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
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
mod bwt_nn;
use super::{FmIndexConfig, FmIndex};
use super::utils::{
SuffixArray,
accumulate_count_array, compress_suffix_array,
};
use bwt_nn::BwtNn;
use libdivsufsort_rs::{divsufsort64, bw_transform64};
use serde::{Serialize, Deserialize};
type CountArray = [u64; 6];
type KmerLookupTable = (usize, Vec<u64>);
const A_UTF8: u8 = 65;
const C_UTF8: u8 = 67;
const G_UTF8: u8 = 71;
const T_UTF8: u8 = 84;
const N_UTF8: u8 = 95;
const A_U8_IDX: u8 = 0b000;
const C_U8_IDX: u8 = 0b001;
const G_U8_IDX: u8 = 0b010;
const T_U8_IDX: u8 = 0b011;
const N_U8_IDX: u8 = 0b100;
const A_IDX: usize = 0b000;
const C_IDX: usize = 0b001;
const G_IDX: usize = 0b010;
const T_IDX: usize = 0b011;
const N_IDX: usize = 0b100;
#[derive(Debug, Serialize, Deserialize, PartialEq)]
pub struct FmIndexNn {
count_array: CountArray,
sampling_ratio: u64,
text_len: u64,
suffix_array: SuffixArray,
pub kmer_lookup_table: Option<KmerLookupTable>,
bwt: BwtNn,
}
impl FmIndexNn {
#[inline]
pub fn new(config: &FmIndexConfig, mut text: Vec<u8>) -> Self {
let text_len = text.len() as u64;
let (count_array, kmer_lookup_table): (CountArray, Option<KmerLookupTable>) = Self::get_ca_and_klt(config, &mut text);
let suffix_array = divsufsort64(&text).unwrap();
let pidx = {
let mut sa = suffix_array.clone();
let pidx = bw_transform64(&mut text, &mut sa).unwrap();
pidx
};
let suffix_array = compress_suffix_array(suffix_array, config.sa_sampling_ratio);
let bwt = BwtNn::new(text, pidx);
Self {
count_array: count_array,
sampling_ratio: config.sa_sampling_ratio,
suffix_array: suffix_array,
text_len: text_len,
kmer_lookup_table: kmer_lookup_table,
bwt: bwt,
}
}
#[inline]
pub fn get_ca_and_klt(config: &FmIndexConfig, text: &mut Vec<u8>) -> (CountArray, Option<KmerLookupTable>) {
match config.kmer_size {
Some(kmer) => {
let mut count_array: CountArray = [0; 6];
let klt_length: usize = 5usize.pow(kmer as u32);
let mut kmer_lookup_table: Vec<u64> = vec![0; klt_length];
let mut klt_index: usize = 0;
let index_for_each_char: [usize; 4] = {
let i = klt_length/5;
[1*i, 2*i, 3*i, 4*i]
};
text.iter_mut().rev().for_each(|chr| {
match *chr {
A_UTF8 => {
count_array[1] += 1;
klt_index /= 5;
kmer_lookup_table[klt_index] += 1;
},
C_UTF8 => {
count_array[2] += 1;
klt_index /= 5;
klt_index += index_for_each_char[0];
kmer_lookup_table[klt_index] += 1;
},
G_UTF8 => {
count_array[3] += 1;
klt_index /= 5;
klt_index += index_for_each_char[1];
kmer_lookup_table[klt_index] += 1;
},
T_UTF8 => {
count_array[4] += 1;
klt_index /= 5;
klt_index += index_for_each_char[2];
kmer_lookup_table[klt_index] += 1;
},
_ => {
count_array[5] += 1;
klt_index /= 5;
klt_index += index_for_each_char[3];
kmer_lookup_table[klt_index] += 1;
*chr = N_UTF8;
},
}
});
accumulate_count_array(&mut count_array);
accumulate_count_array(&mut kmer_lookup_table);
(count_array, Some((kmer, kmer_lookup_table)))
},
None => {
let mut count_array: CountArray = [0; 6];
for chr in text {
match *chr {
A_UTF8 => count_array[1] += 1,
C_UTF8 => count_array[2] += 1,
G_UTF8 => count_array[3] += 1,
T_UTF8 => count_array[4] += 1,
_ => {
count_array[5] += 1;
*chr = N_UTF8;
},
}
}
accumulate_count_array(&mut count_array);
(count_array, None)
}
}
}
#[inline]
fn lf_map(&self, pattern: &[u8]) -> (u64, u64) {
let mut idx = pattern.len();
let chr = pattern[idx-1];
let mut pos_range = self.pos_range_init(chr);
idx -= 1;
while pos_range.0 < pos_range.1 && idx > 0 {
let c = pattern[idx-1];
pos_range = self.bwt.next_pos_range_from_range(pos_range, c, &self.count_array);
idx -= 1;
}
pos_range
}
#[inline]
fn pos_range_init(&self, c: u8) -> (u64, u64) {
let idx = nc_to_idx(&c);
(self.count_array[idx], self.count_array[idx+1])
}
}
impl FmIndex for FmIndexNn {
#[inline]
fn count(&self, pattern: &[u8]) -> u64 {
let pos_range = self.lf_map(pattern);
pos_range.1 - pos_range.0
}
#[inline]
fn locate_wo_klt(&self, pattern: &[u8]) -> Vec<u64> {
let pos_range = self.lf_map(pattern);
let mut locations: Vec<u64> = Vec::with_capacity((pos_range.1 - pos_range.0) as usize);
'each_pos: for mut position in pos_range.0..pos_range.1 {
let mut offset: u64 = 0;
while position % self.sampling_ratio != 0 {
match self.bwt.get_pre_pos(position, &self.count_array) {
Some(v) => {
position = v;
},
None => {
locations.push(offset);
continue 'each_pos;
},
}
offset += 1;
}
let location = self.suffix_array[(position / self.sampling_ratio) as usize] + offset;
locations.push(location);
}
locations
}
#[inline]
fn locate_w_klt(&self, pattern: &[u8]) -> Vec<u64> {
let (kmer_size, klt) = self.kmer_lookup_table.as_ref().unwrap();
let mut idx = pattern.len();
let pattern_len = idx.clone() as u64;
let mut pos_range: (u64, u64) = {
if *kmer_size < idx {
let klt_index = klt_index_of_pattern(&pattern[pattern.len()-*kmer_size..]);
idx -= *kmer_size;
if klt_index == 0 {
(0, klt[klt_index])
} else {
(klt[klt_index-1], klt[klt_index])
}
} else {
let klt_index = klt_index_of_pattern(pattern);
let offset_adder = 5_usize.pow((*kmer_size - idx) as u32);
let klt_index_start = klt_index * offset_adder;
let klt_index_end = klt_index * offset_adder + offset_adder - 1;
idx = 0;
if klt_index_start == 0 {
(0, klt[klt_index_end])
} else {
(klt[klt_index_start-1], klt[klt_index_end])
}
}
};
while pos_range.0 < pos_range.1 && idx > 0 {
let c = pattern[idx-1];
pos_range = self.bwt.next_pos_range_from_range(pos_range, c, &self.count_array);
idx -= 1;
}
let mut locations: Vec<u64> = Vec::with_capacity((pos_range.1 - pos_range.0) as usize);
'each_pos: for mut position in pos_range.0..pos_range.1 {
let mut offset: u64 = 0;
while position % self.sampling_ratio != 0 {
match self.bwt.get_pre_pos(position, &self.count_array) {
Some(v) => {
position = v;
offset += 1;
},
None => {
locations.push(offset);
continue 'each_pos;
},
}
}
let location = self.suffix_array[(position / self.sampling_ratio) as usize] + offset;
if location + pattern_len <= self.text_len {
locations.push(location);
}
}
locations
}
}
#[inline]
fn nc_to_idx(c: &u8) -> usize {
match *c {
A_UTF8 => A_IDX,
C_UTF8 => C_IDX,
G_UTF8 => G_IDX,
T_UTF8 => T_IDX,
_ => N_IDX,
}
}
#[inline]
fn klt_index_of_pattern(pattern: &[u8]) -> usize {
let mut klt_index: usize = 0;
pattern.iter().for_each(|chr| {
match *chr {
A_UTF8 => {
klt_index *= 5;
},
C_UTF8 => {
klt_index *= 5;
klt_index += 1;
},
G_UTF8 => {
klt_index *= 5;
klt_index += 2;
},
T_UTF8 => {
klt_index *= 5;
klt_index += 3;
},
_ => {
klt_index *= 5;
klt_index += 4;
},
}
});
klt_index
}