def process_overlaps(in_f, min_overlap=6):
print("Reading peptides from %s.." % in_f)
if in_f.endswith('.csv'):
peptides = read_peptides_from_csv(in_f)
elif in_f.endswith('.txt'):
peptides = read_peptides_from_txt(in_f)
print("Generating groups...")
make_overlap_groups(peptides, 'groups.yaml', min_overlap)
groups = datafile.load_yaml('groups.yaml')
base = os.path.splitext(in_f)[0]
out_csv = base + '.cluster.csv'
print("Writing groups %s" % out_csv)
write_groups(groups, out_csv)
out_csv = base + '.kernel.csv'
print("Writing groups %s" % out_csv)
write_overlap_kernels(groups, out_csv)
def process_subsets(in_f):
"""
need to look for redundancies
"""
print("Reading peptides from %s.." % in_f)
if in_f.endswith('.csv'):
peptides = read_peptides_from_csv(in_f)
elif in_f.endswith('.txt'):
peptides = read_peptides_from_txt(in_f)
print("Generating groups...")
make_subset_groups(peptides, 'groups.yaml')
groups = datafile.load_yaml('groups.yaml')
base = os.path.splitext(in_f)[0]
out_csv = base + '.cluster.csv'
print("Writing groups %s" % out_csv)
write_groups(groups, out_csv)
out_csv = base + '.kernel.csv'
print("Writing groups %s" % out_csv)
write_subset_kernels(groups, out_csv)
if seq in ref_seq:
return ref_seq.find(seq)
elif ref_seq in seq:
return -seq.find(ref_seq)
elif ref_left_overlap:
return len(ref_seq) - ref_left_overlap
elif ref_right_overlap:
return -(len(seq) - ref_right_overlap)
raise "No overlap"
print("Generating groups...")
make_groups('b57_clean.csv', 'groups.yaml')
print("Loading groups...")
groups = datafile.load_yaml('groups.yaml')
rows = [('i_group', 'sequence', 'modifications', 'protein')]
for group in groups:
sequences = []
for peptide in group['peptides']:
sequences.append(peptide['sequence'])
ref_seq = sequences[0]
indices = [get_i_relative_to_ref(ref_seq, s) for s in sequences]
# find max length
left = -min(indices)
max_len = 0
for index, seq in zip(indices, sequences):
if seq in ref_seq:
return ref_seq.find(seq)
elif ref_seq in seq:
return -seq.find(ref_seq)
elif ref_left_overlap:
return len(ref_seq) - ref_left_overlap
elif ref_right_overlap:
return -(len(seq) - ref_right_overlap)
raise "No overlap"
print("Generating groups...")
make_groups('b57_clean.csv', 'groups.yaml')
print("Loading groups...")
groups = datafile.load_yaml('groups.yaml')
rows = [('i_group', 'sequence', 'modifications', 'protein')]
for group in groups:
sequences = []
for peptide in group['peptides']:
sequences.append(peptide['sequence'])
ref_seq = sequences[0]
indices = [get_i_relative_to_ref(ref_seq, s) for s in sequences]
# find max length
left = -min(indices)
max_len = 0
for index, seq in zip(indices, sequences):
