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
|
#include <string>
#include <vector>
#include <pbdata/Types.h>
#include <alignment/algorithms/sorting/Karkkainen.hpp>
#include <alignment/algorithms/sorting/qsufsort.hpp>
#include <alignment/suffixarray/SuffixArray.hpp>
#include <alignment/suffixarray/ssort.hpp>
#include <pbdata/CompressedSequence.hpp>
#include <pbdata/FASTAReader.hpp>
#include <pbdata/FASTASequence.hpp>
#include <pbdata/NucConversion.hpp>
void PrintUsage()
{
std::cout << "usage: sawriter saOut fastaIn [fastaIn2 fastaIn3 ...] [-blt p] [-larsson] "
"[-4bit] [-manmy] [-kar]"
<< std::endl;
std::cout << " or sawriter fastaIn (writes to fastIn.sa)." << std::endl;
std::cout << " -blt p Build a lookup table on prefixes of length 'p'. This speeds "
<< std::endl
<< " up lookups considerably (more than the LCP table), but misses "
"matches "
<< std::endl
<< " less than p when searching." << std::endl;
std::cout << " -4bit Read in (one) fasta file as a compressed sequence file."
<< std::endl;
std::cout
<< " -larsson (default) Uses the method of Larsson and Sadakane to build the array."
<< std::endl;
std::cout
<< " -mamy Uses the method of MAnber and MYers to build the array (slower than "
"larsson, "
<< std::endl
<< " and produces the same result. This is mainly for double checking"
<< std::endl
<< " the correctness of larsson)." << std::endl
<< " -kark Use Karkkainen DS3 method for building the suffix array. This will "
"probably be more "
<< std::endl
<< " slow than larsson, but takes only an extra N/(sqrt 3) extra space."
<< std::endl
<< " -mafe (disabled for now!) Use the lightweight construction algorithm from "
"Manzini and Ferragina"
<< std::endl
<< " -welter Use lightweight (sort of light) suffix array construction. This is "
"a bit more slow than"
<< std::endl
<< " normal larsson." << std::endl
<< " -welterweight N use a difference cover of size N for building the suffix array. "
" Valid values are 7,32,64,111, and 2281."
<< std::endl;
}
int main(int argc, char* argv[])
{
if (argc < 2) {
PrintUsage();
std::exit(EXIT_FAILURE);
} else if (strcmp(argv[1], "-h") == 0 or strcmp(argv[1], "-help") == 0 or
strcmp(argv[1], "--help") == 0) {
PrintUsage();
std::exit(EXIT_SUCCESS);
}
int argi = 1;
std::string saFile = argv[argi++];
std::vector<std::string> inFiles;
int doBLT = 1;
int bltPrefixLength = 8;
int parsingOptions = 0;
SAType saBuildType = larsson;
int read4BitCompressed = 0;
int diffCoverSize = 0;
while (argi < argc) {
if (strlen(argv[argi]) > 0 and argv[argi][0] == '-') {
parsingOptions = 1;
}
if (!parsingOptions) {
inFiles.push_back(argv[argi]);
} else {
if (strcmp(argv[argi], "-blt") == 0) {
doBLT = 1;
if (argi < argc - 1) {
bltPrefixLength = atoi(argv[++argi]);
if (bltPrefixLength == 0) {
std::cout << argv[argi] << " is not a valid lookup table length."
<< std::endl;
std::exit(EXIT_FAILURE);
}
} else {
std::cout << "Please specify a lookup table length." << std::endl;
std::exit(EXIT_FAILURE);
}
} else if (strcmp(argv[argi], "-mamy") == 0) {
saBuildType = manmy;
} else if (strcmp(argv[argi], "-larsson") == 0) {
saBuildType = larsson;
} else if (strcmp(argv[argi], "-mcilroy") == 0) {
saBuildType = mcilroy;
} else if (strcmp(argv[argi], "-slow") == 0) {
saBuildType = slow;
} else if (strcmp(argv[argi], "-kark") == 0) {
saBuildType = kark;
} else if (strcmp(argv[argi], "-mafe") == 0) {
saBuildType = mafe;
} else if (strcmp(argv[argi], "-welter") == 0) {
saBuildType = welter;
} else if (strcmp(argv[argi], "-welterweight") == 0) {
if (argi < argc - 1) {
diffCoverSize = atoi(argv[++argi]);
} else {
std::cout << "Please specify a difference cover size. Valid values are "
"7,32,64,111, and 2281. Larger values use less memory but may be "
"slower."
<< std::endl;
std::exit(EXIT_FAILURE);
}
if (!(diffCoverSize == 7 or diffCoverSize == 32 or diffCoverSize == 64 or
diffCoverSize == 111 or diffCoverSize == 2281)) {
std::cout << "The difference cover size must be one of 7,32,64,111, or 2281."
<< std::endl;
std::cout << "Larger numbers use less space but are more slow." << std::endl;
std::exit(EXIT_FAILURE);
}
} else if (strcmp(argv[argi], "-4bit") == 0) {
read4BitCompressed = 1;
} else if (strcmp(argv[argi], "-h") == 0 or strcmp(argv[argi], "-help") == 0 or
strcmp(argv[argi], "--help") == 0) {
PrintUsage();
std::exit(EXIT_SUCCESS);
} else {
PrintUsage();
std::cout << "ERROR, bad option: " << argv[argi] << std::endl;
std::exit(EXIT_FAILURE);
}
}
++argi;
}
if (inFiles.size() == 0) {
//
// Special use case: the input file is a fasta file. Write to that file + .sa
//
inFiles.push_back(saFile);
saFile = saFile + ".sa";
}
VectorIndex inFileIndex;
FASTASequence seq;
CompressedSequence<FASTASequence> compSeq;
if (read4BitCompressed == 0) {
for (inFileIndex = 0; inFileIndex < inFiles.size(); ++inFileIndex) {
FASTAReader reader;
reader.Init(inFiles[inFileIndex]);
reader.SetSpacePadding(111);
if (saBuildType == kark) {
//
// The Karkkainen sa building method requires a little extra
// space at the end of the dna sequence so that counting may
// be done mod 3 without adding extra logic for boundaries.
//
}
if (inFileIndex == 0) {
reader.ReadAllSequencesIntoOne(seq);
reader.Close();
} else {
while (reader.ConcatenateNext(seq)) {
std::cout << "added " << seq.title << std::endl;
}
}
}
seq.ToThreeBit();
//seq.ToUpper();
} else {
assert(inFiles.size() == 1);
std::cout << "reading compressed sequence." << std::endl;
compSeq.Read(inFiles[0]);
seq.seq = compSeq.seq;
seq.length = compSeq.length;
compSeq.RemoveCompressionCounts();
std::cout << "done." << std::endl;
}
//
// For now, do not allow creation of suffix arrays on sequences > 4G.
//
if (seq.length >= UINT_MAX) {
std::cout << "ERROR, references greater than " << UINT_MAX << " bases are not supported."
<< std::endl;
std::cout << "Consider breaking the reference into multiple files, running alignment. "
<< std::endl;
std::cout << "against each file, and merging the result." << std::endl;
std::exit(EXIT_FAILURE);
}
std::vector<int> alphabet;
SuffixArray<Nucleotide, std::vector<int> > sa;
// sa.InitTwoBitDNAAlphabet(alphabet);
// sa.InitAsciiCharDNAAlphabet(alphabet);
sa.InitThreeBitDNAAlphabet(alphabet);
if (saBuildType == manmy) {
sa.MMBuildSuffixArray(seq.seq, seq.length, alphabet);
} else if (saBuildType == mcilroy) {
sa.index = new SAIndex[seq.length + 1];
DNALength i;
for (i = 0; i < seq.length; i++) {
sa.index[i] = seq.seq[i] + 1;
}
sa.index[seq.length] = 0;
ssort(sa.index, NULL);
for (i = 1; i < seq.length + 1; i++) {
sa.index[i - 1] = sa.index[i];
};
sa.length = seq.length;
} else if (saBuildType == larsson) {
sa.LarssonBuildSuffixArray(seq.seq, seq.length, alphabet);
} else if (saBuildType == kark) {
sa.index = new SAIndex[seq.length];
seq.ToThreeBit();
DNALength p;
for (p = 0; p < seq.length; p++) {
seq.seq[p]++;
}
KarkkainenBuildSuffixArray<Nucleotide>(seq.seq, sa.index, seq.length, 5);
sa.length = seq.length;
} else if (saBuildType == mafe) {
// sa.MaFeBuildSuffixArray(seq.seq, seq.length);
} else if (saBuildType == welter) {
if (diffCoverSize == 0) {
sa.LightweightBuildSuffixArray(seq.seq, seq.length);
} else {
sa.LightweightBuildSuffixArray(seq.seq, seq.length, diffCoverSize);
}
}
if (doBLT) {
sa.BuildLookupTable(seq.seq, seq.length, bltPrefixLength);
}
sa.Write(saFile);
return 0;
}
|
