def test_sequence_util():
sequence = 'ACTG'
rna_sequence = 'ACUG'
transcribed = transcribe(sequence)
print(transcribed)
reverse_comp = reverse_complement(sequence)
print(reverse_comp)
reverse_transcribed_seq = reverse_transcribe(rna_sequence)
print(reverse_transcribed_seq)
assert is_nuc_acid('A')
assert not is_nuc_acid('L')
assert not is_nuc_acid(None)
assert RNA2DNA('U') == 'T'
assert DNA2RNA('T') == 'U'
try:
RNA2DNA('L')
except Exception as e:
assert 'Input sequence L is not valid nucleic acid sequence' in str(e)
def test_is_nuc_acid():
good_seq = "GCCAat"
assert sequence.is_nuc_acid(good_seq)
bad_seq = "AAAAXGG"
assert not sequence.is_nuc_acid(bad_seq)
not_str = 1234
assert not sequence.is_nuc_acid(not_str)
def multiple_sanger_analysis(definition_file,
output_dir,
data_dir=None,
verbose=False,
single_line=None,
allprops=False):
'''
:param definition_file: input excel file that defines sample/control/data associations
:param output_dir: output directory
:return:
'''
input_df = pd.read_excel(definition_file)
results = []
fails = []
jobs = []
n = 0
for m, experiment in input_df.iterrows():
label = experiment['Label']
base_outputname = os.path.join(output_dir, '%s-%s' % (n, label))
control_sequence_file = experiment['Control File']
edit_sequence_file = experiment['Experiment File']
guide = experiment['Guide Sequence']
if 'Donor' in experiment and is_nuc_acid(experiment['Donor']):
donor = experiment['Donor']
else:
donor = None
print(donor)
try:
if pd.isnull(control_sequence_file):
raise IOError(
"Control filepath not specified at line {} in definition file"
.format(n + 1))
if pd.isnull(edit_sequence_file):
raise IOError(
"Edit filepath not specified at line {} in definition file"
.format(n + 1))
control_sequence_path = os.path.join(data_dir,
control_sequence_file)
edit_sequence_path = os.path.join(data_dir, edit_sequence_file)
if single_line is not None:
if n != single_line:
continue
msg = "analyzing"
print("-" * 50, msg, n, experiment['Label'], guide)
job_args = (control_sequence_path, edit_sequence_path,
base_outputname, guide)
job_kwargs = {
'verbose': verbose,
'allprops': allprops,
'donor': donor
}
result = single_sanger_analysis(*job_args, **job_kwargs)
jobs.append((experiment, result))
except Exception as e:
fails.append(experiment)
print("Single Sanger analysis failed", e)
import traceback, sys
traceback.print_exc(file=sys.stdout)
n += 1
for job in jobs:
r = job[1]
experiment = job[0]
if r is not None:
tmp = [
experiment['Label'], r['ice'], r['ice_d'], r['rsq'],
r['hdr_pct'], r['guides'], r['notes']
]
else:
tmp = [experiment['Label'], 'Failed', '', '', '', '', '']
results.append(tmp)
if results:
input_df = pd.DataFrame(results)
timestamp = '{:%Y-%m-%d-%H%M%S}'.format(datetime.datetime.now())
out_file = os.path.join(output_dir,
"ice.results.{}.xlsx".format(timestamp))
header = [
"sample_name", "ice", 'ice_d', "r_squared", "hdr_pct", "guides",
"notes"
]
input_df.columns = header
# to json
out_dict = []
for r in results:
row = {}
for idx, c in enumerate(header):
row[c] = r[idx]
out_dict.append(row)
with open(out_file.replace('.xlsx', '.json'), 'w') as f:
json.dump(out_dict, f, ensure_ascii=False)
with pd.ExcelWriter(out_file) as writer:
input_df.to_excel(writer, sheet_name="Results")
md = {'version': __version__}
metadata = pd.DataFrame.from_dict([md])
metadata.to_excel(writer, sheet_name='Metadata')
writer.save()
return out_dict
else:
print("None of the samples were able to be analyzed")
return False