1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
|
#!/usr/bin/env python
import unittest
from cogent import LoadTree
from cogent.evolve import substitution_model
def a_c(x, y):
return (x == 'A' and y == 'C') or (x == 'C' and y == 'A')
from cogent.evolve.predicate import MotifChange, replacement
__author__ = "Peter Maxwell and Gavin Huttley"
__copyright__ = "Copyright 2007-2009, The Cogent Project"
__credits__ = ["Peter Maxwell", "Gavin Huttley"]
__license__ = "GPL"
__version__ = "1.4.1"
__maintainer__ = "Gavin Huttley"
__email__ = "gavin.huttley@anu.edu.au"
__status__ = "Production"
a_c = MotifChange('A', 'C')
trans = MotifChange('A', 'G') | MotifChange('T', 'C')
TREE = LoadTree(tip_names='ab')
class ScaleRuleTests(unittest.TestCase):
def _makeModel(self, do_scaling, predicates, scale_rules=[]):
return substitution_model.Nucleotide(
do_scaling=do_scaling, equal_motif_probs=True,
model_gaps=False, predicates=predicates, scales=scale_rules)
def _getScaledLengths(self, model, params):
LF = model.makeLikelihoodFunction(TREE)
for param in params:
LF.setParamRule(param, value=params[param], is_const=True)
result = {}
for predicate in model.scale_masks:
result[predicate] = LF.getScaledLengths(predicate)['a']
return result
def test_scaled(self):
"""Scale rule requiring matrix entries to have all pars specified"""
model = self._makeModel(True, {'k':trans}, {
'ts':trans, 'tv': ~trans})
self.assertEqual(
self._getScaledLengths(model, {'k':6.0, 'length':4.0}),
{'ts': 3.0, 'tv':1.0})
def test_binned(self):
model = self._makeModel(True, {'k':trans}, {
'ts':trans, 'tv': ~trans})
LF = model.makeLikelihoodFunction(TREE, bins=2)
LF.setParamRule('length', value=4.0, is_const=True)
LF.setParamRule('k', value=6.0, bin='bin0', is_const=True)
LF.setParamRule('k', value=1.0, bin='bin1', is_const=True)
for (bin, expected) in [('bin0', 3.0), ('bin1', 4.0/3), (None, 13.0/6)]:
self.assertEqual(LF.getScaledLengths('ts', bin=bin)['a'], expected)
def test_unscaled(self):
"""Scale rule on a model which has scaling performed after calculation
rather than during it"""
model = self._makeModel(False, {'k':trans}, {
'ts':trans, 'tv': ~trans})
self.assertEqual(
self._getScaledLengths(model, {'k':6.0, 'length':2.0}),
{'ts': 3.0, 'tv':1.0})
def test_scaled_or(self):
"""Scale rule where matrix entries can have any of the pars specified"""
model = self._makeModel(True, {'k':trans, 'ac':a_c}, {
'or': (trans | a_c), 'not': ~(trans | a_c)})
self.assertEqual(
self._getScaledLengths(model, {'k':6.0,'length':6.0, 'ac': 3.0}),
{'or': 5.0, 'not': 1.0})
def test_scaling(self):
"""Testing scaling calculations using Dn and Ds as an example."""
model = substitution_model.Codon(
do_scaling=True, model_gaps=False, recode_gaps=True,
predicates = {
'k': trans,
'r': replacement},
motif_probs={
'TAT': 0.0088813702685557206, 'TGT': 0.020511736096426307,
'TCT': 0.024529498836963416, 'TTT': 0.019454430112074435,
'TGC': 0.0010573059843518714, 'TGG': 0.0042292239374074857,
'TAC': 0.002326073165574117, 'TTC': 0.0086699090716853451,
'TCG': 0.0010573059843518714, 'TTA': 0.020723197293296681,
'TTG': 0.01036159864664834, 'TCC': 0.0082469866779445976,
'TCA': 0.022414886868259674, 'GCA': 0.015648128568407697,
'GTA': 0.014590822584055826, 'GCC': 0.0095157538591668436,
'GTC': 0.0063438359061112285, 'GCG': 0.0016916895749629942,
'GTG': 0.0067667582998519769, 'CAA': 0.018185662930852189,
'GTT': 0.021569042080778176, 'GCT': 0.014167900190315077,
'ACC': 0.0042292239374074857, 'GGT': 0.014167900190315077,
'CGA': 0.0012687671812222456, 'CGC': 0.0010573059843518714,
'GAT': 0.030238951152463524, 'AAG': 0.034891097483611758,
'CGG': 0.002326073165574117, 'ACT': 0.028758722774370905,
'GGG': 0.0071896806935927262, 'GGA': 0.016282512159018821,
'GGC': 0.0090928314654260944, 'GAG': 0.031296257136815393,
'AAA': 0.05476844998942694, 'GAC': 0.011207443434129837,
'CGT': 0.0033833791499259885, 'GAA': 0.076337492070205112,
'CTT': 0.010573059843518714, 'ATG': 0.012687671812222457,
'ACA': 0.021991964474518927, 'ACG': 0.00084584478748149711,
'ATC': 0.0076126030873334746, 'AAC': 0.022837809262000422,
'ATA': 0.017762740537111441, 'AGG': 0.013533516599703954,
'CCT': 0.025586804821315288, 'AGC': 0.029393106364982026,
'AGA': 0.021991964474518927, 'CAT': 0.021357580883907802,
'AAT': 0.05772890674561218, 'ATT': 0.019031507718333687,
'CTG': 0.012899133009092831, 'CTA': 0.013744977796574329,
'CTC': 0.0078240642842038483, 'CAC': 0.0050750687248889825,
'CCG': 0.00021146119687037428, 'AGT': 0.03742863184605625,
'CAG': 0.024106576443222668, 'CCA': 0.021357580883907802,
'CCC': 0.0069782194967223515},
scales = {'dN': replacement, 'dS': ~replacement},
mprob_model = 'tuple',
)
length = 0.1115
a = self._getScaledLengths(model,
{'k': 3.6491, 'r': 0.6317, 'length': length})
b = self._getScaledLengths(model,
{'k': 3.6491, 'r': 1.0, 'length': length})
dN = length * a['dN'] / (3.0 * b['dN'])
dS = length * a['dS'] / (3.0 * b['dS'])
# following are results from PAML
self.assertEqual('%.4f' % dN, '0.0325')
self.assertEqual('%.4f' % dS ,'0.0514')
if __name__ == '__main__':
unittest.main()
|
