def seq_to_seq_map(dict_or_file_or_dataframe):
"""Takes a dictionary (in which case it does not modify it) or a dataframe,
or an excel file name and returns a dict mapping segment names to sequences.
Columns must be PID and Sequence"""
import pandas as pd
#if is dictionary just return it
if isinstance(dict_or_file_or_dataframe, dict):
return dict_or_file_or_dataframe
#if is file, load it to a dataframe
if u.is_str(dict_or_file_or_dataframe):
df = pd.read_excel(dict_or_file_or_dataframe)
else:
df = dict_or_file_or_dataframe
#else it must be a dataframe. Do your magic:)
df = df.dropna()
#cleanup whitespace
df.PID = df.PID.str.strip()
peps = {}
for n, pep in df.iterrows():
peps[pep.PID] = pep.Sequence.strip()
return peps
def get_annotated_sequence(segments,
seg_to_seq,
linkers="GSGPG",
N_tag="",
C_tag=""):
"""Returns an anottated amino acid sequnce of the poylhedra.
Linkers can be a single string or an array of string with the required number of linkers.
N_tag, C_tag - are appended to the left and right side of the string"""
seg_to_seq = seq_to_seq_map(seg_to_seq)
N = len(segments)
if u.is_str(linkers):
linkers = [linkers] * (N - 1)
assert len(linkers) == N - 1, (
"Length of linkers must be one less than the number of segments." +
"Is {NL}, but should be {N}".format(NL=len(linkers), N=N))
max_seg_len = max([len(seg_to_seq[s]) for s in segments])
aa_segments = [
seg_to_seq[s].ljust(max_seg_len).replace('-', '') + "\t|" + s
for s in segments
]
lines = [N_tag] + list(u.roundrobin(aa_segments, linkers)) + [C_tag]
lines = "\n".join(lines)
return lines
def load_pairs(yaml_str):
"""Loads a yaml string and returns a list of pairs"""
pairs = yaml.load(yaml_str)
#split the pairs and chains if they are seperated by a colon
for n in range(len(pairs)):
#if it's a calumn split it
if u.is_str(pairs[n]['pair']):
pairs[n]['pair'] = pairs[n]['pair'].split(':')
#strip whitespace
pairs[n]['pair'] = [p.strip() for p in pairs[n]['pair']]
#do the same for chains
if u.is_str(pairs[n]['chains']):
pairs[n]['chains'] = pairs[n]['chains'].split(':')
#strip whitespace
pairs[n]['chains'] = [c.strip() for c in pairs[n]['chains']]
return pairs
def do_assignment_replacements(topology, assignments):
"""Replaces topology segments with replacemnt rules.
The first occurance of a name is replaced witht he first element in the list"""
if u.is_str(assignments):
assignments = segment_assignments_to_dict(assignments)
else:
assignments = assignments.copy()
keys = assignments.keys()
ret = []
for s in topology:
sl = s.upper()
if sl in keys:
ret.append(assignments[sl].pop(0))
else:
ret.append(s)
return ret
def test_is_str():
assert u.is_str("A string")
assert u.is_str(u"A string")
assert not u.is_str({})
assert not u.is_str([])
assert not u.is_str(1)