-
Notifications
You must be signed in to change notification settings - Fork 4
/
barcode.h
199 lines (185 loc) · 9.32 KB
/
barcode.h
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
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
/*
Pheniqs : PHilology ENcoder wIth Quality Statistics
Copyright (C) 2018 Lior Galanti
NYU Center for Genetics and System Biology
Author: Lior Galanti <lior.galanti@nyu.edu>
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as
published by the Free Software Foundation, either version 3 of the
License, or (at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#ifndef PHENIQS_BARCODE_H
#define PHENIQS_BARCODE_H
#include "include.h"
#include "sequence.h"
#include "selector.h"
class Barcode : public SequenceArray< Sequence >, public AccumulatingOption {
friend ostream& operator<<(ostream& o, const Barcode& barcode);
friend bool encode_key_value(const string& key, const Barcode& value, Value& node, Document& document);
public:
void operator=(Barcode const &) = delete;
const int32_t index;
const double concentration;
Barcode(const Value& ontology);
Barcode(const Barcode& other);
inline const bool is_classified() const {
/* by convention, enforced by the job configuration loader, barcode 0 is always the unclassified */
return index > 0;
};
inline const bool is_unclassified() const {
return index == 0;
};
operator string() const {
/* NOTICE this is in BAM encoding not iupac and will not look as expected when printed
Used by MDD for exact match */
string key;
for(const auto& segment : segment_array) {
for(int32_t i(0); i < segment.length; ++i) {
key.push_back(segment.code[i]);
}
}
return key;
};
inline void accurate_decoding_probability(const Observation& observation, double& probability) const {
double sigma_q(0);
for(size_t i(0); i < segment_array.size(); ++i) {
const Sequence& expected = segment_array[i];
const ObservedSequence& observed = observation[i];
for(int32_t j(0); j < expected.length; ++j) {
sigma_q += scale.substitution_quality(expected.code[j], observed.code[j], observed.quality[j]);
}
}
probability = pow(PHRED_PROBABILITY_BASE, sigma_q);
};
inline void accurate_decoding_probability(const Observation& observation, double& probability, int32_t& distance) const {
distance = 0;
double sigma_q(0);
for(size_t i(0); i < segment_array.size(); ++i) {
const Sequence& expected = segment_array[i];
const ObservedSequence& observed = observation[i];
for(int32_t j(0); j < expected.length; ++j) {
sigma_q += scale.substitution_quality(expected.code[j], observed.code[j], observed.quality[j]);
if(observed.code[j] != expected.code[j]) {
++distance;
}
}
}
probability = pow(PHRED_PROBABILITY_BASE, sigma_q);
};
inline void compensated_decoding_probability(const Observation& observation, double& probability) const {
/* use the Kahan summation algorithm to minimize floating point drift
see https://en.wikipedia.org/wiki/Kahan_summation_algorithm
sigma_q accumulates double precision floats from inversed quality scores and UNIFORM_BASE_QUALITY
*/
double y(0);
double t(0);
double sigma_q(0);
double compensation(0);
for(size_t i(0); i < segment_array.size(); ++i) {
const Sequence& expected = segment_array[i];
const ObservedSequence& observed = observation[i];
for(int32_t j(0); j < expected.length; ++j) {
y = scale.substitution_quality(expected.code[j], observed.code[j], observed.quality[j]) - compensation;
t = sigma_q + y;
compensation = (t - sigma_q) - y;
sigma_q = t;
}
}
probability = pow(PHRED_PROBABILITY_BASE, sigma_q);
};
inline void compensated_decoding_probability(const Observation& observation, double& probability, int32_t& distance) const {
/* use the Kahan summation algorithm to minimize floating point drift
see https://en.wikipedia.org/wiki/Kahan_summation_algorithm
sigma_q accumulates double precision floats from inversed quality scores and UNIFORM_BASE_QUALITY
*/
double y(0);
double t(0);
distance = 0;
double sigma_q(0);
double compensation(0);
for(size_t i(0); i < segment_array.size(); ++i) {
const Sequence& expected = segment_array[i];
const ObservedSequence& observed = observation[i];
for(int32_t j(0); j < expected.length; ++j) {
y = scale.substitution_quality(expected.code[j], observed.code[j], observed.quality[j]) - compensation;
t = sigma_q + y;
compensation = (t - sigma_q) - y;
sigma_q = t;
if(observed.code[j] != expected.code[j]) {
++distance;
}
}
}
probability = pow(PHRED_PROBABILITY_BASE, sigma_q);
};
inline void compensated_decoding_probability(const Observation& observation, const uint8_t& high_quality_threshold,
double& probability, int32_t& distance, int32_t& high_quality_distance) const {
/* This version of the function also counts in high_quality_distance mismatches
with quality strictly greater high_quality_threshold.
use the Kahan summation algorithm to minimize floating point drift
see https://en.wikipedia.org/wiki/Kahan_summation_algorithm
sigma_q accumulates double precision floats from inversed quality scores and UNIFORM_BASE_QUALITY
*/
double y(0);
double t(0);
distance = 0;
high_quality_distance = 0;
double sigma_q(0);
double compensation(0);
for(size_t i(0); i < segment_array.size(); ++i) {
const Sequence& expected = segment_array[i];
const ObservedSequence& observed = observation[i];
for(int32_t j(0); j < expected.length; ++j) {
y = scale.substitution_quality(expected.code[j], observed.code[j], observed.quality[j]) - compensation;
t = sigma_q + y;
compensation = (t - sigma_q) - y;
sigma_q = t;
if(observed.code[j] != expected.code[j]) {
++distance;
if(observed.quality[j] >= high_quality_threshold) {
++high_quality_distance;
}
}
}
}
probability = pow(PHRED_PROBABILITY_BASE, sigma_q);
};
inline void compensated_decoding_quality(const Observation& observation, double& quality, int32_t& distance) const {
/* use the Kahan summation algorithm to minimize floating point drift
see https://en.wikipedia.org/wiki/Kahan_summation_algorithm
sigma_q accumulates double precision floats who are either the probability of correct base call
if the expected and observed bases agree or probability of error if they dont.
Since the probabilties are encoded on a logarithmic scale they are added instead of multiplied.
UNIFORM_BASE_QUALITY is used in the of an observed ANY NUCLEOTIDE.
*/
double y(0);
double t(0);
distance = 0;
double sigma_q(0);
double compensation(0);
for(size_t i(0); i < segment_array.size(); ++i) {
const Sequence& expected = segment_array[i];
const ObservedSequence& observed = observation[i];
for(int32_t j(0); j < expected.length; ++j) {
y = scale.substitution_quality(expected.code[j], observed.code[j], observed.quality[j]) - compensation;
t = sigma_q + y;
compensation = (t - sigma_q) - y;
sigma_q = t;
if(observed.code[j] != expected.code[j]) {
++distance;
}
}
}
quality = sigma_q;
};
void encode(Value& container, Document& document) const override;
};
ostream& operator<<(ostream& o, const Barcode& barcode);
template<> vector< Barcode > decode_value_by_key(const Value::Ch* key, const Value& container);
#endif /* PHENIQS_BARCODE_H */