def test_get_counts(self):
"""get_counts: should work with all parameters"""
seq = RnaSequence('UCAG-NAUGU')
seq2 = RnaSequence('UAAG-CACGC')
p = Pairs([(1, 8), (2, 7)])
p2 = Pairs([
(1, 8),
(2, 6),
(3, 6),
(4, 9),
])
exp = {'TP':1,'TN':0, 'FN':1,'FP':3,\
'FP_INCONS':0, 'FP_CONTRA':0, 'FP_COMP':0}
self.assertEqual(get_counts(p, p2), exp)
exp = {'TP':1,'TN':0, 'FN':1,'FP':3,\
'FP_INCONS':1, 'FP_CONTRA':1, 'FP_COMP':1}
self.assertEqual(get_counts(p, p2, split_fp=True), exp)
seq = RnaSequence('UCAG-NACGU')
exp = {'TP':1,'TN':7, 'FN':1,'FP':3,\
'FP_INCONS':1, 'FP_CONTRA':1, 'FP_COMP':1}
self.assertEqual(get_counts(p, p2, split_fp=True,\
sequences=[seq], min_dist=2), exp)
# check against compare_ct.pm
exp = {'TP':4,'TN':266, 'FN':6,'FP':6,\
'FP_INCONS':2, 'FP_CONTRA':2, 'FP_COMP':2}
seq = 'agguugaaggggauccgauccacuccccggcuggucaaccu'.upper()
self.assertEqual(get_counts(self.true, self.predicted, split_fp=True,\
sequences=[seq], min_dist=4), exp)
def get_counts(ref, predicted, split_fp=True, sequences=None, min_dist=1,
accepted=ACCEPTED_STD):
"""Return TP, TN, FPcont, FPconf FPcomp, FN counts
ref: BasePairs object
pred: BasePairs object
split_fp: bool, set flags whether to split false positivies into
different categories or not
sequences: list of sequences of RNA (strings)
min_dist: minimum distance between two bases which make up a pair
accepted: dict, pairs of letters denoting allowed base pairs
"""
assert isinstance(ref, BasePairs), 'ref is not BasePairs instance'
assert isinstance(predicted, BasePairs), 'predicted is not BasePairs instance'
assert isinstance(split_fp, bool), 'split_fp is not bool'
assert type(sequences) in [list, tuple, types.NoneType],\
'sequences is not tuple or list'
assert isinstance(min_dist, int), 'min_dist is not integer'
assert isinstance(accepted, dict), 'accepted is not dict'
from cogent.struct.pairs_util import get_counts
counts = get_counts(ref=ref, predicted=predicted, split_fp=split_fp,
sequences=None, min_dist=min_dist)
if sequences:
num_possible_pairs = get_all_pairs(sequences, min_dist,
accepted=accepted)
counts['TN'] = num_possible_pairs - counts['TP'] -\
counts['FP_INCONS'] - counts['FP_CONTRA']
return counts
def mcc(ref, predicted, seqs, min_dist=1, accepted=ACCEPTED_STD):
"""Return the Matthews correlation coefficient
ref: Pairs object -> reference structure (true structure)
predicted: Pairs object -> predicted structure
seqs: list of sequences, necessary to compute the number of true
negatives. See documentation of extract_seqs function for
accepted formats.
min_dist: minimum distance required between two members of a base pair.
Needed to calculate the number of true negatives.
accepted: dict, pairs of letters denoting allowed base pairs
"""
check_structures(ref, predicted)
if not ref and not predicted:
return 1.0
elif not predicted:
return 0.0
elif not seqs:
raise ValueError, 'No sequence provided!'
sequences = extract_seqs(seqs)
counts = get_counts(ref, predicted, sequences=sequences, split_fp=True,\
min_dist=min_dist, accepted=accepted)
result = mcc_formula(counts)
if result < -1 or result > 1:
raise ValueError("mcc not in range <-1, 1>: %.2f"%(result))
return result
def test_get_counts_pseudo(self):
"""get_counts: should work when pseudo in ref -> classification off"""
# pairs that would normally be compatible, are now contradicting
ref = Pairs([(0, 8), (1, 7), (4, 10)])
pred = Pairs([(0, 8), (3, 6), (4, 10)])
seq = 'GACUGUGUCAU'
exp = {'TP':2,'TN':13-2-1, 'FN':1,'FP':1,\
'FP_INCONS':0, 'FP_CONTRA':1, 'FP_COMP':0}
self.assertEqual(get_counts(ref, pred, split_fp=True,\
sequences=[seq], min_dist=4), exp)
def test_get_counts_pseudo(self):
"""get_counts: should work when pseudo in ref -> classification off"""
# pairs that would normally be compatible, are now contradicting
ref = Pairs([(0,8),(1,7),(4,10)])
pred = Pairs([(0,8),(3,6),(4,10)])
seq = 'GACUGUGUCAU'
exp = {'TP':2,'TN':13-2-1, 'FN':1,'FP':1,\
'FP_INCONS':0, 'FP_CONTRA':1, 'FP_COMP':0}
self.assertEqual(get_counts(ref, pred, split_fp=True,\
sequences=[seq], min_dist=4), exp)
def test_get_counts(self):
"""get_counts: should work with all parameters"""
seq = RnaSequence('UCAG-NAUGU')
seq2 = RnaSequence('UAAG-CACGC')
p = Pairs([(1,8),(2,7)])
p2 = Pairs([(1,8),(2,6),(3,6),(4,9),])
exp = {'TP':1,'TN':0, 'FN':1,'FP':3,\
'FP_INCONS':0, 'FP_CONTRA':0, 'FP_COMP':0}
self.assertEqual(get_counts(p, p2), exp)
exp = {'TP':1,'TN':0, 'FN':1,'FP':3,\
'FP_INCONS':1, 'FP_CONTRA':1, 'FP_COMP':1}
self.assertEqual(get_counts(p, p2, split_fp=True), exp)
seq = RnaSequence('UCAG-NACGU')
exp = {'TP':1,'TN':7, 'FN':1,'FP':3,\
'FP_INCONS':1, 'FP_CONTRA':1, 'FP_COMP':1}
self.assertEqual(get_counts(p, p2, split_fp=True,\
sequences=[seq], min_dist=2), exp)
# check against compare_ct.pm
exp = {'TP':4,'TN':266, 'FN':6,'FP':6,\
'FP_INCONS':2, 'FP_CONTRA':2, 'FP_COMP':2}
seq = 'agguugaaggggauccgauccacuccccggcuggucaaccu'.upper()
self.assertEqual(get_counts(self.true, self.predicted, split_fp=True,\
sequences=[seq], min_dist=4), exp)
def sensitivity(ref, predicted, accepted=ACCEPTED_STD):
"""Return sensitivity of the predicted structure
ref: Pairs object -> reference structure (true structure)
predicted: Pairs object -> predicted structure
accepted: dict, pairs of letters denoting allowed base pairs
Formula: sensitivity = tp/(tp + fn)
tp = True positives
fn = False negatives
"""
check_structures(ref, predicted)
if not ref and not predicted:
return 1.0
elif not predicted:
return 0.0
counts = get_counts(ref, predicted, accepted=accepted)
return sensitivity_formula(counts)
