1 | // ================================================================= // |
---|
2 | // // |
---|
3 | // File : chartable.cxx // |
---|
4 | // Purpose : count characters of multiple sequences // |
---|
5 | // // |
---|
6 | // Institute of Microbiology (Technical University Munich) // |
---|
7 | // http://www.arb-home.de/ // |
---|
8 | // // |
---|
9 | // ================================================================= // |
---|
10 | |
---|
11 | #include "chartable.h" |
---|
12 | |
---|
13 | #include <arb_msg.h> |
---|
14 | #include <arb_mem.h> |
---|
15 | #include <iupac.h> |
---|
16 | #include <arb_defs.h> |
---|
17 | #include <smartptr.h> |
---|
18 | |
---|
19 | #include <cctype> |
---|
20 | |
---|
21 | #define CONSENSUS_AWAR_SOURCE CAS_INTERNAL // not awar-constructable here |
---|
22 | #include <consensus.h> |
---|
23 | |
---|
24 | using namespace chartable; |
---|
25 | |
---|
26 | // ------------------------ |
---|
27 | // SepBaseFreq |
---|
28 | |
---|
29 | #ifdef DEBUG |
---|
30 | // # define COUNT_BASES_TABLE_SIZE |
---|
31 | #endif |
---|
32 | |
---|
33 | #ifdef COUNT_BASES_TABLE_SIZE |
---|
34 | static long bases_table_size = 0L; |
---|
35 | #endif |
---|
36 | |
---|
37 | void SepBaseFreq::init(int length) { |
---|
38 | if (length) { |
---|
39 | if (no_of_bases.shortTable) { |
---|
40 | delete no_of_bases.shortTable; |
---|
41 | no_of_bases.shortTable = NULp; |
---|
42 | #ifdef COUNT_BASES_TABLE_SIZE |
---|
43 | bases_table_size -= (no_of_entries+1)*table_entry_size; |
---|
44 | #endif |
---|
45 | } |
---|
46 | |
---|
47 | no_of_entries = length; |
---|
48 | int new_size = (no_of_entries+1)*table_entry_size; |
---|
49 | no_of_bases.shortTable = (unsigned char*)new char[new_size]; |
---|
50 | memset(no_of_bases.shortTable, 0, new_size); |
---|
51 | #ifdef COUNT_BASES_TABLE_SIZE |
---|
52 | bases_table_size += new_size; |
---|
53 | printf("bases_table_size = %li\n", bases_table_size); |
---|
54 | #endif |
---|
55 | } |
---|
56 | } |
---|
57 | |
---|
58 | void SepBaseFreq::expand_table_entry_size() { // converts short table into long table |
---|
59 | ct_assert(table_entry_size==SHORT_TABLE_ELEM_SIZE); |
---|
60 | |
---|
61 | int *new_table = new int[no_of_entries+1]; |
---|
62 | int i; |
---|
63 | |
---|
64 | for (i=0; i<=no_of_entries; i++) { // LOOP_VECTORIZED[!<5.0] |
---|
65 | new_table[i] = no_of_bases.shortTable[i]; |
---|
66 | } |
---|
67 | delete [] no_of_bases.shortTable; |
---|
68 | |
---|
69 | #ifdef COUNT_BASES_TABLE_SIZE |
---|
70 | bases_table_size += (no_of_entries+1)*(LONG_TABLE_ELEM_SIZE-SHORT_TABLE_ELEM_SIZE); |
---|
71 | printf("bases_table_size = %li\n", bases_table_size); |
---|
72 | #endif |
---|
73 | |
---|
74 | no_of_bases.longTable = new_table; |
---|
75 | table_entry_size = LONG_TABLE_ELEM_SIZE; |
---|
76 | } |
---|
77 | |
---|
78 | #define INVALID_TABLE_OPERATION() GBK_terminatef("SepBaseFreq: invalid operation at %i", __LINE__) |
---|
79 | |
---|
80 | void SepBaseFreq::add(const SepBaseFreq& other, int start, int end) { |
---|
81 | ct_assert(no_of_entries==other.no_of_entries); |
---|
82 | ct_assert(start>=0); |
---|
83 | ct_assert(end<no_of_entries); |
---|
84 | ct_assert(start<=end); |
---|
85 | |
---|
86 | int i; |
---|
87 | if (table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
88 | if (other.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
89 | for (i=start; i<=end; i++) { |
---|
90 | set_elem_short(i, get_elem_short(i)+other.get_elem_short(i)); |
---|
91 | } |
---|
92 | } |
---|
93 | else { |
---|
94 | INVALID_TABLE_OPERATION(); // cannot add long to short |
---|
95 | } |
---|
96 | } |
---|
97 | else { |
---|
98 | if (other.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
99 | for (i=start; i<=end; i++) { // LOOP_VECTORIZED[!>=620<7] |
---|
100 | set_elem_long(i, get_elem_long(i)+other.get_elem_short(i)); |
---|
101 | } |
---|
102 | } |
---|
103 | else { |
---|
104 | for (i=start; i<=end; i++) { // LOOP_VECTORIZED[!>=620<7] |
---|
105 | set_elem_long(i, get_elem_long(i)+other.get_elem_long(i)); |
---|
106 | } |
---|
107 | } |
---|
108 | } |
---|
109 | } |
---|
110 | void SepBaseFreq::sub(const SepBaseFreq& other, int start, int end) { |
---|
111 | ct_assert(no_of_entries==other.no_of_entries); |
---|
112 | ct_assert(start>=0); |
---|
113 | ct_assert(end<no_of_entries); |
---|
114 | ct_assert(start<=end); |
---|
115 | |
---|
116 | int i; |
---|
117 | if (table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
118 | if (other.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
119 | for (i=start; i<=end; i++) { |
---|
120 | set_elem_short(i, get_elem_short(i)-other.get_elem_short(i)); |
---|
121 | } |
---|
122 | } |
---|
123 | else { |
---|
124 | INVALID_TABLE_OPERATION(); // cannot sub long from short |
---|
125 | } |
---|
126 | } |
---|
127 | else { |
---|
128 | if (other.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
129 | for (i=start; i<=end; i++) { // LOOP_VECTORIZED[!>=620<7] |
---|
130 | set_elem_long(i, get_elem_long(i)-other.get_elem_short(i)); |
---|
131 | } |
---|
132 | } |
---|
133 | else { |
---|
134 | for (i=start; i<=end; i++) { // LOOP_VECTORIZED[!>=620<7] |
---|
135 | set_elem_long(i, get_elem_long(i)-other.get_elem_long(i)); |
---|
136 | } |
---|
137 | } |
---|
138 | } |
---|
139 | } |
---|
140 | void SepBaseFreq::sub_and_add(const SepBaseFreq& Sub, const SepBaseFreq& Add, PosRange range) { |
---|
141 | ct_assert(no_of_entries==Sub.no_of_entries); |
---|
142 | ct_assert(no_of_entries==Add.no_of_entries); |
---|
143 | |
---|
144 | int start = range.start(); |
---|
145 | int end = range.end(); |
---|
146 | |
---|
147 | ct_assert(start>=0); |
---|
148 | ct_assert(end<no_of_entries); |
---|
149 | ct_assert(start<=end); |
---|
150 | |
---|
151 | int i; |
---|
152 | if (table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
153 | if (Sub.table_entry_size==SHORT_TABLE_ELEM_SIZE && |
---|
154 | Add.table_entry_size==SHORT_TABLE_ELEM_SIZE) |
---|
155 | { |
---|
156 | for (i=start; i<=end; i++) { |
---|
157 | set_elem_short(i, get_elem_short(i)-Sub.get_elem_short(i)+Add.get_elem_short(i)); |
---|
158 | } |
---|
159 | } |
---|
160 | else { |
---|
161 | INVALID_TABLE_OPERATION(); // cannot add or sub long to/from short |
---|
162 | } |
---|
163 | } |
---|
164 | else { |
---|
165 | if (Sub.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
166 | if (Add.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
167 | for (i=start; i<=end; i++) { // LOOP_VECTORIZED[!>=620<7] |
---|
168 | set_elem_long(i, get_elem_long(i)-Sub.get_elem_short(i)+Add.get_elem_short(i)); |
---|
169 | } |
---|
170 | } |
---|
171 | else { |
---|
172 | for (i=start; i<=end; i++) { // LOOP_VECTORIZED[!>=620<7] |
---|
173 | set_elem_long(i, get_elem_long(i)-Sub.get_elem_short(i)+Add.get_elem_long(i)); |
---|
174 | } |
---|
175 | } |
---|
176 | } |
---|
177 | else { |
---|
178 | if (Add.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
179 | for (i=start; i<=end; i++) { // LOOP_VECTORIZED[!>=620<7] |
---|
180 | set_elem_long(i, get_elem_long(i)-Sub.get_elem_long(i)+Add.get_elem_short(i)); |
---|
181 | } |
---|
182 | } |
---|
183 | else { |
---|
184 | for (i=start; i<=end; i++) { // LOOP_VECTORIZED[!>=620<7] |
---|
185 | set_elem_long(i, get_elem_long(i)-Sub.get_elem_long(i)+Add.get_elem_long(i)); |
---|
186 | } |
---|
187 | } |
---|
188 | } |
---|
189 | } |
---|
190 | } |
---|
191 | |
---|
192 | int SepBaseFreq::firstDifference(const SepBaseFreq& other, int start, int end, int *firstDifferentPos) const { |
---|
193 | int i; |
---|
194 | int result = 0; |
---|
195 | |
---|
196 | if (table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
197 | if (other.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
198 | for (i=start; i<=end; i++) { |
---|
199 | if (get_elem_short(i)!=other.get_elem_short(i)) { |
---|
200 | *firstDifferentPos = i; |
---|
201 | result = 1; |
---|
202 | break; |
---|
203 | } |
---|
204 | } |
---|
205 | } |
---|
206 | else { |
---|
207 | for (i=start; i<=end; i++) { |
---|
208 | if (get_elem_short(i)!=other.get_elem_long(i)) { |
---|
209 | *firstDifferentPos = i; |
---|
210 | result = 1; |
---|
211 | break; |
---|
212 | } |
---|
213 | } |
---|
214 | } |
---|
215 | } |
---|
216 | else { |
---|
217 | if (other.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
218 | for (i=start; i<=end; i++) { |
---|
219 | if (get_elem_long(i)!=other.get_elem_short(i)) { |
---|
220 | *firstDifferentPos = i; |
---|
221 | result = 1; |
---|
222 | break; |
---|
223 | } |
---|
224 | } |
---|
225 | } |
---|
226 | else { |
---|
227 | for (i=start; i<=end; i++) { |
---|
228 | if (get_elem_long(i)!=other.get_elem_long(i)) { |
---|
229 | *firstDifferentPos = i; |
---|
230 | result = 1; |
---|
231 | break; |
---|
232 | } |
---|
233 | } |
---|
234 | } |
---|
235 | } |
---|
236 | |
---|
237 | return result; |
---|
238 | } |
---|
239 | int SepBaseFreq::lastDifference(const SepBaseFreq& other, int start, int end, int *lastDifferentPos) const { |
---|
240 | int i; |
---|
241 | int result = 0; |
---|
242 | |
---|
243 | if (table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
244 | if (other.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
245 | for (i=end; i>=start; i--) { |
---|
246 | if (get_elem_short(i)!=other.get_elem_short(i)) { |
---|
247 | *lastDifferentPos = i; |
---|
248 | result = 1; |
---|
249 | break; |
---|
250 | } |
---|
251 | } |
---|
252 | } |
---|
253 | else { |
---|
254 | for (i=end; i>=start; i--) { |
---|
255 | if (get_elem_short(i)!=other.get_elem_long(i)) { |
---|
256 | *lastDifferentPos = i; |
---|
257 | result = 1; |
---|
258 | break; |
---|
259 | } |
---|
260 | } |
---|
261 | } |
---|
262 | } |
---|
263 | else { |
---|
264 | if (other.table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
265 | for (i=end; i>=start; i--) { |
---|
266 | if (get_elem_long(i)!=other.get_elem_short(i)) { |
---|
267 | *lastDifferentPos = i; |
---|
268 | result = 1; |
---|
269 | break; |
---|
270 | } |
---|
271 | } |
---|
272 | } |
---|
273 | else { |
---|
274 | for (i=end; i>=start; i--) { |
---|
275 | if (get_elem_long(i)!=other.get_elem_long(i)) { |
---|
276 | *lastDifferentPos = i; |
---|
277 | result = 1; |
---|
278 | break; |
---|
279 | } |
---|
280 | } |
---|
281 | } |
---|
282 | } |
---|
283 | |
---|
284 | return result; |
---|
285 | } |
---|
286 | |
---|
287 | |
---|
288 | SepBaseFreq::SepBaseFreq(int maxseqlength) |
---|
289 | : table_entry_size(SHORT_TABLE_ELEM_SIZE), |
---|
290 | no_of_entries(0) |
---|
291 | { |
---|
292 | no_of_bases.shortTable = NULp; |
---|
293 | if (maxseqlength) init(maxseqlength); |
---|
294 | } |
---|
295 | |
---|
296 | SepBaseFreq::~SepBaseFreq() { |
---|
297 | delete [] no_of_bases.shortTable; |
---|
298 | #ifdef COUNT_BASES_TABLE_SIZE |
---|
299 | bases_table_size -= (no_of_entries+1)*table_entry_size; |
---|
300 | printf("bases_table_size = %li\n", bases_table_size); |
---|
301 | #endif |
---|
302 | } |
---|
303 | |
---|
304 | void SepBaseFreq::change_table_length(int new_length, int default_entry) { |
---|
305 | if (new_length!=no_of_entries) { |
---|
306 | int min_length = new_length<no_of_entries ? new_length : no_of_entries; |
---|
307 | int growth = new_length-no_of_entries; |
---|
308 | |
---|
309 | if (table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
310 | unsigned char *new_table = new unsigned char[new_length+1]; |
---|
311 | |
---|
312 | ct_assert(default_entry>=0 && default_entry<256); |
---|
313 | |
---|
314 | memcpy(new_table, no_of_bases.shortTable, min_length*sizeof(*new_table)); |
---|
315 | new_table[new_length] = no_of_bases.shortTable[no_of_entries]; |
---|
316 | if (growth>0) { |
---|
317 | for (int e=no_of_entries; e<new_length; ++e) new_table[e] = default_entry; |
---|
318 | } |
---|
319 | |
---|
320 | delete [] no_of_bases.shortTable; |
---|
321 | no_of_bases.shortTable = new_table; |
---|
322 | } |
---|
323 | else { |
---|
324 | int *new_table = new int[new_length+1]; |
---|
325 | |
---|
326 | memcpy(new_table, no_of_bases.longTable, min_length*sizeof(*new_table)); |
---|
327 | new_table[new_length] = no_of_bases.longTable[no_of_entries]; |
---|
328 | if (growth>0) { |
---|
329 | for (int e=no_of_entries; e<new_length; ++e) { // LOOP_VECTORIZED |
---|
330 | new_table[e] = default_entry; |
---|
331 | } |
---|
332 | } |
---|
333 | |
---|
334 | delete [] no_of_bases.longTable; |
---|
335 | no_of_bases.longTable = new_table; |
---|
336 | } |
---|
337 | |
---|
338 | #ifdef COUNT_BASES_TABLE_SIZE |
---|
339 | bases_table_size += growth*table_entry_size; |
---|
340 | printf("bases_table_size = %li\n", bases_table_size); |
---|
341 | #endif |
---|
342 | no_of_entries = new_length; |
---|
343 | } |
---|
344 | } |
---|
345 | |
---|
346 | #if defined(TEST_CHAR_TABLE_INTEGRITY) || defined(ASSERTION_USED) |
---|
347 | |
---|
348 | int SepBaseFreq::empty() const { |
---|
349 | int i; |
---|
350 | |
---|
351 | if (table_entry_size==SHORT_TABLE_ELEM_SIZE) { |
---|
352 | for (i=0; i<no_of_entries; i++) { |
---|
353 | if (get_elem_short(i)) { |
---|
354 | return 0; |
---|
355 | } |
---|
356 | } |
---|
357 | } |
---|
358 | else { |
---|
359 | for (i=0; i<no_of_entries; i++) { |
---|
360 | if (get_elem_long(i)) { |
---|
361 | return 0; |
---|
362 | } |
---|
363 | } |
---|
364 | } |
---|
365 | |
---|
366 | return 1; |
---|
367 | } |
---|
368 | #endif // ASSERTION_USED |
---|
369 | |
---|
370 | char *BaseFrequencies::build_consensus_string(PosRange r, const ConsensusBuildParams& cbp) const { |
---|
371 | ExplicitRange range(r, size()); |
---|
372 | |
---|
373 | long entries = range.size(); |
---|
374 | char *new_buf = ARB_alloc<char>(entries+1); |
---|
375 | |
---|
376 | build_consensus_string_to(new_buf, range, cbp); |
---|
377 | new_buf[entries] = 0; |
---|
378 | |
---|
379 | return new_buf; |
---|
380 | } |
---|
381 | |
---|
382 | #if defined(DEBUG) |
---|
383 | // #define DEBUG_CONSENSUS |
---|
384 | #endif |
---|
385 | |
---|
386 | void BaseFrequencies::build_consensus_string_to(char *consensus_string, ExplicitRange range, const ConsensusBuildParams& BK) const { |
---|
387 | // 'consensus_string' has to be a buffer of size 'range.size()+1' |
---|
388 | // Note : Always check that consensus behavior is identical to that used in CON_evaluatestatistic() |
---|
389 | |
---|
390 | ct_assert(consensus_string); |
---|
391 | ct_assert(range.end()<size()); |
---|
392 | |
---|
393 | #define PERCENT(part, all) ((100*(part))/(all)) |
---|
394 | #define MAX_BASES_TABLES 41 // 25 |
---|
395 | |
---|
396 | ct_assert(used_bases_tables<=MAX_BASES_TABLES); // this is correct for DNA/RNA -> build_consensus_string() must be corrected for AMI/PRO |
---|
397 | |
---|
398 | const int left_idx = range.start(); |
---|
399 | const int right_idx = range.end(); |
---|
400 | |
---|
401 | #if defined(DEBUG_CONSENSUS) |
---|
402 | #define DUMPINT(var) do { if (dumpcol) fprintf(stderr, "%s=%i ", #var, var); } while(0) |
---|
403 | #else |
---|
404 | #define DUMPINT(var) |
---|
405 | #endif |
---|
406 | |
---|
407 | if (sequenceUnits) { |
---|
408 | for (int i=left_idx; i<=right_idx; i++) { |
---|
409 | #if defined(DEBUG_CONSENSUS) |
---|
410 | int column = i+1; |
---|
411 | bool dumpcol = (column == 21); |
---|
412 | DUMPINT(column); |
---|
413 | #endif |
---|
414 | |
---|
415 | const int o = i-left_idx; // output offset |
---|
416 | |
---|
417 | int bases = 0; // count of all bases together |
---|
418 | int base[MAX_BASES_TABLES]; // count of specific base |
---|
419 | int max_base = -1; // maximum count of specific base |
---|
420 | int max_base_idx = -1; // index of this base |
---|
421 | |
---|
422 | for (int j=0; j<used_bases_tables; j++) { |
---|
423 | base[j] = linear_table(j)[i]; |
---|
424 | if (!isGap(index_to_upperChar(j))) { |
---|
425 | bases += base[j]; |
---|
426 | if (base[j]>max_base) { // search for most used base |
---|
427 | max_base = base[j]; |
---|
428 | max_base_idx = j; |
---|
429 | } |
---|
430 | } |
---|
431 | } |
---|
432 | |
---|
433 | int gaps = sequenceUnits-bases; // count of gaps |
---|
434 | |
---|
435 | DUMPINT(sequenceUnits); |
---|
436 | DUMPINT(gaps); |
---|
437 | DUMPINT(bases); |
---|
438 | |
---|
439 | // What to do with gaps? |
---|
440 | |
---|
441 | if (gaps == sequenceUnits) { |
---|
442 | consensus_string[o] = '='; |
---|
443 | } |
---|
444 | else if (BK.countgaps && PERCENT(gaps, sequenceUnits) >= BK.gapbound) { |
---|
445 | DUMPINT(PERCENT(gaps,sequenceUnits)); |
---|
446 | DUMPINT(BK.gapbound); |
---|
447 | consensus_string[o] = '-'; |
---|
448 | } |
---|
449 | else { |
---|
450 | char kchar = 0; // character for consensus |
---|
451 | int kcount = 0; // count this character |
---|
452 | |
---|
453 | if (BK.group) { // simplify using IUPAC |
---|
454 | if (ali_type == GB_AT_RNA || ali_type == GB_AT_DNA) { |
---|
455 | int no_of_bases = 0; // count of different bases used to create iupac |
---|
456 | char used_bases[MAX_BASES_TABLES+1]; // string containing those bases |
---|
457 | |
---|
458 | for (int j=0; j<used_bases_tables; j++) { |
---|
459 | int bchar = index_to_upperChar(j); |
---|
460 | |
---|
461 | if (!isGap(bchar)) { |
---|
462 | if (base[j]>0 && PERCENT(base[j],bases) >= BK.considbound) { |
---|
463 | #if defined(DEBUG_CONSENSUS) |
---|
464 | if (!kcount) DUMPINT(BK.considbound); |
---|
465 | #endif |
---|
466 | used_bases[no_of_bases++] = index_to_upperChar(j); |
---|
467 | kcount += base[j]; |
---|
468 | |
---|
469 | DUMPINT(base[j]); |
---|
470 | DUMPINT(PERCENT(base[j],bases)); |
---|
471 | DUMPINT(kcount); |
---|
472 | } |
---|
473 | } |
---|
474 | } |
---|
475 | used_bases[no_of_bases] = 0; |
---|
476 | kchar = iupac::encode(used_bases, ali_type); |
---|
477 | } |
---|
478 | else { |
---|
479 | ct_assert(ali_type == GB_AT_AA); |
---|
480 | |
---|
481 | const int amino_groups = iupac::AA_GROUP_COUNT; |
---|
482 | int group_count[amino_groups]; |
---|
483 | |
---|
484 | for (int j=0; j<amino_groups; j++) { |
---|
485 | group_count[j] = 0; |
---|
486 | } |
---|
487 | for (int j=0; j<used_bases_tables; j++) { |
---|
488 | unsigned char bchar = index_to_upperChar(j); |
---|
489 | |
---|
490 | if (!isGap(bchar)) { |
---|
491 | group_count[iupac::get_amino_group_for(bchar)] += base[j]; |
---|
492 | } |
---|
493 | } |
---|
494 | |
---|
495 | int bestGroup = 0; |
---|
496 | for (int j=0; j<amino_groups; j++) { |
---|
497 | if (group_count[j]>kcount) { |
---|
498 | bestGroup = j; |
---|
499 | kcount = group_count[j]; |
---|
500 | } |
---|
501 | } |
---|
502 | |
---|
503 | ct_assert(kcount>0); |
---|
504 | if (PERCENT(kcount, bases) >= BK.considbound) { |
---|
505 | kchar = iupac::get_amino_consensus_char(iupac::Amino_Group(bestGroup)); |
---|
506 | } |
---|
507 | else { |
---|
508 | kchar = 'X'; |
---|
509 | kcount = bases; |
---|
510 | } |
---|
511 | } |
---|
512 | } |
---|
513 | if (!kcount) { // IUPAC grouping is either off OR didnt consider any bases |
---|
514 | ct_assert(max_base); // expect at least one base to occur |
---|
515 | ct_assert(max_base_idx >= 0); |
---|
516 | |
---|
517 | kchar = index_to_upperChar(max_base_idx); |
---|
518 | kcount = max_base; |
---|
519 | } |
---|
520 | |
---|
521 | ct_assert(kchar); |
---|
522 | ct_assert(kcount); |
---|
523 | ct_assert(kcount<=bases); |
---|
524 | |
---|
525 | // show as upper or lower case ? |
---|
526 | int percent = PERCENT(kcount, BK.countgaps ? sequenceUnits : bases); |
---|
527 | DUMPINT(percent); |
---|
528 | DUMPINT(BK.upper); |
---|
529 | DUMPINT(BK.lower); |
---|
530 | if (percent>=BK.upper) { |
---|
531 | consensus_string[o] = kchar; |
---|
532 | } |
---|
533 | else if (percent>=BK.lower) { |
---|
534 | consensus_string[o] = tolower(kchar); |
---|
535 | } |
---|
536 | else { |
---|
537 | consensus_string[o] = '.'; |
---|
538 | } |
---|
539 | } |
---|
540 | ct_assert(consensus_string[o]); |
---|
541 | |
---|
542 | #if defined(DEBUG_CONSENSUS) |
---|
543 | if (dumpcol) fprintf(stderr, "result='%c'\n", consensus_string[o]); |
---|
544 | #endif |
---|
545 | } |
---|
546 | } |
---|
547 | else { |
---|
548 | memset(consensus_string, '?', right_idx-left_idx+1); |
---|
549 | } |
---|
550 | #undef PERCENT |
---|
551 | } |
---|
552 | |
---|
553 | // ----------------------- |
---|
554 | // BaseFrequencies |
---|
555 | |
---|
556 | bool BaseFrequencies::initialized = false; |
---|
557 | Ambiguity BaseFrequencies::ambiguity_table[MAX_AMBIGUITY_CODES]; |
---|
558 | uint8_t BaseFrequencies::unitsPerSequence = 1; |
---|
559 | unsigned char BaseFrequencies::char_to_index_tab[MAXCHARTABLE]; |
---|
560 | bool BaseFrequencies::is_gap[MAXCHARTABLE]; |
---|
561 | unsigned char BaseFrequencies::upper_index_chars[MAX_USED_BASES_TABLES]; |
---|
562 | int BaseFrequencies::used_bases_tables = 0; |
---|
563 | GB_alignment_type BaseFrequencies::ali_type = GB_AT_RNA; |
---|
564 | |
---|
565 | inline void BaseFrequencies::set_char_to_index(unsigned char c, int index) { |
---|
566 | ct_assert(index>=0 && index<used_bases_tables); |
---|
567 | char_to_index_tab[upper_index_chars[index] = toupper(c)] = index; |
---|
568 | char_to_index_tab[tolower(c)] = index; |
---|
569 | } |
---|
570 | |
---|
571 | void BaseFrequencies::expand_tables() { |
---|
572 | int i; |
---|
573 | for (i=0; i<used_bases_tables; i++) { |
---|
574 | linear_table(i).expand_table_entry_size(); |
---|
575 | } |
---|
576 | } |
---|
577 | |
---|
578 | static const char *getAAgroupDefinition() { |
---|
579 | using namespace iupac; |
---|
580 | static SmartCharPtr aadef; |
---|
581 | |
---|
582 | if (aadef.isNull()) { |
---|
583 | char buffer[2*26+1]; |
---|
584 | int off = 0; |
---|
585 | for (int c = 'A'; c<='Z'; ++c) { |
---|
586 | switch (get_amino_group_for(c)) { |
---|
587 | case AA_GROUP_NONE: break; // cannot add: would add invalid characters -> add 'X' manually below. |
---|
588 | case AA_GROUP_ILLEGAL: ct_assert(0); break; |
---|
589 | default: |
---|
590 | buffer[off++] = c; |
---|
591 | buffer[off++] = ','; |
---|
592 | break; |
---|
593 | } |
---|
594 | } |
---|
595 | buffer[off] = 'X'; |
---|
596 | buffer[off+1] = 0; |
---|
597 | |
---|
598 | aadef = strdup(buffer); |
---|
599 | } |
---|
600 | |
---|
601 | return &*aadef; |
---|
602 | } |
---|
603 | |
---|
604 | void BaseFrequencies::setup(const char *gap_chars, GB_alignment_type ali_type_) { |
---|
605 | const char *groups = NULp; |
---|
606 | const char *ambiguous = NULp; |
---|
607 | |
---|
608 | ali_type = ali_type_; |
---|
609 | |
---|
610 | switch (ali_type) { |
---|
611 | case GB_AT_RNA: |
---|
612 | groups = "A,C,G,TU,"; // Note: terminal ',' defines a group for unknown characters |
---|
613 | ambiguous = "MRWSYKVHDBN"; |
---|
614 | unitsPerSequence = 12; |
---|
615 | break; |
---|
616 | |
---|
617 | case GB_AT_DNA: |
---|
618 | groups = "A,C,G,UT,"; |
---|
619 | ambiguous = "MRWSYKVHDBN"; |
---|
620 | unitsPerSequence = 12; |
---|
621 | break; |
---|
622 | |
---|
623 | case GB_AT_AA: |
---|
624 | groups = getAAgroupDefinition(); |
---|
625 | unitsPerSequence = 1; |
---|
626 | break; |
---|
627 | |
---|
628 | default: |
---|
629 | ct_assert(0); |
---|
630 | break; |
---|
631 | } |
---|
632 | |
---|
633 | ct_assert(groups); |
---|
634 | #if defined(ASSERTION_USED) |
---|
635 | if (ali_type == GB_AT_AA) { |
---|
636 | ct_assert(!ambiguous); |
---|
637 | // ct_assert(strlen(ambiguous) <= MAX_AMBIGUITY_CODES); // @@@ later |
---|
638 | } |
---|
639 | else { |
---|
640 | ct_assert(strlen(ambiguous) == MAX_AMBIGUITY_CODES); |
---|
641 | } |
---|
642 | #endif |
---|
643 | |
---|
644 | for (int i = 0; i<MAXCHARTABLE; ++i) is_gap[i] = false; |
---|
645 | |
---|
646 | int gapcharcount = 0; |
---|
647 | char *unique_gap_chars = strdup(gap_chars); |
---|
648 | for (int i = 0; gap_chars[i]; ++i) { |
---|
649 | if (!is_gap[safeCharIndex(gap_chars[i])]) { // ignore duplicate occurrences in 'gap_chars' |
---|
650 | is_gap[safeCharIndex(gap_chars[i])] = true; |
---|
651 | unique_gap_chars[gapcharcount++] = gap_chars[i]; |
---|
652 | } |
---|
653 | } |
---|
654 | |
---|
655 | used_bases_tables = gapcharcount; |
---|
656 | |
---|
657 | for (int i=0; groups[i]; i++) { |
---|
658 | if (groups[i]==',') { |
---|
659 | used_bases_tables++; |
---|
660 | } |
---|
661 | } |
---|
662 | |
---|
663 | ct_assert(used_bases_tables<=MAX_USED_BASES_TABLES); |
---|
664 | |
---|
665 | int idx = 0; |
---|
666 | unsigned char init_val = used_bases_tables-1; // map all unknown stuff to last table |
---|
667 | memset(char_to_index_tab, init_val, MAXCHARTABLE*sizeof(*char_to_index_tab)); |
---|
668 | |
---|
669 | for (int i=0; groups[i]; i++) { |
---|
670 | if (groups[i]==',') { |
---|
671 | idx++; |
---|
672 | } |
---|
673 | else { |
---|
674 | ct_assert(isupper(groups[i])); |
---|
675 | set_char_to_index(groups[i], idx); |
---|
676 | } |
---|
677 | } |
---|
678 | |
---|
679 | const char *align_string_ptr = unique_gap_chars; |
---|
680 | while (1) { |
---|
681 | char c = *align_string_ptr++; |
---|
682 | if (!c) break; |
---|
683 | set_char_to_index(c, idx++); |
---|
684 | } |
---|
685 | |
---|
686 | if (ambiguous) { |
---|
687 | ct_assert(ali_type == GB_AT_DNA || ali_type == GB_AT_RNA); // @@@ amino ambiguities not impl |
---|
688 | |
---|
689 | const uint8_t indices2increment[MAX_TARGET_INDICES+1] = { 0, 0, 6, 4, 3 }; |
---|
690 | for (int i = 0; ambiguous[i]; ++i) { |
---|
691 | const char *contained = iupac::decode(ambiguous[i], ali_type, false); |
---|
692 | |
---|
693 | Ambiguity& amb = ambiguity_table[i]; |
---|
694 | |
---|
695 | amb.indices = strlen(contained); |
---|
696 | ct_assert(amb.indices>=2 && amb.indices<=MAX_TARGET_INDICES); |
---|
697 | amb.increment = indices2increment[amb.indices]; |
---|
698 | |
---|
699 | for (int j = 0; j<amb.indices; ++j) { |
---|
700 | int cidx = char_to_index_tab[safeCharIndex(contained[j])]; |
---|
701 | ct_assert(cidx<MAX_INDEX_TABLES); // has to be non-ambiguous |
---|
702 | amb.index[j] = cidx; |
---|
703 | } |
---|
704 | |
---|
705 | char_to_index_tab[toupper(ambiguous[i])] = |
---|
706 | char_to_index_tab[tolower(ambiguous[i])] = i+MAX_INDEX_TABLES; |
---|
707 | } |
---|
708 | } |
---|
709 | |
---|
710 | free(unique_gap_chars); |
---|
711 | ct_assert(idx==used_bases_tables); |
---|
712 | initialized = true; |
---|
713 | } |
---|
714 | |
---|
715 | BaseFrequencies::BaseFrequencies(int maxseqlength) |
---|
716 | : ignore(0) |
---|
717 | { |
---|
718 | ct_assert(initialized); |
---|
719 | |
---|
720 | bases_table = new SepBaseFreqPtr[used_bases_tables]; |
---|
721 | |
---|
722 | for (int i=0; i<used_bases_tables; i++) { |
---|
723 | bases_table[i] = new SepBaseFreq(maxseqlength); |
---|
724 | } |
---|
725 | |
---|
726 | sequenceUnits = 0; |
---|
727 | } |
---|
728 | |
---|
729 | void BaseFrequencies::init(int maxseqlength) { |
---|
730 | int i; |
---|
731 | for (i=0; i<used_bases_tables; i++) { |
---|
732 | linear_table(i).init(maxseqlength); |
---|
733 | } |
---|
734 | |
---|
735 | sequenceUnits = 0; |
---|
736 | } |
---|
737 | |
---|
738 | void BaseFrequencies::bases_and_gaps_at(int column, int *bases, int *gaps) const { |
---|
739 | int b = 0; |
---|
740 | int g = 0; |
---|
741 | |
---|
742 | for (int i=0; i<used_bases_tables; i++) { |
---|
743 | char c = upper_index_chars[i]; |
---|
744 | |
---|
745 | if (isGap(c)) { |
---|
746 | g += table(c)[column]; |
---|
747 | } |
---|
748 | else { |
---|
749 | b += table(c)[column]; |
---|
750 | } |
---|
751 | } |
---|
752 | |
---|
753 | if (bases) { |
---|
754 | *bases = b/unitsPerSequence; |
---|
755 | ct_assert((b%unitsPerSequence) == 0); // could happen if an ambiguous code contains a gap |
---|
756 | } |
---|
757 | if (gaps) { |
---|
758 | *gaps = g/unitsPerSequence; |
---|
759 | ct_assert((g%unitsPerSequence) == 0); // could happen if an ambiguous code contains a gap |
---|
760 | } |
---|
761 | } |
---|
762 | |
---|
763 | double BaseFrequencies::max_frequency_at(int column, bool ignore_gaps) const { |
---|
764 | int gaps = 0; |
---|
765 | int minus = 0; |
---|
766 | int maxb = 0; |
---|
767 | |
---|
768 | for (int i=0; i<used_bases_tables; i++) { |
---|
769 | char c = upper_index_chars[i]; |
---|
770 | |
---|
771 | if (isGap(c)) { |
---|
772 | if (c == '-') { |
---|
773 | minus += table(c)[column]; |
---|
774 | } |
---|
775 | else { |
---|
776 | gaps += table(c)[column]; |
---|
777 | } |
---|
778 | } |
---|
779 | else { |
---|
780 | maxb = std::max(maxb, table(c)[column]); |
---|
781 | } |
---|
782 | } |
---|
783 | |
---|
784 | if (ignore_gaps) { |
---|
785 | gaps += minus; |
---|
786 | } |
---|
787 | else { |
---|
788 | maxb = std::max(maxb, minus); |
---|
789 | } |
---|
790 | |
---|
791 | double mfreq; |
---|
792 | if (sequenceUnits>gaps) { |
---|
793 | mfreq = maxb/double(sequenceUnits-gaps); |
---|
794 | } |
---|
795 | else { // only gaps (but no '-') |
---|
796 | ct_assert(sequenceUnits == gaps); |
---|
797 | mfreq = 0.0; |
---|
798 | } |
---|
799 | |
---|
800 | #if defined(ASSERTION_USED) |
---|
801 | if (mfreq != 0.0) { |
---|
802 | if (ali_type != GB_AT_AA) { |
---|
803 | ct_assert(mfreq>=0.2); // wrong for amino acids |
---|
804 | } |
---|
805 | ct_assert(mfreq<=1.0); |
---|
806 | } |
---|
807 | #endif |
---|
808 | |
---|
809 | return mfreq; |
---|
810 | } |
---|
811 | |
---|
812 | BaseFrequencies::~BaseFrequencies() { |
---|
813 | int i; |
---|
814 | for (i=0; i<used_bases_tables; i++) { |
---|
815 | delete bases_table[i]; |
---|
816 | } |
---|
817 | |
---|
818 | delete [] bases_table; |
---|
819 | } |
---|
820 | |
---|
821 | const PosRange *BaseFrequencies::changed_range(const BaseFrequencies& other) const { |
---|
822 | int i; |
---|
823 | int Size = size(); |
---|
824 | int start = Size-1; |
---|
825 | int end = 0; |
---|
826 | |
---|
827 | ct_assert(Size==other.size()); |
---|
828 | for (i=0; i<used_bases_tables; i++) { |
---|
829 | if (linear_table(i).firstDifference(other.linear_table(i), 0, start, &start)) { |
---|
830 | if (end<start) { |
---|
831 | end = start; |
---|
832 | } |
---|
833 | linear_table(i).lastDifference(other.linear_table(i), end+1, Size-1, &end); |
---|
834 | |
---|
835 | for (; i<used_bases_tables; i++) { |
---|
836 | linear_table(i).firstDifference(other.linear_table(i), 0, start-1, &start); |
---|
837 | if (end<start) { |
---|
838 | end = start; |
---|
839 | } |
---|
840 | linear_table(i).lastDifference(other.linear_table(i), end+1, Size-1, &end); |
---|
841 | } |
---|
842 | |
---|
843 | ct_assert(start<=end); |
---|
844 | |
---|
845 | static PosRange range; |
---|
846 | range = PosRange(start, end); |
---|
847 | |
---|
848 | return ⦥ |
---|
849 | } |
---|
850 | } |
---|
851 | return NULp; |
---|
852 | } |
---|
853 | void BaseFrequencies::add(const BaseFrequencies& other) { |
---|
854 | if (other.ignore) return; |
---|
855 | if (other.size()==0) return; |
---|
856 | prepare_to_add_elements(other.added_sequences()); |
---|
857 | add(other, 0, other.size()-1); |
---|
858 | } |
---|
859 | void BaseFrequencies::add(const BaseFrequencies& other, int start, int end) { |
---|
860 | if (other.ignore) return; |
---|
861 | |
---|
862 | test(); |
---|
863 | other.test(); |
---|
864 | |
---|
865 | ct_assert(start<=end); |
---|
866 | |
---|
867 | int i; |
---|
868 | for (i=0; i<used_bases_tables; i++) { |
---|
869 | linear_table(i).add(other.linear_table(i), start, end); |
---|
870 | } |
---|
871 | |
---|
872 | sequenceUnits += other.sequenceUnits; |
---|
873 | |
---|
874 | test(); |
---|
875 | } |
---|
876 | |
---|
877 | void BaseFrequencies::sub(const BaseFrequencies& other) { |
---|
878 | if (other.ignore) return; |
---|
879 | sub(other, 0, other.size()-1); |
---|
880 | } |
---|
881 | |
---|
882 | void BaseFrequencies::sub(const BaseFrequencies& other, int start, int end) { |
---|
883 | if (other.ignore) return; |
---|
884 | |
---|
885 | test(); |
---|
886 | other.test(); |
---|
887 | ct_assert(start<=end); |
---|
888 | |
---|
889 | int i; |
---|
890 | for (i=0; i<used_bases_tables; i++) { |
---|
891 | linear_table(i).sub(other.linear_table(i), start, end); |
---|
892 | } |
---|
893 | |
---|
894 | sequenceUnits -= other.sequenceUnits; |
---|
895 | |
---|
896 | test(); |
---|
897 | } |
---|
898 | |
---|
899 | void BaseFrequencies::sub_and_add(const BaseFrequencies& Sub, const BaseFrequencies& Add, PosRange range) { |
---|
900 | test(); |
---|
901 | Sub.test(); |
---|
902 | Add.test(); |
---|
903 | |
---|
904 | ct_assert(!Sub.ignore && !Add.ignore); |
---|
905 | ct_assert(range.is_part()); |
---|
906 | |
---|
907 | int i; |
---|
908 | for (i=0; i<used_bases_tables; i++) { |
---|
909 | linear_table(i).sub_and_add(Sub.linear_table(i), Add.linear_table(i), range); |
---|
910 | } |
---|
911 | |
---|
912 | test(); |
---|
913 | } |
---|
914 | |
---|
915 | const PosRange *BaseFrequencies::changed_range(const char *string1, const char *string2, int min_len) { |
---|
916 | const unsigned long *range1 = (const unsigned long *)string1; |
---|
917 | const unsigned long *range2 = (const unsigned long *)string2; |
---|
918 | const int step = sizeof(*range1); |
---|
919 | |
---|
920 | int l = min_len/step; |
---|
921 | int r = min_len%step; |
---|
922 | int i; |
---|
923 | int j; |
---|
924 | int k; |
---|
925 | |
---|
926 | int start = -1, end = -1; |
---|
927 | |
---|
928 | for (i=0; i<l; i++) { // search wordwise (it's faster) |
---|
929 | if (range1[i] != range2[i]) { |
---|
930 | k = i*step; |
---|
931 | for (j=0; j<step; j++) { |
---|
932 | if (string1[k+j] != string2[k+j]) { |
---|
933 | start = end = k+j; |
---|
934 | break; |
---|
935 | } |
---|
936 | } |
---|
937 | |
---|
938 | break; |
---|
939 | } |
---|
940 | } |
---|
941 | |
---|
942 | k = l*step; |
---|
943 | if (i==l) { // no difference found in word -> look at rest |
---|
944 | for (j=0; j<r; j++) { |
---|
945 | if (string1[k+j] != string2[k+j]) { |
---|
946 | start = end = k+j; |
---|
947 | break; |
---|
948 | } |
---|
949 | } |
---|
950 | |
---|
951 | if (j==r) { // the strings are equal |
---|
952 | return NULp; |
---|
953 | } |
---|
954 | } |
---|
955 | |
---|
956 | ct_assert(start != -1 && end != -1); |
---|
957 | |
---|
958 | for (j=r-1; j>=0; j--) { // search rest backwards |
---|
959 | if (string1[k+j] != string2[k+j]) { |
---|
960 | end = k+j; |
---|
961 | break; |
---|
962 | } |
---|
963 | } |
---|
964 | |
---|
965 | if (j==-1) { // not found in rest -> search words backwards |
---|
966 | int m = start/step; |
---|
967 | for (i=l-1; i>=m; i--) { |
---|
968 | if (range1[i] != range2[i]) { |
---|
969 | k = i*step; |
---|
970 | for (j=step-1; j>=0; j--) { |
---|
971 | if (string1[k+j] != string2[k+j]) { |
---|
972 | end = k+j; |
---|
973 | break; |
---|
974 | } |
---|
975 | } |
---|
976 | break; |
---|
977 | } |
---|
978 | } |
---|
979 | } |
---|
980 | |
---|
981 | ct_assert(start<=end); |
---|
982 | |
---|
983 | static PosRange range; |
---|
984 | range = PosRange(start, end); |
---|
985 | |
---|
986 | return ⦥ |
---|
987 | } |
---|
988 | |
---|
989 | #define INC_DEC(incdec) do { \ |
---|
990 | int idx = char_to_index_tab[c]; \ |
---|
991 | if (idx<MAX_INDEX_TABLES) { \ |
---|
992 | table(c).incdec(offset, unitsPerSequence); \ |
---|
993 | } \ |
---|
994 | else { \ |
---|
995 | Ambiguity& amb = ambiguity_table[idx-MAX_INDEX_TABLES]; \ |
---|
996 | for (int i = 0; i<amb.indices; ++i) { \ |
---|
997 | linear_table(amb.index[i]).incdec(offset, amb.increment); \ |
---|
998 | } \ |
---|
999 | } \ |
---|
1000 | } while(0) |
---|
1001 | |
---|
1002 | void BaseFrequencies::incr_short(unsigned char c, int offset) { |
---|
1003 | INC_DEC(inc_short); |
---|
1004 | } |
---|
1005 | |
---|
1006 | void BaseFrequencies::decr_short(unsigned char c, int offset) { |
---|
1007 | INC_DEC(dec_short); |
---|
1008 | } |
---|
1009 | |
---|
1010 | void BaseFrequencies::incr_long(unsigned char c, int offset) { |
---|
1011 | INC_DEC(inc_long); |
---|
1012 | } |
---|
1013 | |
---|
1014 | void BaseFrequencies::decr_long(unsigned char c, int offset) { |
---|
1015 | INC_DEC(dec_long); |
---|
1016 | } |
---|
1017 | |
---|
1018 | #undef INC_DEC |
---|
1019 | |
---|
1020 | void BaseFrequencies::add(const char *scan_string, int len) { |
---|
1021 | test(); |
---|
1022 | |
---|
1023 | prepare_to_add_elements(1); |
---|
1024 | |
---|
1025 | int i; |
---|
1026 | int sz = size(); |
---|
1027 | |
---|
1028 | if (get_table_entry_size()==SHORT_TABLE_ELEM_SIZE) { |
---|
1029 | for (i=0; i<len; i++) { |
---|
1030 | unsigned char c = scan_string[i]; |
---|
1031 | if (!c) break; |
---|
1032 | incr_short(c, i); |
---|
1033 | } |
---|
1034 | SepBaseFreq& t = table('.'); |
---|
1035 | for (; i<sz; i++) { |
---|
1036 | t.inc_short(i, unitsPerSequence); |
---|
1037 | } |
---|
1038 | } |
---|
1039 | else { |
---|
1040 | for (i=0; i<len; i++) { |
---|
1041 | unsigned char c = scan_string[i]; |
---|
1042 | if (!c) break; |
---|
1043 | incr_long(c, i); |
---|
1044 | } |
---|
1045 | SepBaseFreq& t = table('.'); |
---|
1046 | for (; i<sz; i++) { // LOOP_VECTORIZED[!>=620<7] |
---|
1047 | t.inc_long(i, unitsPerSequence); |
---|
1048 | } |
---|
1049 | } |
---|
1050 | |
---|
1051 | sequenceUnits += unitsPerSequence; |
---|
1052 | |
---|
1053 | test(); |
---|
1054 | } |
---|
1055 | |
---|
1056 | void BaseFrequencies::sub(const char *scan_string, int len) { |
---|
1057 | test(); |
---|
1058 | |
---|
1059 | int i; |
---|
1060 | int sz = size(); |
---|
1061 | |
---|
1062 | if (get_table_entry_size()==SHORT_TABLE_ELEM_SIZE) { |
---|
1063 | for (i=0; i<len; i++) { |
---|
1064 | unsigned char c = scan_string[i]; |
---|
1065 | if (!c) break; |
---|
1066 | decr_short(c, i); |
---|
1067 | } |
---|
1068 | SepBaseFreq& t = table('.'); |
---|
1069 | for (; i<sz; i++) { |
---|
1070 | t.dec_short(i, unitsPerSequence); |
---|
1071 | } |
---|
1072 | } |
---|
1073 | else { |
---|
1074 | for (i=0; i<len; i++) { |
---|
1075 | unsigned char c = scan_string[i]; |
---|
1076 | if (!c) break; |
---|
1077 | decr_long(c, i); |
---|
1078 | } |
---|
1079 | SepBaseFreq& t = table('.'); |
---|
1080 | for (; i<sz; i++) { // LOOP_VECTORIZED[!>=620<7] |
---|
1081 | t.dec_long(i, unitsPerSequence); |
---|
1082 | } |
---|
1083 | } |
---|
1084 | |
---|
1085 | sequenceUnits -= unitsPerSequence; |
---|
1086 | |
---|
1087 | test(); |
---|
1088 | } |
---|
1089 | |
---|
1090 | void BaseFrequencies::sub_and_add(const char *old_string, const char *new_string, PosRange range) { |
---|
1091 | test(); |
---|
1092 | int start = range.start(); |
---|
1093 | int end = range.end(); |
---|
1094 | ct_assert(start<=end); |
---|
1095 | |
---|
1096 | if (get_table_entry_size()==SHORT_TABLE_ELEM_SIZE) { |
---|
1097 | for (int i=start; i<=end; i++) { |
---|
1098 | unsigned char o = old_string[i]; |
---|
1099 | unsigned char n = new_string[i]; |
---|
1100 | |
---|
1101 | ct_assert(o && n); // passed string have to be defined in range |
---|
1102 | |
---|
1103 | if (o!=n) { |
---|
1104 | decr_short(o, i); |
---|
1105 | incr_short(n, i); |
---|
1106 | } |
---|
1107 | } |
---|
1108 | } |
---|
1109 | else { |
---|
1110 | for (int i=start; i<=end; i++) { |
---|
1111 | unsigned char o = old_string[i]; |
---|
1112 | unsigned char n = new_string[i]; |
---|
1113 | |
---|
1114 | ct_assert(o && n); // passed string have to be defined in range |
---|
1115 | |
---|
1116 | if (o!=n) { |
---|
1117 | decr_long(o, i); |
---|
1118 | incr_long(n, i); |
---|
1119 | } |
---|
1120 | } |
---|
1121 | } |
---|
1122 | |
---|
1123 | test(); |
---|
1124 | } |
---|
1125 | |
---|
1126 | void BaseFrequencies::change_table_length(int new_length) { |
---|
1127 | for (int c=0; c<used_bases_tables; c++) { |
---|
1128 | linear_table(c).change_table_length(new_length, 0); |
---|
1129 | } |
---|
1130 | test(); |
---|
1131 | } |
---|
1132 | |
---|
1133 | // -------------- |
---|
1134 | // tests: |
---|
1135 | |
---|
1136 | #if defined(TEST_CHAR_TABLE_INTEGRITY) || defined(ASSERTION_USED) |
---|
1137 | |
---|
1138 | bool BaseFrequencies::empty() const { |
---|
1139 | int i; |
---|
1140 | for (i=0; i<used_bases_tables; i++) { |
---|
1141 | if (!linear_table(i).empty()) { |
---|
1142 | return false; |
---|
1143 | } |
---|
1144 | } |
---|
1145 | return true; |
---|
1146 | } |
---|
1147 | |
---|
1148 | bool BaseFrequencies::ok() const { |
---|
1149 | if (empty()) return true; |
---|
1150 | if (is_ignored()) return true; |
---|
1151 | |
---|
1152 | if (sequenceUnits<0) { |
---|
1153 | fprintf(stderr, "Negative # of sequences (%i) in BaseFrequencies\n", added_sequences()); |
---|
1154 | return false; |
---|
1155 | } |
---|
1156 | |
---|
1157 | const int table_size = size(); |
---|
1158 | for (int i=0; i<table_size; i++) { |
---|
1159 | int bases, gaps; |
---|
1160 | bases_and_gaps_at(i, &bases, &gaps); |
---|
1161 | |
---|
1162 | if (!bases && !gaps) { // this occurs after insert column |
---|
1163 | for (int j=i+1; j<table_size; j++) { // test if we find any bases till end of table !!! |
---|
1164 | |
---|
1165 | bases_and_gaps_at(j, &bases, NULp); |
---|
1166 | if (bases) { // bases found -> empty position was illegal |
---|
1167 | fprintf(stderr, "Zero in BaseFrequencies at column %i\n", i); |
---|
1168 | return false; |
---|
1169 | } |
---|
1170 | } |
---|
1171 | break; |
---|
1172 | } |
---|
1173 | else { |
---|
1174 | if ((bases+gaps)!=added_sequences()) { |
---|
1175 | fprintf(stderr, "bases+gaps should be equal to # of sequences at column %i (bases=%i, gaps=%i, added_sequences=%i)\n", |
---|
1176 | i, bases, gaps, added_sequences()); |
---|
1177 | return false; |
---|
1178 | } |
---|
1179 | } |
---|
1180 | } |
---|
1181 | |
---|
1182 | return true; |
---|
1183 | } |
---|
1184 | |
---|
1185 | #endif // TEST_CHAR_TABLE_INTEGRITY || ASSERTION_USED |
---|
1186 | |
---|
1187 | #if defined(TEST_CHAR_TABLE_INTEGRITY) |
---|
1188 | |
---|
1189 | void BaseFrequencies::test() const { |
---|
1190 | |
---|
1191 | if (!ok()) { |
---|
1192 | GBK_terminate("BaseFrequencies::test() failed"); |
---|
1193 | } |
---|
1194 | } |
---|
1195 | |
---|
1196 | #endif // TEST_CHAR_TABLE_INTEGRITY |
---|
1197 | |
---|
1198 | |
---|
1199 | // -------------------------------------------------------------------------------- |
---|
1200 | |
---|
1201 | #ifdef UNIT_TESTS |
---|
1202 | #ifndef TEST_UNIT_H |
---|
1203 | #include <test_unit.h> |
---|
1204 | #endif |
---|
1205 | #include <arbdb.h> |
---|
1206 | |
---|
1207 | #define SETUP(gapChars,alitype) BaseFrequencies::setup(gapChars,alitype) |
---|
1208 | |
---|
1209 | void TEST_char_table() { |
---|
1210 | for (int useAminoAcids = 0; useAminoAcids<=1; ++useAminoAcids) { |
---|
1211 | const char *alphabeth = useAminoAcids ? "ABCDEFGHIJKLMNPQRSTVWXYZ-" : "ACGTN-"; |
---|
1212 | const int alphabeth_size = strlen(alphabeth); |
---|
1213 | |
---|
1214 | TEST_EXPECT_EQUAL(alphabeth_size, (useAminoAcids ? 23 : 4)+2); // "+2" is for gap and N/X |
---|
1215 | |
---|
1216 | const int SEQLEN = 30; |
---|
1217 | char seq[SEQLEN+1]; |
---|
1218 | |
---|
1219 | const char *gapChars = ".-"; // @@@ doesnt make any difference which gaps are used - why? |
---|
1220 | // const char *gapChars = "-"; |
---|
1221 | // const char *gapChars = "."; |
---|
1222 | // const char *gapChars = "A"; // makes me fail |
---|
1223 | SETUP(gapChars, useAminoAcids ? GB_AT_AA : GB_AT_RNA); |
---|
1224 | |
---|
1225 | if (useAminoAcids) { |
---|
1226 | // document used amino acid group definition: |
---|
1227 | TEST_EXPECT_EQUAL(getAAgroupDefinition(), "A,B,C,D,E,F,G,H,I,J,K,L,M,N,P,Q,R,S,T,V,W,Y,Z,X"); |
---|
1228 | } |
---|
1229 | |
---|
1230 | ConsensusBuildParams BK; |
---|
1231 | // BK.lower = 70; BK.upper = 95; BK.gapbound = 60; // defaults from awars |
---|
1232 | BK.lower = 40; BK.upper = 70; BK.gapbound = 40; |
---|
1233 | |
---|
1234 | const int LOOPS = 5; |
---|
1235 | GB_random_seed(100); |
---|
1236 | |
---|
1237 | int seed[LOOPS]; |
---|
1238 | for (int loop = 0; loop<5; ++loop) { |
---|
1239 | seed[loop] = GB_random(INT_MAX); |
---|
1240 | } |
---|
1241 | |
---|
1242 | for (int loop = 0; loop<5; ++loop) { |
---|
1243 | GB_random_seed(seed[loop]); |
---|
1244 | |
---|
1245 | BaseFrequencies tab(SEQLEN); |
---|
1246 | |
---|
1247 | // build some seq |
---|
1248 | for (int c = 0; c<SEQLEN; ++c) { |
---|
1249 | seq[c] = alphabeth[GB_random(alphabeth_size)]; |
---|
1250 | } |
---|
1251 | seq[SEQLEN] = 0; |
---|
1252 | |
---|
1253 | TEST_EXPECT_EQUAL(strlen(seq), size_t(SEQLEN)); |
---|
1254 | |
---|
1255 | const int add_count = 300; |
---|
1256 | char *added_seqs[add_count]; |
---|
1257 | |
---|
1258 | const char *s1; // length = SEQLEN |
---|
1259 | const char *s2; |
---|
1260 | if (useAminoAcids) { |
---|
1261 | s1 = "ABCDEFGHIJKLMNPQRSTVWXYZabcdef"; |
---|
1262 | s2 = "ghijklmnpqrstvwxyzABCDEFGHIJKL"; |
---|
1263 | } |
---|
1264 | else { // rna |
---|
1265 | s1 = "ACGTACGTAcgtacgtaCGTACGTACGTAC"; // @@@ insert gaps! -> adapt changed results |
---|
1266 | s2 = "MRWSYKVHDBNmrwsykvhdbnSYKVHDBN"; // use ambiguities |
---|
1267 | } |
---|
1268 | |
---|
1269 | // add seq multiple times |
---|
1270 | for (int a = 0; a<add_count; ++a) { |
---|
1271 | tab.add(seq, SEQLEN); |
---|
1272 | added_seqs[a] = strdup(seq); |
---|
1273 | |
---|
1274 | // modify 1 character in seq: |
---|
1275 | int sidx = GB_random(SEQLEN); |
---|
1276 | int aidx = GB_random(alphabeth_size); |
---|
1277 | seq[sidx] = alphabeth[aidx]; |
---|
1278 | |
---|
1279 | if (a == 16) { // with 15 sequences in tab |
---|
1280 | // test sub_and_add (short table version) |
---|
1281 | char *consensus = tab.build_consensus_string(BK); |
---|
1282 | |
---|
1283 | tab.add(s1, SEQLEN); |
---|
1284 | PosRange r(0, SEQLEN-1); |
---|
1285 | tab.sub_and_add(s1, s2, r); |
---|
1286 | tab.sub_and_add(s2, consensus, r); |
---|
1287 | tab.sub(consensus, SEQLEN); |
---|
1288 | |
---|
1289 | char *consensus2 = tab.build_consensus_string(BK); |
---|
1290 | TEST_EXPECT_EQUAL(consensus, consensus2); |
---|
1291 | |
---|
1292 | free(consensus2); |
---|
1293 | free(consensus); |
---|
1294 | } |
---|
1295 | } |
---|
1296 | |
---|
1297 | // build consensi (just check regression) |
---|
1298 | // Note: semantic tests for consensus are in ../../NTREE/AP_consensus.cxx@TEST_nucleotide_consensus_and_maxFrequency |
---|
1299 | { |
---|
1300 | char *consensus = tab.build_consensus_string(BK); |
---|
1301 | if (useAminoAcids) { |
---|
1302 | switch (seed[loop]) { |
---|
1303 | case 1080710223: TEST_EXPECT_EQUAL(consensus, "a.i.dXi.haai.hifdi.d.adXaa..dX"); break; |
---|
1304 | case 290469779: TEST_EXPECT_EQUAL(consensus, ".......I.D.A......-f..iXi...dX"); break; |
---|
1305 | case 346019116: TEST_EXPECT_EQUAL(consensus, "if.a-hd.hd..a.di-ddf.i..aXIaI."); break; |
---|
1306 | case 1708473987: TEST_EXPECT_EQUAL(consensus, "aia..df...haDaadd.d.fha.ii.X.."); break; |
---|
1307 | case 2080862653: TEST_EXPECT_EQUAL(consensus, "a.ia.id...a.aah.had......i..d."); break; |
---|
1308 | default: TEST_EXPECTATION(all().of(that(consensus).is_equal_to("undef"), that(seed[loop]).is_equal_to(-1))); // default case is unwanted (logs data -> want one case above for each pair) |
---|
1309 | } |
---|
1310 | } |
---|
1311 | else { |
---|
1312 | switch (seed[loop]) { |
---|
1313 | case 1080710223: TEST_EXPECT_EQUAL(consensus, "W.gKyc.sR.urSkGcy.kuy......RY-"); break; |
---|
1314 | case 290469779: TEST_EXPECT_EQUAL(consensus, ".u-SW.Mgcu.uw-uu..-mmb.sgcRuY."); break; |
---|
1315 | case 346019116: TEST_EXPECT_EQUAL(consensus, "y-m.-Suy.c.-u.-g-y.cWgr.cyGrgY"); break; |
---|
1316 | case 1708473987: TEST_EXPECT_EQUAL(consensus, ".K.Kc--.cakyw.uW..a-.guuggwH-a"); break; |
---|
1317 | case 2080862653: TEST_EXPECT_EQUAL(consensus, ".sgaR..--.uu.cMuuycsk-u.cu.Wc."); break; |
---|
1318 | default: TEST_EXPECTATION(all().of(that(consensus).is_equal_to("undef"), that(seed[loop]).is_equal_to(-1))); // default case is unwanted (logs data -> want one case above for each pair) |
---|
1319 | } |
---|
1320 | } |
---|
1321 | |
---|
1322 | // test sub_and_add() |
---|
1323 | |
---|
1324 | tab.add(s1, SEQLEN); |
---|
1325 | PosRange r(0, SEQLEN-1); |
---|
1326 | tab.sub_and_add(s1, s2, r); |
---|
1327 | tab.sub_and_add(s2, consensus, r); |
---|
1328 | tab.sub(consensus, SEQLEN); |
---|
1329 | |
---|
1330 | char *consensus2 = tab.build_consensus_string(BK); |
---|
1331 | TEST_EXPECT_EQUAL(consensus, consensus2); |
---|
1332 | |
---|
1333 | free(consensus2); |
---|
1334 | free(consensus); |
---|
1335 | } |
---|
1336 | |
---|
1337 | for (int a = 0; a<add_count; a++) { |
---|
1338 | tab.sub(added_seqs[a], SEQLEN); |
---|
1339 | freenull(added_seqs[a]); |
---|
1340 | } |
---|
1341 | { |
---|
1342 | char *consensus = tab.build_consensus_string(BK); |
---|
1343 | TEST_EXPECT_EQUAL(consensus, "??????????????????????????????"); // check tab is empty |
---|
1344 | free(consensus); |
---|
1345 | } |
---|
1346 | } |
---|
1347 | } |
---|
1348 | } |
---|
1349 | |
---|
1350 | void TEST_SepBaseFreq() { |
---|
1351 | const int LEN = 20; |
---|
1352 | const int OFF1 = 6; |
---|
1353 | const int OFF2 = 7; // adjacent to OFF1 |
---|
1354 | |
---|
1355 | SepBaseFreq short1(LEN), short2(LEN); |
---|
1356 | for (int i = 0; i<100; ++i) short1.inc_short(OFF1, 1); |
---|
1357 | for (int i = 0; i<70; ++i) short1.inc_short(OFF2, 1); |
---|
1358 | for (int i = 0; i<150; ++i) short2.inc_short(OFF1, 1); |
---|
1359 | for (int i = 0; i<80; ++i) short2.inc_short(OFF2, 1); |
---|
1360 | |
---|
1361 | SepBaseFreq long1(LEN), long2(LEN); |
---|
1362 | long1.expand_table_entry_size(); |
---|
1363 | long2.expand_table_entry_size(); |
---|
1364 | for (int i = 0; i<2000; ++i) long1.inc_long(OFF1, 1); |
---|
1365 | for (int i = 0; i<2500; ++i) long2.inc_long(OFF1, 1); |
---|
1366 | |
---|
1367 | SepBaseFreq shortsum(LEN); |
---|
1368 | shortsum.add(short1, 0, LEN-1); TEST_EXPECT_EQUAL(shortsum[OFF1], 100); TEST_EXPECT_EQUAL(shortsum[OFF2], 70); |
---|
1369 | shortsum.add(short2, 0, LEN-1); TEST_EXPECT_EQUAL(shortsum[OFF1], 250); TEST_EXPECT_EQUAL(shortsum[OFF2], 150); |
---|
1370 | shortsum.sub(short1, 0, LEN-1); TEST_EXPECT_EQUAL(shortsum[OFF1], 150); TEST_EXPECT_EQUAL(shortsum[OFF2], 80); |
---|
1371 | |
---|
1372 | shortsum.sub_and_add(short1, short2, PosRange(0, LEN-1)); TEST_EXPECT_EQUAL(shortsum[OFF1], 200); TEST_EXPECT_EQUAL(shortsum[OFF2], 90); |
---|
1373 | |
---|
1374 | // shortsum.add(long1, 0, LEN-1); // invalid operation -> terminate |
---|
1375 | |
---|
1376 | SepBaseFreq longsum(LEN); |
---|
1377 | longsum.expand_table_entry_size(); |
---|
1378 | longsum.add(long1, 0, LEN-1); TEST_EXPECT_EQUAL(longsum[OFF1], 2000); |
---|
1379 | longsum.add(long2, 0, LEN-1); TEST_EXPECT_EQUAL(longsum[OFF1], 4500); |
---|
1380 | longsum.sub(long1, 0, LEN-1); TEST_EXPECT_EQUAL(longsum[OFF1], 2500); |
---|
1381 | longsum.add(short1, 0, LEN-1); TEST_EXPECT_EQUAL(longsum[OFF1], 2600); |
---|
1382 | longsum.sub(short2, 0, LEN-1); TEST_EXPECT_EQUAL(longsum[OFF1], 2450); |
---|
1383 | |
---|
1384 | longsum.sub_and_add(long1, long2, PosRange(0, LEN-1)); TEST_EXPECT_EQUAL(longsum[OFF1], 2950); |
---|
1385 | longsum.sub_and_add(short1, short2, PosRange(0, LEN-1)); TEST_EXPECT_EQUAL(longsum[OFF1], 3000); |
---|
1386 | longsum.sub_and_add(long1, short2, PosRange(0, LEN-1)); TEST_EXPECT_EQUAL(longsum[OFF1], 1150); |
---|
1387 | longsum.sub_and_add(short1, long2, PosRange(0, LEN-1)); TEST_EXPECT_EQUAL(longsum[OFF1], 3550); |
---|
1388 | |
---|
1389 | int pos = -1; |
---|
1390 | TEST_EXPECT_EQUAL(short1.firstDifference(short2, 0, LEN-1, &pos), 1); TEST_EXPECT_EQUAL(pos, OFF1); |
---|
1391 | TEST_EXPECT_EQUAL(short1.lastDifference (short2, 0, LEN-1, &pos), 1); TEST_EXPECT_EQUAL(pos, OFF2); |
---|
1392 | TEST_EXPECT_EQUAL(short1.firstDifference(long1, 0, LEN-1, &pos), 1); TEST_EXPECT_EQUAL(pos, OFF1); |
---|
1393 | TEST_EXPECT_EQUAL(short1.lastDifference (long1, 0, LEN-1, &pos), 1); TEST_EXPECT_EQUAL(pos, OFF2); |
---|
1394 | TEST_EXPECT_EQUAL(long2.firstDifference (short2, 0, LEN-1, &pos), 1); TEST_EXPECT_EQUAL(pos, OFF1); |
---|
1395 | TEST_EXPECT_EQUAL(long2.lastDifference (short2, 0, LEN-1, &pos), 1); TEST_EXPECT_EQUAL(pos, OFF2); |
---|
1396 | TEST_EXPECT_EQUAL(long2.firstDifference (long1, 0, LEN-1, &pos), 1); TEST_EXPECT_EQUAL(pos, OFF1); |
---|
1397 | TEST_EXPECT_EQUAL(long2.lastDifference (long1, 0, LEN-1, &pos), 1); TEST_EXPECT_EQUAL(pos, OFF1); |
---|
1398 | } |
---|
1399 | |
---|
1400 | #endif // UNIT_TESTS |
---|
1401 | |
---|
1402 | // -------------------------------------------------------------------------------- |
---|
1403 | |
---|
1404 | |
---|
1405 | |
---|