropy figure is being shown (you do not have display? you can avoid this step by using --no-display))')
plt.show()
except:
pass
progress.end()
def generate_html_output(run_info_dict, html_output_directory = None, entropy_figure = None):
if not html_output_directory:
html_output_directory = os.path.join(run_info_dict['output_directory'], 'HTML-OUTPUT')
if not os.path.exists(html_output_directory):
os.makedirs(html_output_directory)
html_dict = copy.deepcopy(run_info_dict)
shutil.copy2(os.path.join(absolute, 'static/style.css'), os.path.join(html_output_directory, 'style.css'))
shutil.copy2(os.path.join(absolute, 'static/header_1.png'), os.path.join(html_output_directory, 'header.png'))
shutil.copy2(os.path.join(absolute, 'static/missing_image.png'), os.path.join(html_output_directory, 'missing.png'))
shutil.copy2(os.path.join(absolute, 'static/colorbar.png'), os.path.join(html_output_directory, 'colorbar.png'))
shutil.copy2(os.path.join(absolute, 'scripts/jquery-1.7.1.js'), os.path.join(html_output_directory, 'jquery-1.7.1.js'))
shutil.copy2(os.path.join(absolute, 'scripts/popup.js'), os.path.join(html_output_directory, 'popup.js'))
shutil.copy2(os.path.join(absolute, 'scripts/g.pie.js'), os.path.join(html_output_directory, 'g.pie.js'))
shutil.copy2(os.path.join(absolute, 'scripts/g.raphael.js'), os.path.join(html_output_directory, 'g.raphael.js'))
shutil.copy2(os.path.join(absolute, 'scripts/raphael.js'), os.path.join(html_output_directory, 'raphael.js'))
shutil.copy2(os.path.join(absolute, 'scripts/morris.js'), os.path.join(html_output_directory, 'morris.js'))
def copy_as(source, dest_name, essential = True):
dest = os.path.join(html_output_directory, dest_name)
if essential:
shutil.copy2(source, dest)
else:
# it is ok if you fail to copy files that are not
# essential..
try:
shutil.copy2(source, dest)
except:
sys.stderr.write('\n\n[HTML] Warning: Source file not found\n\tSource: "%s"\n\tDest: "%s\n\n"' % (source, dest))
return os.path.basename(dest)
# embarrassingly ad-hoc:
if entropy_figure:
if entropy_figure.endswith('.pdf') or entropy_figure.endswith('.png'):
entropy_figure = entropy_figure[:-4]
CP = lambda e, o: copy_as(os.path.join(e + ('.%s' % ext)), o, essential = True if ext == 'png' else False)
for ext in ['png', 'pdf']:
output_file = 'entropy.%s' % ext
if entropy_figure:
html_dict['entropy_figure_%s' % ext] = CP(entropy_figure, output_file)
else:
try:
html_dict['entropy_figure_%s' % ext] = CP(run_info_dict['entropy'], output_file)
except:
html_dict['entropy_figure_%s' % ext] = CP(run_info_dict['entropy'][:-4], output_file)
if run_info_dict['gexf_network_file_path']:
html_dict['gexf_network_file_path'] = copy_as(run_info_dict['gexf_network_file_path'], 'network.gexf')
if run_info_dict['sample_mapping']:
html_dict['sample_mapping'] = copy_as(run_info_dict['sample_mapping'], 'sample_mapping.txt')
else:
html_dict['sample_mapping'] = None
html_dict['matrix_count_file_path'] = copy_as(run_info_dict['matrix_count_file_path'], 'matrix_counts.txt')
html_dict['matrix_percent_file_path'] = copy_as(run_info_dict['matrix_percent_file_path'], 'matrix_percents.txt')
html_dict['read_distribution_table_path'] = copy_as(run_info_dict['read_distribution_table_path'], 'read_distribution.txt')
html_dict['environment_file_path'] = copy_as(run_info_dict['environment_file_path'], 'environment.txt')
html_dict['oligos_fasta_file_path'] = copy_as(run_info_dict['oligos_fasta_file_path'], 'oligos.fa.txt')
html_dict['oligos_nexus_file_path'] = copy_as(run_info_dict['oligos_nexus_file_path'], 'oligos.nex.txt')
def get_figures_dict(html_dict_prefix):
html_dict_key = '%s_file_path' % html_dict_prefix
if html_dict.has_key(html_dict_key):
figures_dict = cPickle.load(open(html_dict[html_dict_key]))
for _map in figures_dict:
for _func in figures_dict[_map]:
for _op in figures_dict[_map][_func]:
if os.path.exists(figures_dict[_map][_func][_op] + '.pdf') and os.path.exists(figures_dict[_map][_func][_op] + '.png'):
prefix = copy_as(figures_dict[_map][_func][_op] + '.pdf', '%s.pdf' % '-'.join([_map, _func, _op]))
prefix = copy_as(figures_dict[_map][_func][_op] + '.png', '%s.png' % '-'.join([_map, _func, _op]))
figures_dict[_map][_func][_op] = '.'.join(prefix.split('.')[:-1])
else:
figures_dict[_map][_func][_op] = None
return figures_dict
else:
return None
html_dict['figures_dict'] = get_figures_dict('figures_dict')
html_dict['exclusive_figures_dict'] = get_figures_dict('exclusive_figures_dict')
if html_dict['generate_sets']:
html_dict['across_samples_MN_file_path'] = copy_as(run_info_dict['across_samples_MN_file_path'], 'across_samples_max_normalized.txt')
html_dict['across_samples_SN_file_path'] = copy_as(run_info_dict['across_samples_SN_file_path'], 'across_samples_sum_normalized.txt')
html_dict['oligo_sets_stackbar_figure'] = copy_as(run_info_dict['stack_bar_with_agglomerated_oligos_file_path'], 'stackbar_with_oligo_sets.png')
html_dict['oligos_across_samples_figure'] = copy_as(run_info_dict['oligos_across_samples_file_path'], 'oligos_across_samples.png')
html_dict['oligotype_sets_figure'] = copy_as(run_info_dict['oligotype_sets_across_samples_figure_path'], 'oligotype_sets.png')
html_dict['matrix_count_oligo_sets_file_path'] = copy_as(run_info_dict['matrix_count_oligo_sets_file_path'], 'matrix_counts_oligo_sets.txt')
html_dict['matrix_percent_oligo_sets_file_path'] = copy_as(run_info_dict['matrix_percent_oligo_sets_file_path'], 'matrix_percents_oligo_sets.txt')
html_dict['oligotype_sets_file'] = copy_as(run_info_dict['oligotype_sets_file_path'], 'oligotype_sets.txt')
html_dict['oligotype_sets'] = [l.strip().split('\t')[1].split(',') for l in open(run_info_dict['oligotype_sets_file_path'])]
if html_dict.has_key('representative_seqs_fasta_file_path'):
html_dict['representative_seqs_fasta_file_path'] = copy_as(run_info_dict['representative_seqs_fasta_file_path'], 'oligo-representatives.fa.txt')
else:
html_dict['representative_seqs_fasta_file_path'] = None
if run_info_dict.has_key('blast_ref_db') and os.path.exists(run_info_dict['blast_ref_db']):
html_dict['blast_ref_db_path'] = copy_as(run_info_dict['blast_ref_db'], 'reference_db.fa')
html_dict['entropy_components'] = [int(x) for x in html_dict['bases_of_interest_locs'].split(',')]
html_dict['samples_dict'] = get_samples_dict_from_environment_file(run_info_dict['environment_file_path'])
html_dict['samples'] = sorted(html_dict['samples_dict'].keys())
html_dict['blast_results_found'] = False
# get alignment length
html_dict['alignment_length'] = get_alignment_length(run_info_dict['alignment'])
# include pretty names
html_dict['pretty_names'] = pretty_names
# get purity score colors dict
html_dict['score_color_dict'] = {}
gradient = get_list_of_colors(26, colormap = 'RdYlGn')
for oligo in run_info_dict['final_purity_score_dict']:
html_dict['score_color_dict'][oligo] = gradient[int(run_info_dict['final_purity_score_dict'][oligo] * 25)]
# get total purity score color dict
html_dict['total_score_color'] = gradient[int(float(run_info_dict['total_purity_score_dict']) * 25)]
# get colors dict
html_dict['color_dict'] = get_colors_dict(run_info_dict['colors_file_path'])
# get abundant oligos list
html_dict['oligos'] = get_oligos_list(run_info_dict['oligos_fasta_file_path'])
# get oligo frequencies
html_dict['frequency'] = {}
for oligo in html_dict['oligos']:
html_dict['frequency'][oligo] = pretty_print(sum([d[oligo] for d in html_dict['samples_dict'].values() if d.has_key(oligo)]))
# get purity score
html_dict['purity_score'] = run_info_dict['final_purity_score_dict']
# get total purity score
html_dict['total_purity_score'] = run_info_dict['total_purity_score_dict']
# get unique sequence dict (which will contain the most frequent unique sequence for given oligotype)
if html_dict.has_key('output_directory_for_reps'):
html_dict['rep_oligo_seqs_clean_dict'], html_dict['rep_oligo_seqs_fancy_dict'] = get_unique_sequences_dict(html_dict)
html_dict['oligo_reps_dict'] = get_oligo_reps_dict(html_dict, html_output_directory)
html_dict['component_reference'] = ''.join(['<a onmouseover="popup(\'\#%d\', 50)" href="">|</a>' % i for i in range(0, html_dict['alignment_length'])])
# get javascript code for sample pie-charts
html_dict['pie_charts_js'] = render_to_string('pie_charts_js.tmpl', html_dict)
# FIXME: code below is very inefficient and causes a huge
# memory issue. fix it by not using deepcopy.
# generate individual oligotype pages
if html_dict.has_key('output_directory_for_reps'):
for i in range(0, len(html_dict['oligos'])):
oligo = html_dict['oligos'][i]
tmp_dict = copy.deepcopy(html_dict)
tmp_dict['oligo'] = oligo
tmp_dict['distribution'] = get_oligo_distribution_dict(oligo, html_dict)
oligo_page = os.path.join(html_output_directory, 'oligo_%s.html' % oligo)
tmp_dict['index'] = i + 1
tmp_dict['total'] = len(html_dict['oligos'])
tmp_dict['prev'] = None
tmp_dict['next'] = None
if i > 0:
tmp_dict['prev'] = 'oligo_%s.html' % html_dict['oligos'][i - 1]
if i < (len(html_dict['oligos']) - 1):
tmp_dict['next'] = 'oligo_%s.html' % html_dict['oligos'][i + 1]
rendered = render_to_string('single_oligo.tmpl', tmp_dict)
open(oligo_page, 'w').write(rendered.encode("utf-8"))
# generate index
index_page = os.path.join(html_output_directory, 'index.html')
rendered = render_to_string('index_for_oligo.tmpl', html_dict)
open(index_page, 'w').write(rendered.encode("utf-8"))
return index_page
quals_dict_filtered = {}
ids_in_alignment_file = []
while alignment.next():
ids_in_alignment_file.append(alignment.id)
ids_in_alignment_file = set(ids_in_alignment_file)
for read_id in quals_dict:
if read_id in ids_in_alignment_file:
quals_dict_filtered[read_id] = quals_dict[read_id]
ids_in_alignment_file.remove(read_id)
qual_stats_dict = get_qual_stats_dict(quals_dict_filtered)
colors = get_list_of_colors(21, colormap="RdYlGn")
colors = [colors[0] for _ in range(0, 20)] + colors
max_count = max([
qual_stats_dict[q]['count'] for q in qual_stats_dict if qual_stats_dict[q]
])
alignment_length = len(quals_dict.values()[0])
fig = plt.figure(figsize=(25, 8))
plt.rc('grid', color='0.50', linestyle='-', linewidth=0.1)
plt.grid(True)
plt.subplots_adjust(left=0.02, bottom=0.09, top=0.95, right=0.98)
for position in range(0, alignment_length):
print position
def vis_oligotype_sets_distribution(partitions, vectors, samples, colors_dict = None, output_file = None, legend = False, project_title = None, display = True):
if colors_dict == None:
colors_dict = {}
list_of_colors = get_list_of_colors(len(partitions), colormap = 'Accent')
for i in range(0, len(partitions)):
colors_dict[partitions[i][0]] = list_of_colors[i]
#for oligo in partitions[i]:
# print '%s,%s' % (oligo, list_of_colors[i])
# figure..
plt.figure(figsize=(20, 7))
if legend:
plt.subplots_adjust(left=0.03, bottom = 0.25, top = 0.95, right = 0.87)
else:
plt.subplots_adjust(left=0.03, bottom = 0.25, top = 0.95, right = 0.99)
plt.rcParams.update({'axes.linewidth' : 0.1})
plt.rc('grid', color='0.70', linestyle='-', linewidth=0.1)
plt.grid(True)
N = len(samples)
ind = np.arange(N)
width = 0.75
number_of_dimensions = len(vectors.values()[0])
for i in range(0, len(partitions)):
group = partitions[i]
vector = []
mins = []
maxs = []
for d in range(0, number_of_dimensions):
vector.append(np.mean([vectors[oligo][d] for oligo in group]))
mins.append(np.min([vectors[oligo][d] for oligo in group]))
maxs.append(np.max([vectors[oligo][d] for oligo in group]))
try:
color = HTMLColorToRGB(colors_dict[group[0]])
except:
color = 'black'
plt.fill_between(range(0, len(vector)), maxs, mins, color=color, alpha = 0.1)
plt.plot(vector, color=color, linewidth = 1, alpha = 0.95, label = 'Set #%d' % i)
if len(vector) < 50:
plt.plot(vector, color=color, linewidth = 3, alpha = 0.7, label = '_nolegend_')
plt.plot(vector, color=color, linewidth = 7, alpha = 0.6, zorder = i, label = '_nolegend_')
plt.ylabel('Oligotype Set Abundance', size='large')
plt.title(project_title if project_title else 'Oligotype Sets Across Samples')
plt.xticks(ind, samples, rotation=90, size='small')
plt.yticks([])
plt.ylim(ymax = 100)
plt.xlim(xmin = -(width) / 2, xmax = len(samples) - 0.5)
if legend:
plt.legend(bbox_to_anchor=(1.01, 1), loc=2, borderaxespad=0.0, shadow=True, fancybox=True)
leg = plt.gca().get_legend()
ltext = leg.get_texts()
llines = leg.get_lines()
frame = leg.get_frame()
frame.set_facecolor('0.80')
plt.setp(ltext, fontsize='small', fontname='arial', family='monospace')
plt.setp(llines, linewidth=1.5)
if output_file:
plt.savefig(output_file)
if display:
try:
plt.show()
except:
pass
def vis_oligotype_sets_distribution(partitions,
vectors,
samples,
colors_dict=None,
output_file=None,
legend=False,
project_title=None,
display=True):
if colors_dict == None:
colors_dict = {}
list_of_colors = get_list_of_colors(len(partitions), colormap='Accent')
for i in range(0, len(partitions)):
colors_dict[partitions[i][0]] = list_of_colors[i]
#for oligo in partitions[i]:
# print '%s,%s' % (oligo, list_of_colors[i])
# figure..
plt.figure(figsize=(20, 7))
if legend:
plt.subplots_adjust(left=0.03, bottom=0.25, top=0.95, right=0.87)
else:
plt.subplots_adjust(left=0.03, bottom=0.25, top=0.95, right=0.99)
plt.rcParams.update({'axes.linewidth': 0.1})
plt.rc('grid', color='0.70', linestyle='-', linewidth=0.1)
plt.grid(True)
N = len(samples)
ind = np.arange(N)
width = 0.75
number_of_dimensions = len(vectors.values()[0])
for i in range(0, len(partitions)):
group = partitions[i]
vector = []
mins = []
maxs = []
for d in range(0, number_of_dimensions):
vector.append(np.mean([vectors[oligo][d] for oligo in group]))
mins.append(np.min([vectors[oligo][d] for oligo in group]))
maxs.append(np.max([vectors[oligo][d] for oligo in group]))
try:
color = HTMLColorToRGB(colors_dict[group[0]])
except:
color = 'black'
plt.fill_between(range(0, len(vector)),
maxs,
mins,
color=color,
alpha=0.1)
plt.plot(vector,
color=color,
linewidth=1,
alpha=0.95,
label='Set #%d' % i)
if len(vector) < 50:
plt.plot(vector,
color=color,
linewidth=3,
alpha=0.7,
label='_nolegend_')
plt.plot(vector,
color=color,
linewidth=7,
alpha=0.6,
zorder=i,
label='_nolegend_')
plt.ylabel('Oligotype Set Abundance', size='large')
plt.title(
project_title if project_title else 'Oligotype Sets Across Samples')
plt.xticks(ind, samples, rotation=90, size='small')
plt.yticks([])
plt.ylim(ymax=100)
plt.xlim(xmin=-(width) / 2, xmax=len(samples) - 0.5)
if legend:
plt.legend(bbox_to_anchor=(1.01, 1),
loc=2,
borderaxespad=0.0,
shadow=True,
fancybox=True)
leg = plt.gca().get_legend()
ltext = leg.get_texts()
llines = leg.get_lines()
frame = leg.get_frame()
frame.set_facecolor('0.80')
plt.setp(ltext, fontsize='small', fontname='arial', family='monospace')
plt.setp(llines, linewidth=1.5)
if output_file:
plt.savefig(output_file)
if display:
try:
plt.show()
except:
pass
quals_dict_filtered = {}
ids_in_alignment_file = []
while alignment.next():
ids_in_alignment_file.append(alignment.id)
ids_in_alignment_file = set(ids_in_alignment_file)
for read_id in quals_dict:
if read_id in ids_in_alignment_file:
quals_dict_filtered[read_id] = quals_dict[read_id]
ids_in_alignment_file.remove(read_id)
qual_stats_dict = get_qual_stats_dict(quals_dict_filtered)
colors = get_list_of_colors(21, colormap="RdYlGn")
colors = [colors[0] for _ in range(0, 20)] + colors
max_count = max([qual_stats_dict[q]['count'] for q in qual_stats_dict if qual_stats_dict[q]])
alignment_length = len(quals_dict.values()[0])
fig = plt.figure(figsize = (25, 8))
plt.rc('grid', color='0.50', linestyle='-', linewidth=0.1)
plt.grid(True)
plt.subplots_adjust(left=0.02, bottom = 0.09, top = 0.95, right = 0.98)
for position in range(0, alignment_length):
print position
if not qual_stats_dict[position]:
