def test_StockholmParser_strict_invalid_structure(self):
"""StockholmParser: functions when toggling strict record w/ bad struct
"""
# strict
self.assertRaises(RecordError,list,\
StockholmParser(self._fake_record_bad_structure_2))
#not strict
self.assertEqual(list(StockholmParser(\
self._fake_record_bad_structure_2,\
strict=False)),[])
def test_StockholmParser_strict_missing_fields(self):
"""StockholmParser: toggle strict functions correctly """
# strict = True
self.assertRaises(RecordError,list,\
StockholmParser(self._fake_record_no_headers))
# strict = False
self.assertEqual(list(StockholmParser(self._fake_record_no_headers,\
strict=False)), [])
self.assertEqual(list(StockholmParser(self._fake_record_no_sequences,\
strict=False)), [])
def test_StockholmParser_strict_invalid_headers(self):
"""StockholmParser: functions when toggling strict record w/ bad header
"""
self.assertRaises(RecordError,list,\
StockholmParser(self._fake_record_bad_header_1))
self.assertRaises(RecordError,list,\
StockholmParser(self._fake_record_bad_header_2))
# strict = False
x = list(StockholmParser(self._fake_record_bad_header_1, strict=False))
obs = list(StockholmParser(self._fake_record_bad_header_1,\
strict=False))[0].Info.GF.keys()
self.assertEqual(len(obs), 1)
obs = list(StockholmParser(self._fake_record_bad_header_2,\
strict=False))[0].Info.GF.keys()
self.assertEqual(len(obs), 1)
def test_StockholmParser_strict_invalid_structure(self):
"""StockholmParser: toggle strict functions w/ invalid structure
"""
#strict = True
self.assertRaises(RecordError,list,\
StockholmParser(self._fake_record_bad_structure_1))
# strict = False
self.assertEqual(list(MinimalStockholmParser(\
self._fake_record_bad_structure_1,strict=False))[0][2],None)
def test_StockholmParser_strict_invalid_sequences(self):
"""StockholmParser: functions when toggling strict w/ record w/ bad seq
"""
self.assertRaises(
RecordError, list,
MinimalStockholmParser(self._fake_record_bad_sequence_1))
# strict = False
# in 'False' mode you expect to get back as much as possible, also
# parts of sequences
self.assertEqual(len(list(StockholmParser(\
self._fake_record_bad_sequence_1,\
strict=False))[0].NamedSeqs), 3)
def test_StockholmParser_single_family(self):
"""StockholmParser: should work on a family in stockholm format"""
exp_header = {}
exp_aln = {'K02120.1/628-682':\
'AUGGGAAAUUCCCCCUCCUAUAACCCCCCCGCUGGUAUCUCCCCCUCAGACUGGC',\
'D00647.1/629-683':\
'AUGGGAAACUCCCCCUCCUAUAACCCCCCCGCUGGCAUCUCCCCCUCAGACUGGC'}
exp_struct = '<<<<<<.........>>>>>>.........<<<<<<.............>>>>>>'
aln = list(StockholmParser(self.single_family))[0]
h = aln.Info['GF']
a = aln
s = aln.Info['Struct']
self.assertEqual(h, exp_header)
self.assertEqual(a, exp_aln)
self.assertEqual(s, exp_struct)
def test_StockholmParser(self):
"""StockholmParser: integrity of output """
expected_sequences =\
[''.join(['AACACAUCAGAUUUCCUGGUGUAACGAAUUUUUUAAGUGCUUCUUGCUUA',\
'AGCAAGUUUCAUCCCGACCCCCUCAGGGUCGGGAUUU']),\
''.join(['AACGCAUCGGAUUUCCCGGUGUAACGAA-UUUUCAAGUGCUUCUUGCAUU',\
'AGCAAGUUUGAUCCCGACUCCUG-CGAGUCGGGAUUU']),\
''.join(['CUCACAUCAGAUUUCCUGGUGUAACGAA-UUUUCAAGUGCUUCUUGCAUA',\
'AGCAAGUUUGAUCCCGACCCGU--AGGGCCGGGAUUU'])]
expected_structure = ''.join(\
['...<<<<<<<.....>>>>>>>....................<<<<<...',\
'.>>>>>....<<<<<<<<<<.....>>>>>>>>>>..'])
for r in StockholmParser(self._fake_record):
headers = r.Info
sequences = r
structure = r.Info['Struct']
self.assertEqual(headers['GF']['AccessionNumber'], 'RF00014')
self.assertEqual(headers['GF']['Author'], 'Mifsud W')
self.assertEqualItems(sequences.values(), expected_sequences)
assert isinstance(sequences, Alignment)
self.assertEqual(structure, expected_structure)
assert isinstance(structure, WussStructure)
from cogent import LoadSeqs, RNA
from cogent.parse.stockholm import StockholmParser
from sys import argv
from numpy import zeros
#stats.py /path/to/file.sto /path/to/folder/out/
if __name__ == "__main__":
if argv[2][-1] != "/":
argv[2] += "/"
fin = open(argv[1])
aln = LoadSeqs(data=StockholmParser(fin).next(), moltype=RNA)
fin.close()
consensus = aln.majorityConsensus()
#counts A:0 U:1 G:2 C:3 -:4
counts = zeros(5, dtype=int)
countsout = open(argv[2] + "counts.txt", 'w')
countsout.write('\t'.join(['pos', 'maj', 'A', 'T', 'G', 'C', '-', '\n']))
#count all nucs that do not conform to consensus for each position
for pos, nucs in enumerate(aln.iterPositions()):
majnuc = consensus[pos]
for nuc in nucs:
if nuc != majnuc:
if nuc == 'A':
counts[0] += 1
elif nuc == 'U':
counts[1] += 1
elif nuc == 'G':
counts[2] += 1
elif nuc == 'C':
counts[3] += 1