def main():
try:
start_time = datetime.now()
start_time_string = start_time.strftime('%Y-%m-%d %H:%M:%S')
description = [
'~~~CRISPRessoAggregate~~~', '-Aggregation of CRISPResso Run Data-'
]
aggregate_string = r'''
___________________________________
| __ __ _ _ __ ___ _ |
| /\ /__ /__ |_) |_ /__ /\ | |_ |
|/--\ \_| \_| | \ |_ \_| /--\ | |_ |
|__________________________________|
'''
print(
CRISPRessoShared.get_crispresso_header(description,
aggregate_string))
parser = argparse.ArgumentParser(
description="Aggreate CRISPResso2 Runs")
parser.add_argument(
"-p",
"--prefix",
action='append',
help=
"Prefix for CRISPResso folders to aggregate (may be specified multiple times)",
default=[])
parser.add_argument("-s",
"--suffix",
type=str,
help="Suffix for CRISPResso folders to aggregate",
default="")
parser.add_argument("-n",
"--name",
type=str,
help="Output name of the report",
required=True)
parser.add_argument(
'--min_reads_for_inclusion',
help=
'Minimum number of reads for a run to be included in the run summary',
type=int,
default=0)
parser.add_argument(
'--place_report_in_output_folder',
help=
'If true, report will be written inside the CRISPResso output folder. By default, the report will be written one directory up from the report output.',
action='store_true')
parser.add_argument('--suppress_report',
help='Suppress output report',
action='store_true')
parser.add_argument('--suppress_plots',
help='Suppress output plots',
action='store_true')
parser.add_argument('--debug',
help='Show debug messages',
action='store_true')
args = parser.parse_args()
output_folder_name = 'CRISPRessoAggregate_on_%s' % args.name
OUTPUT_DIRECTORY = os.path.abspath(output_folder_name)
_jp = lambda filename: os.path.join(
OUTPUT_DIRECTORY, filename
) #handy function to put a file in the output directory
try:
info('Creating Folder %s' % OUTPUT_DIRECTORY)
os.makedirs(OUTPUT_DIRECTORY)
except:
warn('Folder %s already exists.' % OUTPUT_DIRECTORY)
log_filename = _jp('CRISPRessoAggregate_RUNNING_LOG.txt')
logging.getLogger().addHandler(logging.FileHandler(log_filename))
with open(log_filename, 'w+') as outfile:
outfile.write('[Command used]:\n%s\n\n[Execution log]:\n' %
' '.join(sys.argv))
crispresso2Aggregate_info_file = os.path.join(
OUTPUT_DIRECTORY, 'CRISPResso2Aggregate_info.pickle')
crispresso2_info = {
} #keep track of all information for this run to be pickled and saved at the end of the run
crispresso2_info['version'] = CRISPRessoShared.__version__
crispresso2_info['args'] = deepcopy(args)
crispresso2_info['log_filename'] = os.path.basename(log_filename)
#glob returns paths including the original prefix
all_files = []
for prefix in args.prefix:
all_files.extend(glob.glob(prefix + '*' + args.suffix))
if args.prefix != "":
all_files.extend(glob.glob(
prefix + '/*' +
args.suffix)) #if a folder is given, add all subfolders
seen_folders = {}
crispresso2_folder_infos = {
} #file_loc->crispresso_info; these are only CRISPResso runs -- this bit unrolls batch, pooled, and wgs runs
successfully_imported_count = 0
not_imported_count = 0
for folder in all_files:
if folder in seen_folders: #skip if we've seen this folder (glob could have added it twice)
continue
seen_folders[folder] = 1
if os.path.isdir(folder) and str(folder).endswith(args.suffix):
#first, try to import a plain CRISPResso2 run
crispresso_info_file = os.path.join(folder,
'CRISPResso2_info.pickle')
if os.path.exists(crispresso_info_file):
try:
run_data = CRISPRessoShared.load_crispresso_info(
folder)
crispresso2_folder_infos[folder] = run_data
successfully_imported_count += 1
except Exception as e:
warn('Could not open CRISPResso2 info file in ' +
folder)
not_imported_count += 1
#second, check pooled
pooled_info_file = os.path.join(
folder, 'CRISPResso2Pooled_info.pickle')
if os.path.exists(pooled_info_file):
pooled_data = cp.load(open(pooled_info_file, 'rb'))
if 'good_region_names' in pooled_data:
run_names = pooled_data['good_region_names']
for run_name in run_names:
run_folder_loc = os.path.join(
folder, 'CRISPResso_on_%s' % run_name)
try:
run_data = CRISPRessoShared.load_crispresso_info(
run_folder_loc)
crispresso2_folder_infos[
run_folder_loc] = run_data
successfully_imported_count += 1
except Exception as e:
warn('Could not open CRISPResso2 info file in '
+ run_folder_loc)
not_imported_count += 1
else:
warn('Could not process pooled folder ' + folder)
not_imported_count += 1
#third, check batch
batch_info_file = os.path.join(folder,
'CRISPResso2Batch_info.pickle')
if os.path.exists(batch_info_file):
batch_data = cp.load(open(batch_info_file, 'rb'))
if 'completed_batch_arr' in batch_data:
run_names = batch_data['completed_batch_arr']
for run_name in run_names:
run_folder_loc = os.path.join(
folder, 'CRISPResso_on_%s' % run_name)
try:
run_data = CRISPRessoShared.load_crispresso_info(
run_folder_loc)
crispresso2_folder_infos[
run_folder_loc] = run_data
successfully_imported_count += 1
except Exception as e:
warn('Could not open CRISPResso2 info file in '
+ run_folder_loc)
not_imported_count += 1
else:
warn('Could not process batch folder ' + folder)
not_imported_count += 1
#fourth, check WGS
wgs_info_file = os.path.join(folder,
'CRISPResso2WGS_info.pickle')
if os.path.exists(wgs_info_file):
wgs_data = cp.load(open(wgs_info_file, 'rb'))
if 'good_region_folders' in wgs_data:
run_names = wgs_data['good_region_folders']
for run_name in run_names:
run_folder_loc = os.path.join(
folder, 'CRISPResso_on_%s' % run_name)
try:
run_data = CRISPRessoShared.load_crispresso_info(
run_folder_loc)
crispresso2_folder_infos[
run_folder_loc] = run_data
successfully_imported_count += 1
except Exception as e:
warn('Could not open CRISPResso2 info file in '
+ run_folder_loc)
not_imported_count += 1
else:
warn('Could not process WGS folder ' + folder)
not_imported_count += 1
info('Read ' + str(successfully_imported_count) + ' folders (' +
str(not_imported_count) + ' not imported)')
save_png = True
if args.suppress_report:
save_png = False
if successfully_imported_count > 0:
crispresso2_folders = crispresso2_folder_infos.keys()
crispresso2_folder_names = {}
crispresso2_folder_htmls = {} #file_loc->html folder loc
for crispresso2_folder in crispresso2_folders:
crispresso2_folder_names[
crispresso2_folder] = CRISPRessoShared.slugify(
crispresso2_folder)
this_sub_html_file = crispresso2_folder + ".html"
if crispresso2_folder_infos[crispresso2_folder][
'args'].place_report_in_output_folder:
this_sub_html_file = os.path.join(
crispresso2_folder,
crispresso2_folder_infos[crispresso2_folder]
['report_filename'])
crispresso2_folder_htmls[crispresso2_folder] = os.path.abspath(
this_sub_html_file)
all_amplicons = set()
amplicon_names = {
} #sequence -> ref name (to check for amplicons with the same name but different sequences)
amplicon_counts = {}
amplicon_sources = {}
completed_batch_arr = []
for crispresso2_folder in crispresso2_folders:
run_data = crispresso2_folder_infos[crispresso2_folder]
for ref_name in run_data['ref_names']:
ref_seq = run_data['refs'][ref_name]['sequence']
all_amplicons.add(ref_seq)
#if this amplicon is called something else in another sample, just call it the amplicon
if ref_name in amplicon_names and amplicon_names[
ref_seq] != ref_name:
amplicon_names[ref_seq] = ref_seq
else:
amplicon_names[ref_seq] = ref_name
if ref_seq not in amplicon_counts:
amplicon_counts[ref_seq] = 0
amplicon_sources[ref_seq] = []
amplicon_counts[ref_seq] += 1
amplicon_sources[ref_seq].append(crispresso2_folder + '(' +
ref_name + ')')
#make sure amplicon names aren't super long
for amplicon in all_amplicons:
if len(amplicon_names[amplicon]) > 21:
amplicon_names[amplicon] = amplicon_names[amplicon][0:21]
#make sure no duplicate amplicon names (same name for the different amplicons)
seen_names = []
for amplicon in all_amplicons:
suffix_counter = 2
orig_name = amplicon_names[amplicon]
while amplicon_names[amplicon] in seen_names:
amplicon_names[amplicon] = orig_name + "_" + str(
suffix_counter)
suffix_counter += 1
seen_names.append(amplicon_names[amplicon])
crispresso2_info['ref_names'] = seen_names
crispresso2_info['refs'] = {}
crispresso2_info['summary_plot_names'] = []
crispresso2_info['summary_plot_titles'] = {}
crispresso2_info['summary_plot_labels'] = {}
crispresso2_info['summary_plot_datas'] = {}
with open(_jp('CRISPRessoAggregate_amplicon_information.txt'),
'w') as outfile:
outfile.write("\t".join([
'Amplicon Name', 'Number of sources', 'Amplicon sources',
'Amplicon sequence'
]) + "\n")
for amplicon in all_amplicons:
outfile.write("\t".join([
amplicon_names[amplicon],
str(amplicon_counts[amplicon]), ';'.join(
amplicon_sources[amplicon]), amplicon
]) + "\n")
window_nuc_pct_quilt_plot_names = []
nuc_pct_quilt_plot_names = []
window_nuc_conv_plot_names = []
nuc_conv_plot_names = []
#report for amplicons that appear multiple times
for amplicon_index, amplicon_seq in enumerate(all_amplicons):
amplicon_name = amplicon_names[amplicon_seq]
crispresso2_info['refs'][amplicon_name] = {}
#only perform comparison if amplicon seen in more than one sample
if amplicon_counts[amplicon_seq] < 2:
continue
info('Reporting summary for amplicon: "' + amplicon_name + '"')
consensus_sequence = ""
nucleotide_frequency_summary = []
nucleotide_percentage_summary = []
modification_frequency_summary = []
modification_percentage_summary = []
amp_found_count = 0 #how many folders had information for this amplicon
consensus_guides = []
consensus_include_idxs = []
consensus_sgRNA_plot_idxs = []
consensus_sgRNA_intervals = []
guides_all_same = True
runs_with_this_amplicon = []
for crispresso2_folder in crispresso2_folders:
run_data = crispresso2_folder_infos[crispresso2_folder]
run_has_amplicon = False
run_amplicon_name = ''
for ref_name in run_data['ref_names']:
if amplicon_seq == run_data['refs'][ref_name][
'sequence']:
run_has_amplicon = True
run_amplicon_name = ref_name
if not run_has_amplicon:
continue
runs_with_this_amplicon.append(crispresso2_folder)
if consensus_guides == []:
consensus_guides = run_data['refs'][run_amplicon_name][
'sgRNA_sequences']
consensus_include_idxs = run_data['refs'][
run_amplicon_name]['include_idxs']
consensus_sgRNA_intervals = run_data['refs'][
run_amplicon_name]['sgRNA_intervals']
consensus_sgRNA_plot_idxs = run_data['refs'][
run_amplicon_name]['sgRNA_plot_idxs']
if run_data['refs'][run_amplicon_name][
'sgRNA_sequences'] != consensus_guides:
guides_all_same = False
if set(run_data['refs'][run_amplicon_name]
['include_idxs']) != set(consensus_include_idxs):
guides_all_same = False
if 'nuc_freq_filename' not in run_data['refs'][
run_amplicon_name]:
info(
"Skipping the amplicon '%s' in folder '%s'. Cannot find nucleotide information."
% (run_amplicon_name, crispresso2_folder))
continue
nucleotide_frequency_file = os.path.join(
crispresso2_folder, run_data['refs'][run_amplicon_name]
['nuc_freq_filename'])
ampSeq_nf, nuc_freqs = CRISPRessoShared.parse_count_file(
nucleotide_frequency_file)
nucleotide_pct_file = os.path.join(
crispresso2_folder, run_data['refs'][run_amplicon_name]
['nuc_pct_filename'])
ampSeq_np, nuc_pcts = CRISPRessoShared.parse_count_file(
nucleotide_pct_file)
count_file = os.path.join(
crispresso2_folder, run_data['refs'][run_amplicon_name]
['mod_count_filename'])
ampSeq_cf, mod_freqs = CRISPRessoShared.parse_count_file(
count_file)
if ampSeq_nf is None or ampSeq_np is None or ampSeq_cf is None:
info(
"Skipping the amplicon '%s' in folder '%s'. Could not parse run output."
% (run_amplicon_name, crispresso2_folder))
info(
"Nucleotide frequency amplicon: '%s', Nucleotide percentage amplicon: '%s', Count vectors amplicon: '%s'"
% (ampSeq_nf, ampSeq_np, ampSeq_cf))
continue
if ampSeq_nf != ampSeq_np or ampSeq_np != ampSeq_cf:
warn(
"Skipping the amplicon '%s' in folder '%s'. Parsed amplicon sequences do not match\nnf:%s\nnp:%s\ncf:%s\nrf:%s"
% (run_amplicon_name, crispresso2_folder,
ampSeq_nf, ampSeq_np, ampSeq_cf, amplicon_seq))
continue
if consensus_sequence == "":
consensus_sequence = ampSeq_nf
if ampSeq_nf != consensus_sequence:
info(
"Skipping the amplicon '%s' in folder '%s'. Amplicon sequences do not match."
% (run_amplicon_name, crispresso2_folder))
continue
if 'Total' not in mod_freqs:
info(
"Skipping the amplicon '%s' in folder '%s'. Processing did not complete."
% (run_amplicon_name, crispresso2_folder))
continue
if mod_freqs['Total'][0] == 0 or mod_freqs['Total'][
0] == "0":
info(
"Skipping the amplicon '%s' in folder '%s'. Got no reads for amplicon."
% (run_amplicon_name, crispresso2_folder))
continue
this_amp_total_reads = run_data['counts_total'][
run_amplicon_name]
if this_amp_total_reads < args.min_reads_for_inclusion:
info(
"Skipping the amplicon '%s' in folder '%s'. Got %s reads (min_reads_for_inclusion is %d)."
% (run_amplicon_name, crispresso2_folder,
str(this_amp_total_reads),
args.min_reads_for_inclusion))
continue
mod_pcts = {}
for key in mod_freqs:
mod_pcts[key] = np.array(mod_freqs[key]).astype(
np.float) / float(this_amp_total_reads)
amp_found_count += 1
run_name = crispresso2_folder_names[crispresso2_folder]
for nuc in ['A', 'T', 'C', 'G', 'N', '-']:
row = [run_name, nuc]
row.extend(nuc_freqs[nuc])
nucleotide_frequency_summary.append(row)
pct_row = [run_name, nuc]
pct_row.extend(nuc_pcts[nuc])
nucleotide_percentage_summary.append(pct_row)
for mod in [
'Insertions', 'Insertions_Left', 'Deletions',
'Substitutions', 'All_modifications'
]:
row = [run_name, mod]
row.extend(mod_freqs[mod])
modification_frequency_summary.append(row)
pct_row = [run_name, mod]
pct_row.extend(mod_pcts[mod])
modification_percentage_summary.append(pct_row)
if amp_found_count == 0:
info(
"Couldn't find any data for amplicon '%s'. Not compiling results."
% amplicon_name)
else:
amplicon_plot_name = amplicon_name + "."
if len(amplicon_names
) == 1 and amplicon_name == "Reference":
amplicon_plot_name = ""
colnames = ['Folder', 'Nucleotide']
colnames.extend(list(consensus_sequence))
nucleotide_frequency_summary_df = pd.DataFrame(
nucleotide_frequency_summary, columns=colnames)
nucleotide_frequency_summary_df = pd.concat([
nucleotide_frequency_summary_df.iloc[:, 0:2],
nucleotide_frequency_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
nucleotide_frequency_summary_filename = _jp(
amplicon_plot_name +
'Nucleotide_frequency_summary.txt')
nucleotide_frequency_summary_df.to_csv(
nucleotide_frequency_summary_filename,
sep='\t',
index=None)
nucleotide_percentage_summary_df = pd.DataFrame(
nucleotide_percentage_summary, columns=colnames)
nucleotide_percentage_summary_df = pd.concat([
nucleotide_percentage_summary_df.iloc[:, 0:2],
nucleotide_percentage_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
nucleotide_percentage_summary_filename = _jp(
amplicon_plot_name +
'Nucleotide_percentage_summary.txt')
nucleotide_percentage_summary_df.to_csv(
nucleotide_percentage_summary_filename,
sep='\t',
index=None)
colnames = ['Folder', 'Modification']
colnames.extend(list(consensus_sequence))
modification_frequency_summary_df = pd.DataFrame(
modification_frequency_summary, columns=colnames)
modification_frequency_summary_df = pd.concat([
modification_frequency_summary_df.iloc[:, 0:2],
modification_frequency_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
modification_frequency_summary_filename = _jp(
amplicon_plot_name +
'MODIFICATION_FREQUENCY_SUMMARY.txt')
modification_frequency_summary_df.to_csv(
modification_frequency_summary_filename,
sep='\t',
index=None)
modification_percentage_summary_df = pd.DataFrame(
modification_percentage_summary, columns=colnames)
modification_percentage_summary_df = pd.concat([
modification_percentage_summary_df.iloc[:, 0:2],
modification_percentage_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
modification_percentage_summary_filename = _jp(
amplicon_plot_name +
'MODIFICATION_PERCENTAGE_SUMMARY.txt')
modification_percentage_summary_df.to_csv(
modification_percentage_summary_filename,
sep='\t',
index=None)
crispresso2_info['refs'][amplicon_name][
'nucleotide_frequency_summary_filename'] = os.path.basename(
nucleotide_frequency_summary_filename)
crispresso2_info['refs'][amplicon_name][
'nucleotide_percentage_summary_filename'] = os.path.basename(
nucleotide_percentage_summary_filename)
crispresso2_info['refs'][amplicon_name][
'modification_frequency_summary_filename'] = os.path.basename(
modification_frequency_summary_filename)
crispresso2_info['refs'][amplicon_name][
'modification_percentage_summary_filename'] = os.path.basename(
modification_percentage_summary_filename)
#if guides are all the same, merge substitutions and perform base editor comparison at guide quantification window
if guides_all_same and consensus_guides != []:
info(
"All guides are equal. Performing comparison of runs for amplicon '%s'"
% amplicon_name)
include_idxs = consensus_include_idxs #include indexes are the same for all guides
for idx, sgRNA in enumerate(consensus_guides):
sgRNA_intervals = consensus_sgRNA_intervals[idx]
sgRNA_plot_idxs = consensus_sgRNA_plot_idxs[idx]
plot_idxs_flat = [0, 1] # guide, nucleotide
plot_idxs_flat.extend(
[plot_idx + 2 for plot_idx in sgRNA_plot_idxs])
sub_nucleotide_frequency_summary_df = nucleotide_frequency_summary_df.iloc[:,
plot_idxs_flat]
sub_nucleotide_percentage_summary_df = nucleotide_percentage_summary_df.iloc[:,
plot_idxs_flat]
sub_modification_percentage_summary_df = modification_percentage_summary_df.iloc[:,
plot_idxs_flat]
#show all sgRNA's on the plot
sub_sgRNA_intervals = []
for sgRNA_interval in consensus_sgRNA_intervals:
newstart = None
newend = None
for idx, i in enumerate(sgRNA_plot_idxs):
if i <= sgRNA_interval[0]:
newstart = idx
if newend is None and i >= sgRNA_interval[
1]:
newend = idx
#if guide doesn't overlap with plot idxs
if newend == 0 or newstart == len(
sgRNA_plot_idxs):
continue
#otherwise, correct partial overlaps
elif newstart == None and newend == None:
newstart = 0
newend = len(include_idxs) - 1
elif newstart == None:
newstart = 0
elif newend == None:
newend = len(include_idxs) - 1
#and add it to the list
sub_sgRNA_intervals.append((newstart, newend))
if not args.suppress_plots:
#plot for each guide
this_window_nuc_pct_quilt_plot_name = _jp(
amplicon_plot_name +
'Nucleotide_percentage_quilt_around_sgRNA_'
+ sgRNA)
CRISPRessoPlot.plot_nucleotide_quilt(
sub_nucleotide_percentage_summary_df,
sub_modification_percentage_summary_df,
this_window_nuc_pct_quilt_plot_name,
save_png,
sgRNA_intervals=sub_sgRNA_intervals,
quantification_window_idxs=include_idxs,
group_column='Folder')
plot_name = os.path.basename(
this_window_nuc_pct_quilt_plot_name)
window_nuc_pct_quilt_plot_names.append(
plot_name)
crispresso2_info['summary_plot_titles'][
plot_name] = 'sgRNA: ' + sgRNA + ' Amplicon: ' + amplicon_name
if len(consensus_guides) == 1:
crispresso2_info['summary_plot_titles'][
plot_name] = ''
crispresso2_info['summary_plot_labels'][
plot_name] = 'Composition of each base around the guide ' + sgRNA + ' for the amplicon ' + amplicon_name
crispresso2_info['summary_plot_datas'][
plot_name] = [
(amplicon_name +
' nucleotide frequencies',
os.path.basename(
nucleotide_frequency_summary_filename
)),
(amplicon_name +
' modification frequencies',
os.path.basename(
modification_frequency_summary_filename
))
]
sub_nucleotide_frequency_summary_df = pd.concat(
[
sub_nucleotide_frequency_summary_df.
iloc[:, 0:2],
sub_nucleotide_frequency_summary_df.
iloc[:, 2:].apply(pd.to_numeric)
],
axis=1)
sub_nucleotide_frequency_summary_filename = _jp(
amplicon_plot_name +
'Nucleotide_frequency_summary_around_sgRNA_'
+ sgRNA + '.txt')
sub_nucleotide_frequency_summary_df.to_csv(
sub_nucleotide_frequency_summary_filename,
sep='\t',
index=None)
sub_nucleotide_percentage_summary_df = pd.concat(
[
sub_nucleotide_percentage_summary_df.
iloc[:, 0:2],
sub_nucleotide_percentage_summary_df.
iloc[:, 2:].apply(pd.to_numeric)
],
axis=1)
sub_nucleotide_percentage_summary_filename = _jp(
amplicon_plot_name +
'Nucleotide_percentage_summary_around_sgRNA_'
+ sgRNA + '.txt')
sub_nucleotide_percentage_summary_df.to_csv(
sub_nucleotide_percentage_summary_filename,
sep='\t',
index=None)
if not args.suppress_plots: # plot the whole region
this_nuc_pct_quilt_plot_name = _jp(
amplicon_plot_name +
'Nucleotide_percentage_quilt')
CRISPRessoPlot.plot_nucleotide_quilt(
nucleotide_percentage_summary_df,
modification_percentage_summary_df,
this_nuc_pct_quilt_plot_name,
save_png,
sgRNA_intervals=consensus_sgRNA_intervals,
quantification_window_idxs=include_idxs,
group_column='Folder')
plot_name = os.path.basename(
this_nuc_pct_quilt_plot_name)
nuc_pct_quilt_plot_names.append(plot_name)
crispresso2_info['summary_plot_titles'][
plot_name] = 'Amplicon: ' + amplicon_name
if len(amplicon_names) == 1:
crispresso2_info['summary_plot_titles'][
plot_name] = ''
crispresso2_info['summary_plot_labels'][
plot_name] = 'Composition of each base for the amplicon ' + amplicon_name
crispresso2_info['summary_plot_datas'][plot_name] = [
(amplicon_name + ' nucleotide frequencies',
os.path.basename(
nucleotide_frequency_summary_filename)),
(amplicon_name + ' modification frequencies',
os.path.basename(
modification_frequency_summary_filename))
]
else: #guides are not the same
if not args.suppress_plots:
this_nuc_pct_quilt_plot_name = _jp(
amplicon_plot_name +
'Nucleotide_percentage_quilt')
CRISPRessoPlot.plot_nucleotide_quilt(
nucleotide_percentage_summary_df,
modification_percentage_summary_df,
this_nuc_pct_quilt_plot_name,
save_png,
group_column='Folder')
plot_name = os.path.basename(
this_nuc_pct_quilt_plot_name)
nuc_pct_quilt_plot_names.append(plot_name)
crispresso2_info['summary_plot_labels'][
plot_name] = 'Composition of each base for the amplicon ' + amplicon_name
crispresso2_info['summary_plot_datas'][plot_name] = [
(amplicon_name + ' nucleotide frequencies',
os.path.basename(
nucleotide_frequency_summary_filename)),
(amplicon_name + ' modification frequencies',
os.path.basename(
modification_frequency_summary_filename))
]
crispresso2_info[
'window_nuc_pct_quilt_plot_names'] = window_nuc_pct_quilt_plot_names
crispresso2_info[
'nuc_pct_quilt_plot_names'] = nuc_pct_quilt_plot_names
crispresso2_info[
'window_nuc_conv_plot_names'] = window_nuc_conv_plot_names
crispresso2_info['nuc_conv_plot_names'] = nuc_conv_plot_names
quantification_summary = []
#summarize amplicon modifications
samples_quantification_summary_by_amplicon_filename = _jp(
'CRISPRessoAggregate_quantification_of_editing_frequency_by_amplicon.txt'
) #this file has separate lines for each amplicon in each run
with open(samples_quantification_summary_by_amplicon_filename,
'w') as outfile:
wrote_header = False
for crispresso2_folder in crispresso2_folders:
run_data = crispresso2_folder_infos[crispresso2_folder]
run_name = crispresso2_folder_names[crispresso2_folder]
amplicon_modification_file = os.path.join(
crispresso2_folder,
run_data['quant_of_editing_freq_filename'])
with open(amplicon_modification_file, 'r') as infile:
file_head = infile.readline()
if not wrote_header:
outfile.write('Folder\t' + file_head)
wrote_header = True
for line in infile:
outfile.write(crispresso2_folder + "\t" + line)
n_tot = run_data['aln_stats']['N_TOT_READS']
n_aligned = 0
n_unmod = 0
n_mod = 0
n_discarded = 0
n_insertion = 0
n_deletion = 0
n_substitution = 0
n_only_insertion = 0
n_only_deletion = 0
n_only_substitution = 0
n_insertion_and_deletion = 0
n_insertion_and_substitution = 0
n_deletion_and_substitution = 0
n_insertion_and_deletion_and_substitution = 0
for ref_name in run_data[
'ref_names']: #multiple alleles could be provided
n_aligned += run_data['counts_total'][ref_name]
n_unmod += run_data['counts_unmodified'][ref_name]
n_mod += run_data['counts_modified'][ref_name]
n_discarded += run_data['counts_discarded'][ref_name]
n_insertion += run_data['counts_insertion'][ref_name]
n_deletion += run_data['counts_deletion'][ref_name]
n_substitution += run_data['counts_substitution'][
ref_name]
n_only_insertion += run_data['counts_only_insertion'][
ref_name]
n_only_deletion += run_data['counts_only_deletion'][
ref_name]
n_only_substitution += run_data[
'counts_only_substitution'][ref_name]
n_insertion_and_deletion += run_data[
'counts_insertion_and_deletion'][ref_name]
n_insertion_and_substitution += run_data[
'counts_insertion_and_substitution'][ref_name]
n_deletion_and_substitution += run_data[
'counts_deletion_and_substitution'][ref_name]
n_insertion_and_deletion_and_substitution += run_data[
'counts_insertion_and_deletion_and_substitution'][
ref_name]
unmod_pct = np.nan
mod_pct = np.nan
if n_aligned > 0:
unmod_pct = 100 * n_unmod / float(n_aligned)
mod_pct = 100 * n_mod / float(n_aligned)
vals = [run_name]
vals.extend([
round(unmod_pct, 8),
round(mod_pct, 8), n_aligned, n_tot, n_unmod, n_mod,
n_discarded, n_insertion, n_deletion, n_substitution,
n_only_insertion, n_only_deletion, n_only_substitution,
n_insertion_and_deletion, n_insertion_and_substitution,
n_deletion_and_substitution,
n_insertion_and_deletion_and_substitution
])
quantification_summary.append(vals)
header = 'Name\tUnmodified%\tModified%\tReads_total\tReads_aligned\tUnmodified\tModified\tDiscarded\tInsertions\tDeletions\tSubstitutions\tOnly Insertions\tOnly Deletions\tOnly Substitutions\tInsertions and Deletions\tInsertions and Substitutions\tDeletions and Substitutions\tInsertions Deletions and Substitutions'
header_els = header.split("\t")
df_summary_quantification = pd.DataFrame(quantification_summary,
columns=header_els)
samples_quantification_summary_filename = _jp(
'CRISPRessoAggregate_quantification_of_editing_frequency.txt'
) #this file has one line for each run (sum of all amplicons)
df_summary_quantification.fillna('NA').to_csv(
samples_quantification_summary_filename, sep='\t', index=None)
crispresso2_info[
'samples_quantification_summary_filename'] = os.path.basename(
samples_quantification_summary_filename)
crispresso2_info[
'samples_quantification_summary_by_amplicon_filename'] = os.path.basename(
samples_quantification_summary_by_amplicon_filename)
df_summary_quantification.set_index('Name')
if not args.suppress_plots:
plot_root = _jp("CRISPRessoAggregate_reads_summary")
CRISPRessoPlot.plot_reads_total(plot_root,
df_summary_quantification,
save_png,
args.min_reads_for_inclusion)
plot_name = os.path.basename(plot_root)
crispresso2_info['summary_plot_root'] = plot_name
crispresso2_info['summary_plot_names'].append(plot_name)
crispresso2_info['summary_plot_titles'][
plot_name] = 'CRISPRessoAggregate Mapping Statistics Summary'
crispresso2_info['summary_plot_labels'][
plot_name] = 'Each bar shows the total number of reads in each sample. The vertical line shows the cutoff for analysis, set using the --min_reads_for_inclusion parameter.'
crispresso2_info['summary_plot_datas'][plot_name] = [
('CRISPRessoAggregate summary',
os.path.basename(samples_quantification_summary_filename)
),
('CRISPRessoAggregate summary by amplicon',
os.path.basename(
samples_quantification_summary_by_amplicon_filename))
]
plot_root = _jp(
"CRISPRessoAggregate_quantification_of_editing_frequency")
CRISPRessoPlot.plot_unmod_mod_pcts(
plot_root, df_summary_quantification, save_png,
args.min_reads_for_inclusion)
plot_name = os.path.basename(plot_root)
crispresso2_info['summary_plot_root'] = plot_name
crispresso2_info['summary_plot_names'].append(plot_name)
crispresso2_info['summary_plot_titles'][
plot_name] = 'CRISPRessoAggregate Modification Summary'
crispresso2_info['summary_plot_labels'][
plot_name] = 'Each bar shows the total number of reads aligned to each amplicon, divided into the reads that are modified and unmodified. The vertical line shows the cutoff for analysis, set using the --min_reads_for_inclusion parameter.'
crispresso2_info['summary_plot_datas'][plot_name] = [
('CRISPRessoAggregate summary',
os.path.basename(samples_quantification_summary_filename)
),
('CRISPRessoAggregate summary by amplicon',
os.path.basename(
samples_quantification_summary_by_amplicon_filename))
]
#summarize alignment
with open(_jp('CRISPRessoAggregate_mapping_statistics.txt'),
'w') as outfile:
wrote_header = False
for crispresso2_folder in crispresso2_folders:
run_data = crispresso2_folder_infos[crispresso2_folder]
run_name = crispresso2_folder_names[crispresso2_folder]
mapping_file = os.path.join(
crispresso2_folder, run_data['mapping_stats_filename'])
with open(mapping_file, 'r') as infile:
file_head = infile.readline()
if not wrote_header:
outfile.write('Folder\t' + file_head)
wrote_header = True
for line in infile:
outfile.write(crispresso2_folder + "\t" + line)
if not args.suppress_report:
report_filename = OUTPUT_DIRECTORY + '.html'
if (args.place_report_in_output_folder):
report_filename = _jp("CRISPResso2Aggregate_report.html")
CRISPRessoReport.make_aggregate_report(
crispresso2_info, args.name, report_filename,
OUTPUT_DIRECTORY, _ROOT, crispresso2_folders,
crispresso2_folder_htmls)
crispresso2_info['report_location'] = report_filename
crispresso2_info['report_filename'] = os.path.basename(
report_filename)
end_time = datetime.now()
end_time_string = end_time.strftime('%Y-%m-%d %H:%M:%S')
running_time = end_time - start_time
running_time_string = str(running_time)
crispresso2_info['end_time'] = end_time
crispresso2_info['end_time_string'] = end_time_string
crispresso2_info['running_time'] = running_time
crispresso2_info['running_time_string'] = running_time_string
cp.dump(crispresso2_info, open(crispresso2Aggregate_info_file, 'wb'))
info('Analysis Complete!')
print(CRISPRessoShared.get_crispresso_footer())
sys.exit(0)
except Exception as e:
debug_flag = False
if 'args' in vars() and 'debug' in args:
debug_flag = args.debug
if debug_flag:
traceback.print_exc(file=sys.stdout)
error('\n\nERROR: %s' % e)
sys.exit(-1)
def main():
try:
description = [
'~~~CRISPRessoBatch~~~',
'-Analysis of CRISPR/Cas9 outcomes from batch deep sequencing data-'
]
batch_string = r'''
_________________
| __ ___ __ |
||__) /\ | / |__||
||__)/--\| \__| ||
|_________________|
'''
print(CRISPRessoShared.get_crispresso_header(description,
batch_string))
parser = CRISPRessoShared.getCRISPRessoArgParser(
parserTitle='CRISPRessoBatch Parameters')
#batch specific params
parser.add_argument(
'-bs',
'--batch_settings',
type=str,
help=
'Settings file for batch. Must be tab-separated text file. The header row contains CRISPResso parameters (e.g., fastq_r1, fastq_r2, amplicon_seq, and other optional parameters). Each following row sets parameters for an additional batch.',
required=True)
parser.add_argument(
'--skip_failed',
help='Continue with batch analysis even if one sample fails',
action='store_true')
parser.add_argument(
'--min_reads_for_inclusion',
help=
'Minimum number of reads for a batch to be included in the batch summary',
type=int)
parser.add_argument(
'-p',
'--n_processes',
type=int,
help='Specify the number of processes to use for quantification.\
Please use with caution since increasing this parameter will increase the memory required to run CRISPResso.',
default=1)
parser.add_argument(
'-bo',
'--batch_output_folder',
help='Directory where batch analysis output will be stored')
parser.add_argument('--crispresso_command',
help='CRISPResso command to call',
default='CRISPResso')
args = parser.parse_args()
debug_flag = args.debug
crispresso_options = CRISPRessoShared.get_crispresso_options()
options_to_ignore = set(['name', 'output_folder'])
crispresso_options_for_batch = list(crispresso_options -
options_to_ignore)
CRISPRessoShared.check_file(args.batch_settings)
##parse excel sheet
batch_params = pd.read_csv(args.batch_settings, comment='#', sep='\t')
#pandas either allows for auto-detect sep or for comment. not both
# batch_params=pd.read_csv(args.batch_settings,sep=None,engine='python',error_bad_lines=False)
batch_params.columns = batch_params.columns.str.strip(' -\xd0')
#rename column "a" to "amplicon_seq", etc
batch_params.rename(
index=str,
columns=CRISPRessoShared.get_crispresso_options_lookup(),
inplace=True)
batch_count = batch_params.shape[0]
batch_params.index = range(batch_count)
if 'fastq_r1' not in batch_params and 'bam_input' not in batch_params:
raise CRISPRessoShared.BadParameterException(
"fastq_r1 must be specified in the batch settings file. Current headings are: "
+ str(batch_params.columns.values))
#add args from the command line to batch_params_df
for arg in vars(args):
if arg not in batch_params:
batch_params[arg] = getattr(args, arg)
else:
if (getattr(args, arg) is not None):
batch_params[arg].fillna(value=getattr(args, arg),
inplace=True)
#assert that all names are unique
#and clean names
for i in range(batch_count):
if batch_params.loc[i, 'name'] == '':
batch_params.at[i, 'name'] = i
batch_params.at[i, 'name'] = CRISPRessoShared.clean_filename(
batch_params.loc[i, 'name'])
if batch_params.drop_duplicates(
'name').shape[0] != batch_params.shape[0]:
raise CRISPRessoShared.BadParameterException(
'Batch input names must be unique. The given names are not unique: '
+ str(batch_params.loc[:, 'name']))
#Check files
batch_params[
"sgRNA_intervals"] = '' #create empty array for sgRNA intervals
batch_params["sgRNA_intervals"] = batch_params[
"sgRNA_intervals"].apply(list)
batch_params[
"cut_point_include_idx"] = '' #create empty array for cut point intervals for each batch based on sgRNA
batch_params["cut_point_include_idx"] = batch_params[
"cut_point_include_idx"].apply(list)
for idx, row in batch_params.iterrows():
if 'fastq_r1' in row:
if row.fastq_r1 is None:
raise CRISPRessoShared.BadParameterException(
"At least one fastq file must be given as a command line parameter or be specified in the batch settings file with the heading 'fastq_r1' (fastq_r1 on row %s '%s' is invalid)"
% (int(idx) + 1, row.fastq_r1))
else:
CRISPRessoShared.check_file(row.fastq_r1)
if 'fastq_r2' in row and row.fastq_r2 != "":
CRISPRessoShared.check_file(row.fastq_r2)
if 'input_bam' in row:
if row.input_bam is None:
raise CRISPRessoShared.BadParameterException(
"At least one input file must be given as a command line parameter or be specified in the batch settings file with the heading 'fastq_r1' or 'input_bam' (input_bam on row %s '%s' is invalid)"
% (int(idx) + 1, row.input_bam))
else:
CRISPRessoShared.check_file(row.input_bam)
if args.auto:
continue
curr_amplicon_seq_str = row.amplicon_seq
if curr_amplicon_seq_str is None:
raise CRISPRessoShared.BadParameterException(
"Amplicon sequence must be given as a command line parameter or be specified in the batch settings file with the heading 'amplicon_seq' (Amplicon seq on row %s '%s' is invalid)"
% (int(idx) + 1, curr_amplicon_seq_str))
guides_are_in_amplicon = {
} #dict of whether a guide is in at least one amplicon sequence
#iterate through amplicons
for curr_amplicon_seq in curr_amplicon_seq_str.split(','):
this_include_idxs = [
] #mask for bp to include for this amplicon seq, as specified by sgRNA cut points
this_sgRNA_intervals = []
wrong_nt = CRISPRessoShared.find_wrong_nt(curr_amplicon_seq)
if wrong_nt:
raise CRISPRessoShared.NTException(
'The amplicon sequence in row %d (%s) contains incorrect characters:%s'
% (idx + 1, curr_amplicon_seq_str, ' '.join(wrong_nt)))
#iterate through guides
curr_guide_seq_string = row.guide_seq
if curr_guide_seq_string is not None and curr_guide_seq_string != "":
guides = curr_guide_seq_string.strip().upper().split(',')
for curr_guide_seq in guides:
wrong_nt = CRISPRessoShared.find_wrong_nt(
curr_guide_seq)
if wrong_nt:
raise CRISPRessoShared.NTException(
'The sgRNA sequence in row %d (%s) contains incorrect characters:%s'
%
(idx + 1, curr_guide_seq, ' '.join(wrong_nt)))
guide_mismatches = [[]] * len(guides)
guide_names = [""] * len(guides)
guide_qw_centers = CRISPRessoShared.set_guide_array(
row.quantification_window_center, guides,
'guide quantification center')
guide_qw_sizes = CRISPRessoShared.set_guide_array(
row.quantification_window_size, guides,
'guide quantification size')
guide_plot_cut_points = [1] * len(guides)
(this_sgRNA_sequences, this_sgRNA_intervals,
this_sgRNA_cut_points, this_sgRNA_plot_cut_points,
this_sgRNA_plot_idxs, this_sgRNA_mismatches,
this_sgRNA_names, this_include_idxs, this_exclude_idxs
) = CRISPRessoShared.get_amplicon_info_for_guides(
curr_amplicon_seq, guides, guide_mismatches,
guide_names, guide_qw_centers, guide_qw_sizes,
row.quantification_window_coordinates,
row.exclude_bp_from_left, row.exclude_bp_from_right,
row.plot_window_size, guide_plot_cut_points)
for guide_seq in this_sgRNA_sequences:
guides_are_in_amplicon[guide_seq] = 1
batch_params.ix[idx, "cut_point_include_idx"].append(
this_include_idxs)
batch_params.ix[idx,
"sgRNA_intervals"].append(this_sgRNA_intervals)
for guide_seq in guides_are_in_amplicon:
if guides_are_in_amplicon[guide_seq] != 1:
warn(
'\nThe guide sequence provided on row %d (%s) is not present in any amplicon sequence:%s! \nNOTE: The guide will be ignored for the analysis. Please check your input!'
% (idx + 1, row.guide_seq, curr_amplicon_seq))
batch_folder_name = os.path.splitext(
os.path.basename(args.batch_settings))[0]
if args.name and args.name != "":
batch_folder_name = args.name
output_folder_name = 'CRISPRessoBatch_on_%s' % batch_folder_name
OUTPUT_DIRECTORY = os.path.abspath(output_folder_name)
if args.batch_output_folder:
OUTPUT_DIRECTORY = os.path.join(
os.path.abspath(args.batch_output_folder), output_folder_name)
_jp = lambda filename: os.path.join(
OUTPUT_DIRECTORY, filename
) #handy function to put a file in the output directory
try:
info('Creating Folder %s' % OUTPUT_DIRECTORY)
os.makedirs(OUTPUT_DIRECTORY)
except:
warn('Folder %s already exists.' % OUTPUT_DIRECTORY)
log_filename = _jp('CRISPRessoBatch_RUNNING_LOG.txt')
logging.getLogger().addHandler(logging.FileHandler(log_filename))
with open(log_filename, 'w+') as outfile:
outfile.write('[Command used]:\n%s\n\n[Execution log]:\n' %
' '.join(sys.argv))
crispresso2Batch_info_file = os.path.join(
OUTPUT_DIRECTORY, 'CRISPResso2Batch_info.pickle')
crispresso2_info = {
} #keep track of all information for this run to be pickled and saved at the end of the run
crispresso2_info['version'] = CRISPRessoShared.__version__
crispresso2_info['args'] = deepcopy(args)
crispresso2_info['log_filename'] = os.path.basename(log_filename)
crispresso_cmds = []
batch_names_arr = []
batch_input_names = {}
for idx, row in batch_params.iterrows():
batchName = CRISPRessoShared.slugify(row["name"])
batch_names_arr.append(batchName)
batch_input_names[batchName] = row["name"]
crispresso_cmd = args.crispresso_command + ' -o %s --name %s' % (
OUTPUT_DIRECTORY, batchName)
crispresso_cmd = propagate_options(crispresso_cmd,
crispresso_options_for_batch,
batch_params, idx)
crispresso_cmds.append(crispresso_cmd)
crispresso2_info['batch_names_arr'] = batch_names_arr
crispresso2_info['batch_input_names'] = batch_input_names
CRISPRessoMultiProcessing.run_crispresso_cmds(crispresso_cmds,
args.n_processes,
'batch',
args.skip_failed)
run_datas = [] #crispresso2 info from each row
all_amplicons = set()
amplicon_names = {}
amplicon_counts = {}
completed_batch_arr = []
for idx, row in batch_params.iterrows():
batchName = CRISPRessoShared.slugify(row["name"])
file_prefix = row['file_prefix']
folder_name = os.path.join(OUTPUT_DIRECTORY,
'CRISPResso_on_%s' % batchName)
run_data_file = os.path.join(folder_name,
'CRISPResso2_info.pickle')
if os.path.isfile(run_data_file) is False:
info("Skipping folder '%s'. Cannot find run data at '%s'." %
(folder_name, run_data_file))
run_datas.append(None)
continue
run_data = cp.load(open(run_data_file, 'rb'))
run_datas.append(run_data)
for ref_name in run_data['ref_names']:
ref_seq = run_data['refs'][ref_name]['sequence']
all_amplicons.add(ref_seq)
#if this amplicon is called something else in another sample, just call it the amplicon
if ref_name in amplicon_names and amplicon_names[
ref_seq] != ref_name:
amplicon_names[ref_seq] = ref_seq
else:
amplicon_names[ref_seq] = ref_name
if ref_seq not in amplicon_counts:
amplicon_counts[ref_seq] = 0
amplicon_counts[ref_seq] += 1
completed_batch_arr.append(batchName)
crispresso2_info['completed_batch_arr'] = completed_batch_arr
#make sure amplicon names aren't super long
for amplicon in all_amplicons:
if len(amplicon_names[amplicon]) > 20:
amplicon_names[amplicon] = amplicon_names[amplicon][0:20]
#make sure no duplicate names (same name for the different amplicons)
seen_names = {}
for amplicon in all_amplicons:
suffix_counter = 2
orig_name = amplicon_names[amplicon]
while amplicon_names[amplicon] in seen_names:
amplicon_names[amplicon] = orig_name + "_" + str(
suffix_counter)
suffix_counter += 1
seen_names[amplicon_names[amplicon]] = 1
save_png = True
if args.suppress_report:
save_png = False
window_nuc_pct_quilt_plot_names = []
nuc_pct_quilt_plot_names = []
window_nuc_conv_plot_names = []
nuc_conv_plot_names = []
#report for amplicons that appear multiple times
for amplicon_index, amplicon_seq in enumerate(all_amplicons):
#only perform comparison if amplicon seen in more than one sample
if amplicon_counts[amplicon_seq] < 2:
continue
amplicon_name = amplicon_names[amplicon_seq]
info('Reporting summary for amplicon: "' + amplicon_name + '"')
consensus_sequence = ""
nucleotide_frequency_summary = []
nucleotide_percentage_summary = []
modification_frequency_summary = []
modification_percentage_summary = []
amp_found_count = 0 #how many folders had information for this amplicon
consensus_guides = []
consensus_include_idxs = []
consensus_sgRNA_plot_idxs = []
consensus_sgRNA_intervals = []
guides_all_same = True
batches_with_this_amplicon = []
for idx, row in batch_params.iterrows():
batchName = CRISPRessoShared.slugify(row["name"])
file_prefix = row['file_prefix']
folder_name = os.path.join(OUTPUT_DIRECTORY,
'CRISPResso_on_%s' % batchName)
run_data = run_datas[idx]
if run_data is None:
continue
batch_has_amplicon = False
batch_amplicon_name = ''
for ref_name in run_data['ref_names']:
if amplicon_seq == run_data['refs'][ref_name]['sequence']:
batch_has_amplicon = True
batch_amplicon_name = ref_name
if not batch_has_amplicon:
continue
batches_with_this_amplicon.append(idx)
if consensus_guides == []:
consensus_guides = run_data['refs'][batch_amplicon_name][
'sgRNA_sequences']
consensus_include_idxs = run_data['refs'][
batch_amplicon_name]['include_idxs']
consensus_sgRNA_intervals = run_data['refs'][
batch_amplicon_name]['sgRNA_intervals']
consensus_sgRNA_plot_idxs = run_data['refs'][
batch_amplicon_name]['sgRNA_plot_idxs']
if run_data['refs'][batch_amplicon_name][
'sgRNA_sequences'] != consensus_guides:
guides_all_same = False
if set(run_data['refs'][batch_amplicon_name]
['include_idxs']) != set(consensus_include_idxs):
guides_all_same = False
if 'nuc_freq_filename' not in run_data['refs'][
batch_amplicon_name]:
info(
"Skipping the amplicon '%s' in folder '%s'. Cannot find nucleotide information."
% (batch_amplicon_name, folder_name))
continue
nucleotide_frequency_file = os.path.join(
folder_name,
run_data['refs'][batch_amplicon_name]['nuc_freq_filename'])
ampSeq_nf, nuc_freqs = CRISPRessoShared.parse_count_file(
nucleotide_frequency_file)
nucleotide_pct_file = os.path.join(
folder_name,
run_data['refs'][batch_amplicon_name]['nuc_pct_filename'])
ampSeq_np, nuc_pcts = CRISPRessoShared.parse_count_file(
nucleotide_pct_file)
count_file = os.path.join(
folder_name, run_data['refs'][batch_amplicon_name]
['mod_count_filename'])
ampSeq_cf, mod_freqs = CRISPRessoShared.parse_count_file(
count_file)
if ampSeq_nf is None or ampSeq_np is None or ampSeq_cf is None:
info(
"Skipping the amplicon '%s' in folder '%s'. Could not parse batch output."
% (batch_amplicon_name, folder_name))
info(
"Nucleotide frequency amplicon: '%s', Nucleotide percentage amplicon: '%s', Count vectors amplicon: '%s'"
% (ampSeq_nf, ampSeq_np, ampSeq_cf))
continue
if ampSeq_nf != ampSeq_np or ampSeq_np != ampSeq_cf:
warn(
"Skipping the amplicon '%s' in folder '%s'. Parsed amplicon sequences do not match\nnf:%s\nnp:%s\ncf:%s\nrf:%s"
% (batch_amplicon_name, folder_name, ampSeq_nf,
ampSeq_np, ampSeq_cf, amplicon_seq))
continue
if consensus_sequence == "":
consensus_sequence = ampSeq_nf
if ampSeq_nf != consensus_sequence:
info(
"Skipping the amplicon '%s' in folder '%s'. Amplicon sequences do not match."
% (batch_amplicon_name, folder_name))
continue
if 'Total' not in mod_freqs:
info(
"Skipping the amplicon '%s' in folder '%s'. Processing did not complete."
% (batch_amplicon_name, folder_name))
continue
if mod_freqs['Total'][0] == 0 or mod_freqs['Total'][0] == "0":
info(
"Skipping the amplicon '%s' in folder '%s'. Got no reads for amplicon."
% (batch_amplicon_name, folder_name))
continue
if (args.min_reads_for_inclusion is not None) and (int(
mod_freqs['Total'][0]) < args.min_reads_for_inclusion):
info(
"Skipping the amplicon '%s' in folder '%s'. Got %s reads (min_reads_for_inclusion is %d)."
% (batch_amplicon_name, folder_name,
str(mod_freqs['Total'][0]),
args.min_reads_for_inclusion))
continue
mod_pcts = {}
for key in mod_freqs:
mod_pcts[key] = np.array(mod_freqs[key]).astype(
np.float) / float(mod_freqs['Total'][0])
amp_found_count += 1
for nuc in ['A', 'T', 'C', 'G', 'N', '-']:
row = [batchName, nuc]
row.extend(nuc_freqs[nuc])
nucleotide_frequency_summary.append(row)
pct_row = [batchName, nuc]
pct_row.extend(nuc_pcts[nuc])
nucleotide_percentage_summary.append(pct_row)
for mod in [
'Insertions', 'Insertions_Left', 'Deletions',
'Substitutions', 'All_modifications'
]:
row = [batchName, mod]
row.extend(mod_freqs[mod])
modification_frequency_summary.append(row)
pct_row = [batchName, mod]
pct_row.extend(mod_pcts[mod])
modification_percentage_summary.append(pct_row)
if amp_found_count == 0:
info(
"Couldn't find any data for amplicon '%s'. Not compiling results."
% amplicon_name)
else:
amplicon_plot_name = amplicon_name + "."
if len(amplicon_names) == 1 and amplicon_name == "Reference":
amplicon_plot_name = ""
colnames = ['Batch', 'Nucleotide']
colnames.extend(list(consensus_sequence))
nucleotide_frequency_summary_df = pd.DataFrame(
nucleotide_frequency_summary, columns=colnames)
nucleotide_frequency_summary_df = pd.concat([
nucleotide_frequency_summary_df.iloc[:, 0:2],
nucleotide_frequency_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
nucleotide_frequency_summary_filename = _jp(
amplicon_plot_name + 'Nucleotide_frequency_summary.txt')
nucleotide_frequency_summary_df.to_csv(
nucleotide_frequency_summary_filename,
sep='\t',
index=None)
nucleotide_percentage_summary_df = pd.DataFrame(
nucleotide_percentage_summary, columns=colnames)
nucleotide_percentage_summary_df = pd.concat([
nucleotide_percentage_summary_df.iloc[:, 0:2],
nucleotide_percentage_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
nucleotide_percentage_summary_filename = _jp(
amplicon_plot_name + 'Nucleotide_percentage_summary.txt')
nucleotide_percentage_summary_df.to_csv(
nucleotide_percentage_summary_filename,
sep='\t',
index=None)
colnames = ['Batch', 'Modification']
colnames.extend(list(consensus_sequence))
modification_frequency_summary_df = pd.DataFrame(
modification_frequency_summary, columns=colnames)
modification_frequency_summary_df = pd.concat([
modification_frequency_summary_df.iloc[:, 0:2],
modification_frequency_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
modification_frequency_summary_filename = _jp(
amplicon_plot_name + 'MODIFICATION_FREQUENCY_SUMMARY.txt')
modification_frequency_summary_df.to_csv(
modification_frequency_summary_filename,
sep='\t',
index=None)
modification_percentage_summary_df = pd.DataFrame(
modification_percentage_summary, columns=colnames)
modification_percentage_summary_df = pd.concat([
modification_percentage_summary_df.iloc[:, 0:2],
modification_percentage_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
modification_percentage_summary_filename = _jp(
amplicon_plot_name + 'MODIFICATION_PERCENTAGE_SUMMARY.txt')
modification_percentage_summary_df.to_csv(
modification_percentage_summary_filename,
sep='\t',
index=None)
crispresso2_info[
'nucleotide_frequency_summary_filename'] = os.path.basename(
nucleotide_frequency_summary_filename)
crispresso2_info[
'nucleotide_percentage_summary_filename'] = os.path.basename(
nucleotide_percentage_summary_filename)
crispresso2_info[
'modification_frequency_summary_filename'] = os.path.basename(
modification_frequency_summary_filename)
crispresso2_info[
'modification_percentage_summary_filename'] = os.path.basename(
modification_percentage_summary_filename)
crispresso2_info['summary_plot_titles'] = {}
crispresso2_info['summary_plot_labels'] = {}
crispresso2_info['summary_plot_datas'] = {}
#if guides are all the same, merge substitutions and perform base editor comparison at guide quantification window
if guides_all_same and consensus_guides != []:
info(
"All guides are equal. Performing comparison of batches for amplicon '%s'"
% amplicon_name)
include_idxs = consensus_include_idxs #include indexes are the same for all guides
for idx, sgRNA in enumerate(consensus_guides):
sgRNA_intervals = consensus_sgRNA_intervals[idx]
sgRNA_plot_idxs = consensus_sgRNA_plot_idxs[idx]
plot_idxs_flat = [0, 1] # guide, nucleotide
plot_idxs_flat.extend(
[plot_idx + 2 for plot_idx in sgRNA_plot_idxs])
sub_nucleotide_frequency_summary_df = nucleotide_frequency_summary_df.iloc[:,
plot_idxs_flat]
sub_nucleotide_percentage_summary_df = nucleotide_percentage_summary_df.iloc[:,
plot_idxs_flat]
sub_modification_percentage_summary_df = modification_percentage_summary_df.iloc[:,
plot_idxs_flat]
#show all sgRNA's on the plot
sub_sgRNA_intervals = []
for sgRNA_interval in consensus_sgRNA_intervals:
newstart = None
newend = None
for idx, i in enumerate(sgRNA_plot_idxs):
if i <= sgRNA_interval[0]:
newstart = idx
if newend is None and i >= sgRNA_interval[1]:
newend = idx
#if guide doesn't overlap with plot idxs
if newend == 0 or newstart == len(sgRNA_plot_idxs):
continue
#otherwise, correct partial overlaps
elif newstart == None and newend == None:
newstart = 0
newend = len(include_idxs) - 1
elif newstart == None:
newstart = 0
elif newend == None:
newend = len(include_idxs) - 1
#and add it to the list
sub_sgRNA_intervals.append((newstart, newend))
if not args.suppress_plots:
#plot for each guide
this_window_nuc_pct_quilt_plot_name = _jp(
amplicon_plot_name +
'Nucleotide_percentage_quilt_around_sgRNA_' +
sgRNA)
CRISPRessoPlot.plot_nucleotide_quilt(
sub_nucleotide_percentage_summary_df,
sub_modification_percentage_summary_df,
this_window_nuc_pct_quilt_plot_name,
save_png,
sgRNA_intervals=sub_sgRNA_intervals,
quantification_window_idxs=include_idxs)
plot_name = os.path.basename(
this_window_nuc_pct_quilt_plot_name)
window_nuc_pct_quilt_plot_names.append(plot_name)
crispresso2_info['summary_plot_titles'][
plot_name] = 'sgRNA: ' + sgRNA + ' Amplicon: ' + amplicon_name
if len(consensus_guides) == 1:
crispresso2_info['summary_plot_titles'][
plot_name] = ''
crispresso2_info['summary_plot_labels'][
plot_name] = 'Composition of each base around the guide ' + sgRNA + ' for the amplicon ' + amplicon_name
crispresso2_info['summary_plot_datas'][plot_name] = [
('Nucleotide frequencies',
os.path.basename(
nucleotide_frequency_summary_filename)),
('Modification frequencies',
os.path.basename(
modification_frequency_summary_filename))
]
sub_nucleotide_frequency_summary_df = pd.concat(
[
sub_nucleotide_frequency_summary_df.
iloc[:, 0:2],
sub_nucleotide_frequency_summary_df.
iloc[:, 2:].apply(pd.to_numeric)
],
axis=1)
sub_nucleotide_frequency_summary_filename = _jp(
amplicon_plot_name +
'Nucleotide_frequency_summary_around_sgRNA_' +
sgRNA + '.txt')
sub_nucleotide_frequency_summary_df.to_csv(
sub_nucleotide_frequency_summary_filename,
sep='\t',
index=None)
sub_nucleotide_percentage_summary_df = pd.concat(
[
sub_nucleotide_percentage_summary_df.
iloc[:, 0:2],
sub_nucleotide_percentage_summary_df.
iloc[:, 2:].apply(pd.to_numeric)
],
axis=1)
sub_nucleotide_percentage_summary_filename = _jp(
amplicon_plot_name +
'Nucleotide_percentage_summary_around_sgRNA_' +
sgRNA + '.txt')
sub_nucleotide_percentage_summary_df.to_csv(
sub_nucleotide_percentage_summary_filename,
sep='\t',
index=None)
if args.base_editor_output:
this_window_nuc_conv_plot_name = _jp(
amplicon_plot_name +
'Nucleotide_conversion_map_around_sgRNA_' +
sgRNA)
CRISPRessoPlot.plot_conversion_map(
sub_nucleotide_percentage_summary_df,
this_window_nuc_conv_plot_name,
args.conversion_nuc_from,
args.conversion_nuc_to,
save_png,
sgRNA_intervals=sub_sgRNA_intervals,
quantification_window_idxs=include_idxs)
plot_name = os.path.basename(
this_window_nuc_conv_plot_name)
window_nuc_conv_plot_names.append(plot_name)
crispresso2_info['summary_plot_titles'][
plot_name] = 'sgRNA: ' + sgRNA + ' Amplicon: ' + amplicon_name
if len(consensus_guides) == 1:
crispresso2_info['summary_plot_titles'][
plot_name] = ''
crispresso2_info['summary_plot_labels'][
plot_name] = args.conversion_nuc_from + '->' + args.conversion_nuc_to + ' conversion rates around the guide ' + sgRNA + ' for the amplicon ' + amplicon_name
crispresso2_info['summary_plot_datas'][
plot_name] = [
('Nucleotide frequencies around sgRNA',
os.path.basename(
sub_nucleotide_frequency_summary_filename
)),
('Nucleotide percentages around sgRNA',
os.path.basename(
sub_nucleotide_percentage_summary_filename
))
]
if not args.suppress_plots: # plot the whole region
this_nuc_pct_quilt_plot_name = _jp(
amplicon_plot_name + 'Nucleotide_percentage_quilt')
CRISPRessoPlot.plot_nucleotide_quilt(
nucleotide_percentage_summary_df,
modification_percentage_summary_df,
this_nuc_pct_quilt_plot_name,
save_png,
sgRNA_intervals=consensus_sgRNA_intervals,
quantification_window_idxs=include_idxs)
plot_name = os.path.basename(
this_nuc_pct_quilt_plot_name)
nuc_pct_quilt_plot_names.append(plot_name)
crispresso2_info['summary_plot_titles'][
plot_name] = 'Amplicon: ' + amplicon_name
if len(amplicon_names) == 1:
crispresso2_info['summary_plot_titles'][
plot_name] = ''
crispresso2_info['summary_plot_labels'][
plot_name] = 'Composition of each base for the amplicon ' + amplicon_name
crispresso2_info['summary_plot_datas'][plot_name] = [
('Nucleotide frequencies',
os.path.basename(
nucleotide_frequency_summary_filename)),
('Modification frequencies',
os.path.basename(
modification_frequency_summary_filename))
]
if args.base_editor_output:
this_nuc_conv_plot_name = _jp(
amplicon_plot_name +
'Nucleotide_conversion_map')
CRISPRessoPlot.plot_conversion_map(
nucleotide_percentage_summary_df,
this_nuc_conv_plot_name,
args.conversion_nuc_from,
args.conversion_nuc_to,
save_png,
sgRNA_intervals=consensus_sgRNA_intervals,
quantification_window_idxs=include_idxs)
plot_name = os.path.basename(
this_nuc_conv_plot_name)
nuc_conv_plot_names.append(plot_name)
crispresso2_info['summary_plot_titles'][
plot_name] = 'Amplicon: ' + amplicon_name
if len(amplicon_names) == 1:
crispresso2_info['summary_plot_titles'][
plot_name] = ''
crispresso2_info['summary_plot_titles'][
plot_name] = ''
crispresso2_info['summary_plot_labels'][
plot_name] = args.conversion_nuc_from + '->' + args.conversion_nuc_to + ' conversion rates for the amplicon ' + amplicon_name
crispresso2_info['summary_plot_datas'][plot_name] = [
('Nucleotide frequencies',
os.path.basename(
nucleotide_frequency_summary_filename)),
('Modification frequencies',
os.path.basename(
modification_frequency_summary_filename))
]
else: #guides are not the same
if not args.suppress_plots:
this_nuc_pct_quilt_plot_name = _jp(
amplicon_plot_name + 'Nucleotide_percentage_quilt')
CRISPRessoPlot.plot_nucleotide_quilt(
nucleotide_percentage_summary_df,
modification_percentage_summary_df,
this_nuc_pct_quilt_plot_name, save_png)
plot_name = os.path.basename(
this_nuc_pct_quilt_plot_name)
nuc_pct_quilt_plot_names.append(plot_name)
crispresso2_info['summary_plot_labels'][
plot_name] = 'Composition of each base for the amplicon ' + amplicon_name
crispresso2_info['summary_plot_datas'][plot_name] = [
('Nucleotide frequencies',
os.path.basename(
nucleotide_frequency_summary_filename)),
('Modification frequencies',
os.path.basename(
modification_frequency_summary_filename))
]
if args.base_editor_output:
this_nuc_conv_plot_name = _jp(
amplicon_plot_name +
'Nucleotide_percentage_quilt')
CRISPRessoPlot.plot_conversion_map(
nucleotide_percentage_summary_df,
this_nuc_conv_plot_name,
args.conversion_nuc_from,
args.conversion_nuc_to, save_png)
plot_name = os.path.basename(
this_nuc_conv_plot_name)
nuc_conv_plot_names.append(plot_name)
crispresso2_info['summary_plot_labels'][
plot_name] = args.conversion_nuc_from + '->' + args.conversion_nuc_to + ' conversion rates for the amplicon ' + amplicon_name
crispresso2_info['summary_plot_datas'][plot_name] = [
('Nucleotide frequencies',
os.path.basename(
nucleotide_frequency_summary_filename)),
('Modification frequencies',
os.path.basename(
modification_frequency_summary_filename))
]
crispresso2_info[
'window_nuc_pct_quilt_plot_names'] = window_nuc_pct_quilt_plot_names
crispresso2_info['nuc_pct_quilt_plot_names'] = nuc_pct_quilt_plot_names
crispresso2_info[
'window_nuc_conv_plot_names'] = window_nuc_conv_plot_names
crispresso2_info['nuc_conv_plot_names'] = nuc_conv_plot_names
#summarize amplicon modifications
with open(
_jp('CRISPRessoBatch_quantification_of_editing_frequency.txt'),
'w') as outfile:
wrote_header = False
for idx, row in batch_params.iterrows():
batchName = CRISPRessoShared.slugify(row["name"])
file_prefix = row['file_prefix']
folder_name = os.path.join(OUTPUT_DIRECTORY,
'CRISPResso_on_%s' % batchName)
run_data = run_datas[idx]
if run_data is None:
continue
amplicon_modification_file = os.path.join(
folder_name, run_data['quant_of_editing_freq_filename'])
with open(amplicon_modification_file, 'r') as infile:
file_head = infile.readline()
if not wrote_header:
outfile.write('Batch\t' + file_head)
wrote_header = True
for line in infile:
outfile.write(batchName + "\t" + line)
#summarize alignment
with open(_jp('CRISPRessoBatch_mapping_statistics.txt'),
'w') as outfile:
wrote_header = False
for idx, row in batch_params.iterrows():
batchName = CRISPRessoShared.slugify(row["name"])
folder_name = os.path.join(OUTPUT_DIRECTORY,
'CRISPResso_on_%s' % batchName)
run_data = run_datas[idx]
if run_data is None:
continue
amplicon_modification_file = os.path.join(
folder_name, run_data['mapping_stats_filename'])
with open(amplicon_modification_file, 'r') as infile:
file_head = infile.readline()
if not wrote_header:
outfile.write('Batch\t' + file_head)
wrote_header = True
for line in infile:
outfile.write(batchName + "\t" + line)
if not args.suppress_report:
if (args.place_report_in_output_folder):
report_name = _jp("CRISPResso2Batch_report.html")
else:
report_name = OUTPUT_DIRECTORY + '.html'
CRISPRessoReport.make_batch_report_from_folder(
report_name, crispresso2_info, OUTPUT_DIRECTORY, _ROOT)
crispresso2_info['report_location'] = report_name
crispresso2_info['report_filename'] = os.path.basename(report_name)
cp.dump(crispresso2_info, open(crispresso2Batch_info_file, 'wb'))
info('Analysis Complete!')
print(CRISPRessoShared.get_crispresso_footer())
sys.exit(0)
except Exception as e:
debug_flag = False
if 'args' in vars() and 'debug' in args:
debug_flag = args.debug
if debug_flag:
traceback.print_exc(file=sys.stdout)
error('\n\nERROR: %s' % e)
sys.exit(-1)
def report_nucleotide_summary(amplicon_seq, amplicon_name,
amplicon_index):
consensus_sequence = ""
nucleotide_frequency_summary = []
nucleotide_percentage_summary = []
modification_frequency_summary = []
modification_percentage_summary = []
amp_found_count = 0 #how many folders had information for this amplicon
consensus_guides = []
consensus_include_idxs = []
consensus_sgRNA_intervals = []
guides_all_same = True
batches_with_this_amplicon = []
for idx, row in batch_params.iterrows():
batchName = CRISPRessoShared.slugify(row["name"])
file_prefix = row['file_prefix']
folder_name = os.path.join(OUTPUT_DIRECTORY,
'CRISPResso_on_%s' % batchName)
run_data = run_datas[idx]
if run_data is None:
continue
batch_has_amplicon = False
batch_amplicon_name = ''
for ref_name in run_data['ref_names']:
if amplicon_seq == run_data['refs'][ref_name]['sequence']:
batch_has_amplicon = True
batch_amplicon_name = ref_name
if not batch_has_amplicon:
continue
batches_with_this_amplicon.append(idx)
if consensus_guides == []:
consensus_guides = run_data['refs'][batch_amplicon_name][
'sgRNA_sequences']
consensus_include_idxs = run_data['refs'][
batch_amplicon_name]['include_idxs']
consensus_sgRNA_intervals = run_data['refs'][
batch_amplicon_name]['sgRNA_intervals']
if run_data['refs'][batch_amplicon_name][
'sgRNA_sequences'] != consensus_guides:
guides_all_same = False
if 'nuc_freq_filename' not in run_data['refs'][
batch_amplicon_name]:
info(
"Skipping the amplicon '%s' in folder '%s'. Cannot find nucleotide information."
% (batch_amplicon_name, folder_name))
continue
nucleotide_frequency_file = run_data['refs'][
batch_amplicon_name]['nuc_freq_filename']
ampSeq_nf, nuc_freqs = CRISPRessoShared.parse_count_file(
nucleotide_frequency_file)
nucleotide_pct_file = run_data['refs'][batch_amplicon_name][
'nuc_pct_filename']
ampSeq_np, nuc_pcts = CRISPRessoShared.parse_count_file(
nucleotide_pct_file)
count_file = run_data['refs'][batch_amplicon_name][
'mod_count_filename']
ampSeq_cf, mod_freqs = CRISPRessoShared.parse_count_file(
count_file)
if ampSeq_nf is None or ampSeq_np is None or ampSeq_cf is None:
info(
"Skipping the amplicon '%s' in folder '%s'. Could not parse batch output."
% (batch_amplicon_name, folder_name))
info(
"Nucleotide frequency amplicon: '%s', Nucleotide percentage amplicon: '%s', Count vectors amplicon: '%s'"
% (ampSeq_nf, ampSeq_np, ampSeq_cf))
continue
if ampSeq_nf != ampSeq_np or ampSeq_np != ampSeq_cf:
warn(
"Skipping the amplicon '%s' in folder '%s'. Parsed amplicon sequences do not match\nnf:%s\nnp:%s\ncf:%s\nrf:%s"
% (batch_amplicon_name, folder_name, ampSeq_nf,
ampSeq_np, ampSeq_cf, amplicon_seq))
continue
if consensus_sequence == "":
consensus_sequence = ampSeq_nf
if ampSeq_nf != consensus_sequence:
info(
"Skipping the amplicon '%s' in folder '%s'. Amplicon sequences do not match."
% (batch_amplicon_name, folder_name))
continue
if 'Total' not in mod_freqs:
info(
"Skipping the amplicon '%s' in folder '%s'. Processing did not complete."
% (batch_amplicon_name, folder_name))
continue
if mod_freqs['Total'][0] == 0 or mod_freqs['Total'][0] == "0":
info(
"Skipping the amplicon '%s' in folder '%s'. Got no reads for amplicon."
% (batch_amplicon_name, folder_name))
continue
if (args.min_reads_for_inclusion is not None) and (int(
mod_freqs['Total'][0]) < args.min_reads_for_inclusion):
info(
"Skipping the amplicon '%s' in folder '%s'. Got %s reads (min_reads_for_inclusion is %d)."
% (batch_amplicon_name, folder_name,
str(mod_freqs['Total'][0]),
args.min_reads_for_inclusion))
continue
mod_pcts = {}
for key in mod_freqs:
mod_pcts[key] = np.array(mod_freqs[key]).astype(
np.float) / float(mod_freqs['Total'][0])
amp_found_count += 1
for nuc in ['A', 'T', 'C', 'G', 'N', '-']:
row = [batchName, nuc]
row.extend(nuc_freqs[nuc])
nucleotide_frequency_summary.append(row)
pct_row = [batchName, nuc]
pct_row.extend(nuc_pcts[nuc])
nucleotide_percentage_summary.append(pct_row)
for mod in [
'Insertions', 'Insertions_Left', 'Deletions',
'Substitutions', 'All_modifications'
]:
row = [batchName, mod]
row.extend(mod_freqs[mod])
modification_frequency_summary.append(row)
pct_row = [batchName, mod]
pct_row.extend(mod_pcts[mod])
modification_percentage_summary.append(pct_row)
if amp_found_count == 0:
info(
"Couldn't find any data for amplicon '%s'. Not compiling results."
% amplicon_name)
return ()
colnames = ['Batch', 'Nucleotide']
colnames.extend(list(consensus_sequence))
nucleotide_frequency_summary_df = pd.DataFrame(
nucleotide_frequency_summary, columns=colnames)
nucleotide_frequency_summary_df = pd.concat([
nucleotide_frequency_summary_df.iloc[:, 0:2],
nucleotide_frequency_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
nucleotide_frequency_summary_df.to_csv(
_jp(amplicon_name + '.NUCLEOTIDE_FREQUENCY_SUMMARY.txt'),
sep='\t',
index=None)
nucleotide_percentage_summary_df = pd.DataFrame(
nucleotide_percentage_summary, columns=colnames)
nucleotide_percentage_summary_df = pd.concat([
nucleotide_percentage_summary_df.iloc[:, 0:2],
nucleotide_percentage_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
nucleotide_percentage_summary_df.to_csv(
_jp(amplicon_name + '.NUCLEOTIDE_PERCENTAGE_SUMMARY.txt'),
sep='\t',
index=None)
colnames = ['Batch', 'Modification']
colnames.extend(list(consensus_sequence))
modification_frequency_summary_df = pd.DataFrame(
modification_frequency_summary, columns=colnames)
modification_frequency_summary_df = pd.concat([
modification_frequency_summary_df.iloc[:, 0:2],
modification_frequency_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
modification_frequency_summary_df.to_csv(
_jp(amplicon_name + '.MODIFICATION_FREQUENCY_SUMMARY.txt'),
sep='\t',
index=None)
modification_percentage_summary_df = pd.DataFrame(
modification_percentage_summary, columns=colnames)
modification_percentage_summary_df = pd.concat([
modification_percentage_summary_df.iloc[:, 0:2],
modification_percentage_summary_df.iloc[:, 2:].apply(
pd.to_numeric)
],
axis=1)
modification_percentage_summary_df.to_csv(
_jp(amplicon_name + '.MODIFICATION_PERCENTAGE_SUMMARY.txt'),
sep='\t',
index=None)
#if guides are all the same, merge substitutions and perform base editor comparison at guide quantification window
if guides_all_same and consensus_guides != []:
include_idxs = consensus_include_idxs
sgRNA_intervals = consensus_sgRNA_intervals
info(
"All guides are equal. Performing comparison of batches for amplicon '%s'"
% amplicon_name)
include_idxs_flat = [0, 1] # guide, nucleotide
include_idxs_flat.extend(
[cutidx + 2 for cutidx in include_idxs])
sub_nucleotide_frequency_summary_df = nucleotide_frequency_summary_df.iloc[:,
include_idxs_flat]
sub_nucleotide_percentage_summary_df = nucleotide_percentage_summary_df.iloc[:,
include_idxs_flat]
sub_modification_percentage_summary_df = modification_percentage_summary_df.iloc[:,
include_idxs_flat]
sub_sgRNA_intervals = []
for sgRNA_interval in sgRNA_intervals:
newstart = None
newend = None
for idx, i in enumerate(include_idxs):
if i <= sgRNA_interval[0]:
newstart = idx
if newend is None and i >= sgRNA_interval[1]:
newend = idx
#if guide doesn't overlap with include indexes
if newend == 0 or newstart == len(include_idxs):
continue
#otherwise, correct partial overlaps
elif newstart == None and newend == None:
newstart = 0
newend = len(include_idxs) - 1
elif newstart == None:
newstart = 0
elif newend == None:
newend = len(include_idxs) - 1
#and add it to the list
sub_sgRNA_intervals.append((newstart, newend))
if not args.suppress_plots:
CRISPRessoPlot.plot_nucleotide_quilt(
sub_nucleotide_percentage_summary_df,
sub_modification_percentage_summary_df,
_jp(amplicon_name +
'.Quantification_Window_Nucleotide_Percentage_Quilt'
),
save_png,
sgRNA_intervals=sub_sgRNA_intervals)
if args.base_editor_output:
CRISPRessoPlot.plot_conversion_map(
sub_nucleotide_percentage_summary_df,
_jp(amplicon_name +
'.Quantification_Window_Nucleotide_Conversion'
),
args.conversion_nuc_from,
args.conversion_nuc_to,
save_png,
sgRNA_intervals=sub_sgRNA_intervals)
CRISPRessoPlot.plot_nucleotide_quilt(
nucleotide_percentage_summary_df,
modification_percentage_summary_df,
_jp(amplicon_name + '.Nucleotide_Percentage_Quilt'),
save_png,
sgRNA_intervals=sgRNA_intervals,
quantification_window_idxs=include_idxs)
if args.base_editor_output:
CRISPRessoPlot.plot_conversion_map(
nucleotide_percentage_summary_df,
_jp(amplicon_name + '.Nucleotide_Conversion'),
args.conversion_nuc_from,
args.conversion_nuc_to,
save_png,
sgRNA_intervals=sgRNA_intervals)
else: #guides are not the same
if not args.suppress_plots:
CRISPRessoPlot.plot_nucleotide_quilt(
nucleotide_percentage_summary_df,
modification_percentage_summary_df,
_jp(amplicon_name + '.Nucleotide_Percentage_Quilt'),
save_png)
if args.base_editor_output:
CRISPRessoPlot.plot_conversion_map(
nucleotide_percentage_summary_df,
_jp(amplicon_name + '.Nucleotide_Conversion'),
args.conversion_nuc_from, args.conversion_nuc_to,
save_png)