ecoprimer.c 35.4 KB
Newer Older
1 2 3 4 5 6 7 8
/*
 * ecoprimer.c
 *
 *  Created on: 7 nov. 2008
 *      Author: coissac
 */

#include "libecoprimer/ecoprimer.h"
9
#include "libecoprimer/PrimerSets.h"
10
#include "libecoprimer/ahocorasick.h"
11 12 13 14 15 16 17
#include <stdio.h>
#include <string.h>
#include <ctype.h>
#include <stdlib.h>
#include <getopt.h>
#include <time.h>
#include <sys/time.h>
18
#include <dlfcn.h>
19 20 21
#include"libthermo/nnparams.h"
#include"libthermo/thermostats.h"

22

23
#define VERSION "0.3"
24
  /* TR: by default, statistics are made on species level*/
Eric Coissac committed
25 26 27
#define DEFAULTTAXONRANK "species"

static int cmpprintedpairs(const void* p1,const void* p2);
28
//float _Z27calculateMeltingTemperature_ (char * seq1, char * seq2);
29 30
pwordcount_t reduce_words_to_debug (pwordcount_t words, poptions_t options);
void print_wordwith_positions (primer_t prm, uint32_t seqdbsize, poptions_t options);
Eric Coissac committed
31

32 33
void* lib_handle = NULL;
float (*calcMelTemp)(char*, char*);
34 35 36 37 38 39 40 41

/* ----------------------------------------------- */
/* printout help                                   */
/* ----------------------------------------------- */
#define PP fprintf(stdout,

static void PrintHelp()
{
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
			  PP      "------------------------------------------\n");
			  PP      " ecoPrimer Version %s\n", VERSION);
			  PP      "------------------------------------------\n");
              PP      "synopsis : finding primers and measureing the quality of primers and barcode region\n");
              PP      "usage: ./ecoPrimer [options] \n");
              PP      "------------------------------------------\n");
              PP      "options:\n");
              PP      "-d    : [D]atabase : to match the expected format, the database\n");
              PP      "        has to be formated first by the ecoPCRFormat.py program located.\n");
              PP      "        in the ecoPCR/tools directory.\n");
              PP      "        ecoPCRFormat.py creates three file types :\n");
              PP      "            .sdx : contains the sequences\n");
              PP      "            .tdx : contains information concerning the taxonomy\n");
              PP      "            .rdx : contains the taxonomy rank\n\n");
              PP      "        ecoPrimer needs all the file type. As a result, you have to write the\n");
Eric Coissac committed
57
              PP      "        database radical without any extension. For example /ecoPrimerDB/fstvert\n\n");
58 59
              PP      "-e    : [E]rror : max error allowed by oligonucleotide (0 by default)\n\n");
              PP      "-h    : [H]elp - print <this> help\n\n");
60
              PP      "-i    : [I]gnore the given taxonomy id (define the counterexample taxon set).\n\n");
61 62
              PP      "-l    : minimum [L]ength : define the minimum amplication length. \n\n");
              PP      "-L    : maximum [L]ength : define the maximum amplicationlength. \n\n");
63
              PP      "-r    : [R]estricts the search to the given taxonomic id (restrict the example taxon set).\n\n");
64
              PP      "-E    : [E]xception taxid allows to indicate than some subclade of example sequences are conterexamples.\n\n");
65
              PP      "-c    : Consider that the database sequences are [c]ircular\n\n");
66
              PP      "-3 	 : Three prime strict match\n\n");
67 68 69 70 71
              PP      "-q    : Strict matching [q]uorum, percentage of the sequences in which strict primers are found. By default it is 70\n\n");
              PP      "-s    : [S]ensitivity quorum\n\n");
              PP      "-t    : required [t]axon level for results, by default the results are computed at species level\n\n");
              PP      "-x    : false positive quorum\n\n");
              PP      "-D    : set in [d]ouble strand mode\n\n");
72
              PP      "-O    : set the primer length (default 18) \n\n");
73
              PP      "-S    : Set in [s]ingle strand mode\n\n");
74 75
              PP      "-m    : Salt correction method for Tm computation (SANTALUCIA : 1 or OWCZARZY:2, default=1)\n\n");
              PP      "-a    : Salt contentration in M for Tm computation (default 0.05 M)\n\n");
76
              PP      "-U    : No multi match\n\n");
77
              PP      "-R    : Define the [R]eference sequence identifier (must be part of example set)\n\n");
78 79 80
              PP      "-A    : Print the list of all identifier of sequences present in the database\n\n");
              PP      "-f    : Remove data mining step during  strict primer identification\n\n");
              PP      "-v    : Store statistic file about memory usage during strict primer identification\n\n");
81 82
              PP      "-p    : Print sets of primers (may take several minutes after primers have been designed!)\n\n");
              PP      "-T    : Ignore pairs having specificity below this Threshold\n\n");
83 84 85
              PP      "\n");
              PP      "------------------------------------------\n");
              PP      "Table result description : \n");
86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106
              PP      "column  1 : serial number\n");
              PP      "column  2 : primer1\n");
              PP      "column  3 : primer2\n");
              PP      "column  4 : primer1 Tm without mismatch\n");
              PP      "column  5 : primer1 lowest Tm against exemple sequences\n");
              PP      "column  6 : primer2 Tm without mismatch\n");
              PP      "column  7 : primer2 lowest Tm against exemple sequences\n");
              PP      "column  8 : primer1 G+C count\n");
              PP      "column  9 : primer2 G+C count\n");
              PP      "column 10 : good/bad\n");
              PP      "column 11 : amplified example sequence count\n");
              PP      "column 12 : amplified counterexample sequence count\n");
              PP      "column 13 : yule\n");
              PP      "column 14 : amplified example taxa count\n");
              PP      "column 15 : amplified counterexample taxa count\n");
              PP      "column 16 : ratio of amplified example taxa versus all example taxa (Bc index)\n");
              PP      "column 17 : unambiguously identified example taxa count\n");
              PP      "column 18 : ratio of specificity unambiguously identified example taxa versus all example taxa (Bs index)\n");
              PP      "column 19 : minimum amplified length\n");
              PP      "column 20 : maximum amplified length\n");
              PP      "column 21 : average amplified length\n");
107 108
              PP      "------------------------------------------\n");
              PP		" http://www.grenoble.prabi.fr/trac/ecoPrimer/\n");
Eric Coissac committed
109
              PP      "------------------------------------------\n\n");
110 111
              PP      "\n");

112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
}

static void ExitUsage(int stat)
{
        PP      "usage: ecoprimer [-d database] [-l value] [-L value] [-e value] [-r taxid] [-i taxid] [-R rank] [-t taxon level]\n");
        PP      "type \"ecoprimer -h\" for help\n");

        if (stat)
            exit(stat);
}

#undef  PP

void initoptions(poptions_t options)
{
127
	options->statistics=FALSE;
128
	options->filtering=TRUE;
129
	options->lmin=0;                   //< Amplifia minimal length
130
	options->lmax=1000;                   //< Amplifia maximal length
131 132 133 134
	options->error_max=3;              //**< maximum error count in fuzzy search
	options->primer_length=18;          //**< minimal length of the primers
	options->restricted_taxid=NULL;      //**< limit amplification below these taxid
	options->ignored_taxid=NULL;         //**< no amplification below these taxid
135
	options->exception_taxid=NULL;         //**< no amplification below these taxid
136
	options->prefix=NULL;
137 138
	options->reference=NULL;
	options->refseq=NULL;
139 140
	options->circular=0;
	options->doublestrand=1;
141
	options->strict_quorum=0.7;
142
	options->strict_exclude_quorum=0.1;
143
	options->sensitivity_quorum=0.9;
144
	options->false_positive_quorum=0.1;
145
	options->strict_three_prime=0;
146 147
	options->r=0;
	options->g=0;
148
	options->e=0;
149
	options->no_multi_match=FALSE;
150
	options->pnparm = NULL;
Eric Coissac committed
151
	strcpy(options->taxonrank, DEFAULTTAXONRANK);			/*taxon level for results, species by default*/
152
	options->saltmethod = SALT_METHOD_SANTALUCIA;
153
	options->salt = DEF_SALT;
154
	options->printAC=FALSE;
155
	options->print_sets_of_primers = FALSE;
156
	options->specificity_threshold = 0.6;
157 158 159
	options->links_cnt = 1;
	options->max_links_percent = -1;   /*graph only those primers having maximum 15% links*/
	options->filter_on_links = FALSE;
160 161
}

162 163
void printapair(int32_t index,ppair_t pair, poptions_t options)
{
Eric Coissac committed
164 165 166 167 168 169 170 171 172
	bool_t asdirect1=pair->asdirect1;
	bool_t asdirect2=pair->asdirect2;
	bool_t asdirecttmp;
	word_t w1=pair->p1->word;
	word_t w2=pair->p2->word;
	word_t wtmp;
	bool_t good1=pair->p1->good;
	bool_t good2=pair->p2->good;
	bool_t goodtmp;
173
	bool_t strand;
174
	uint32_t i, j;
175
	float temp;
176
	CNNParams nnparams;
177

178
	//nparam_InitParams(&nnparams, DEF_CONC_PRIMERS,DEF_CONC_SEQUENCES,DEF_SALT,SALT_METHOD_SANTALUCIA);
179 180

	char *c;
181 182
	char p1[32];
	char p2[32];
Eric Coissac committed
183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204

	if (!asdirect1)
		w1=ecoComplementWord(w1,options->primer_length);

	if (!asdirect2)
		w2=ecoComplementWord(w2,options->primer_length);


	if (w2 < w1)
	{
		wtmp=w1;
		w1=w2;
		w2=wtmp;

		asdirecttmp=asdirect1;
		asdirect1=asdirect2;
		asdirect2=asdirecttmp;

		goodtmp=good1;
		good1=good2;
		good2=goodtmp;
	}
Eric Coissac committed
205

206
	//print serial number
207 208
	printf("%6d\t",index);

209 210 211 212 213
	c = ecoUnhashWord(w1,options->primer_length);
	strcpy (p1, c);
	c = ecoUnhashWord(w2,options->primer_length);
	strcpy (p2, c);

214
	//print primer1
215
	printf("%s\t", p1);
216

217
	//print primer2
218
	printf("%s", p2);
219

220
	//print primer1 melting temperature
221
	printf ("\t%3.1f", pair->p1temp);
222

223
	//print minimum melting temperature of approximate versions of primer1
224 225
	printf ("\t%3.1f", pair->p1mintemp);

226
	//print primer2 melting temperature
227 228
	printf ("\t%3.1f", pair->p2temp);

229
	//print minimum melting temperature of approximate versions of primer2
230 231
	printf ("\t%3.1f", pair->p2mintemp);

232 233
	//print gc contents of primer1
	printf ("\t%d",nparam_CountGCContent(p1));
234

235 236
	//print gc contents of primer2
	printf ("\t%d",nparam_CountGCContent(p2));
237

238
	//print good/bad pair indicator
Eric Coissac committed
239
	printf("\t%c%c", "bG"[(int)good1],"bG"[(int)good2]);
240

241
	//print inexample count
242
	printf("\t%d", pair->inexample);
243

244
	//print out example count
245
	printf("\t%d", pair->outexample);
246

247
	//print yule
248 249
	printf("\t%4.3f", pair->yule);

250
	//print in taxa count
251
	printf("\t%d", pair->intaxa);
252

253
	//print out taxa count
254
	printf("\t%d", pair->outtaxa);
255

256
	//print coverage
Eric Coissac committed
257
	printf("\t%4.3f", (float)pair->bc);
258

259
	//print well identified taxa count
Eric Coissac committed
260
	printf("\t%d", pair->intaxa - pair->notwellidentifiedtaxa);
261

262
	//print specificity
Eric Coissac committed
263
	printf("\t%4.3f", pair->bs);
264

265
	//print min amplifia lenght
266
	printf("\t%d", pair->mind);
267

268
	//print max amplifia lenght
269
	printf("\t%d", pair->maxd);
270

271
	//print average amplifia lenght
272
	printf("\t%3.2f", (float)pair->sumd/pair->amplifiacount);
273

274
	//print amplifia information about reference sequence if specified
275 276 277
	if (options->refseq && pair->refsequence >=0)
	{
		printf("\t%s:",options->reference);
278
		strand = pair->pcr.amplifias[pair->refsequence].strand;
279

280
		if (strand)
281 282 283 284 285 286 287 288 289 290 291 292
			printf("join(");
		else
			printf("complement(");

		printf("%d..%d,%d..%d",pair->pcr.amplifias[pair->refsequence].begin - options->primer_length + 1,
				                 pair->pcr.amplifias[pair->refsequence].begin,
							  	 pair->pcr.amplifias[pair->refsequence].end + 2,
							  	 pair->pcr.amplifias[pair->refsequence].end + options->primer_length + 1
							  	 );
		printf(")");
		printf("\t");

293 294 295 296 297 298 299 300
		for (c=pair->pcr.amplifias[pair->refsequence].amplifia,
		     i=pair->pcr.amplifias[pair->refsequence].begin;
			 i<=pair->pcr.amplifias[pair->refsequence].end;
			 i++,
			 c+=(strand)? 1:-1)
			printf("%c","acgt"[(strand)? (*c):(~*c)&3]);


301
	}
302 303
	else
		printf("\t\t");
304

305 306 307 308 309 310
/*	j=0;
	for (i=0; i<options->dbsize; i++)
		if (pair->wellIdentifiedSeqs[i] == 1)
			j++;
	printf("%d", j);*/

311
	printf("\n");
312 313 314

}

Eric Coissac committed
315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333
static int cmpprintedpairs(const void* p1,const void* p2)
{
	float s1,s2;
	ppair_t pair1,pair2;

	pair1=*((ppair_t*)p1);
	pair2=*((ppair_t*)p2);

	s1 = pair1->yule * pair1->bs;
	s2 = pair2->yule * pair2->bs;

//	fprintf(stderr,"s1 : %4.3f  %4.3f   %4.3f\n",pair1->yule , pair1->bs,s1);
//	fprintf(stderr,"s2 : %4.3f  %4.3f   %4.3f\n\n",pair2->yule , pair2->bs,s2);

	if (s1 > s2) return -1;
	if (s1 < s2) return 1;
	return 0;
}

334
uint32_t filterandsortpairs(ppair_t* sortedpairs,uint32_t count, poptions_t options, pecodnadb_t seqdb)
335 336 337 338 339 340 341 342 343 344 345 346 347 348
{
	uint32_t i,j;
	float q,qfp;

	for (i=0,j=0;i < count;i++)
	{
		if (options->insamples)
			q = (float)sortedpairs[i]->inexample/options->insamples;
		else q=1.0;

		if (options->outsamples)
			qfp = (float)sortedpairs[i]->outexample/options->outsamples;
		else qfp=0.0;

349
		sortedpairs[i]->wellIdentifiedSeqs = NULL; //TR 05/09/10 - wellIdentified needed for primer sets
350
		sortedpairs[i]->coveredSeqs = NULL; //TR 05/09/10 - wellIdentified needed for primer sets
351 352
		sortedpairs[i]->quorumin = q;
		sortedpairs[i]->quorumout = qfp;
Eric Coissac committed
353
		sortedpairs[i]->yule = q - qfp;
354 355 356 357 358
		sortedpairs[j]=sortedpairs[i];

		if (q > options->sensitivity_quorum &&
			qfp < options->false_positive_quorum)
		{
359 360
			//TR 05/09/10 - wellIdentified needed for primer sets
			sortedpairs[j]->wellIdentifiedSeqs = ECOMALLOC(options->dbsize * sizeof(int),"Cannot allocate well_identified_array");
361 362
			sortedpairs[j]->coveredSeqs = ECOMALLOC(options->dbsize * sizeof(int),"Cannot allocate well_identified_array");
			(void)taxonomycoverage(sortedpairs[j],options, seqdb, options->dbsize);
363
			taxonomyspecificity(sortedpairs[j], seqdb, options->dbsize);
364 365 366 367
			//j++;
			//if specificity less than user provieded threshold (default 60%) then ignore this pair
			if (sortedpairs[j]->bs >= options->specificity_threshold)
				j++;
368
		}
Eric Coissac committed
369

370
	}
Eric Coissac committed
371
	qsort(sortedpairs,j,sizeof(ppair_t),cmpprintedpairs);
372 373 374
	return j;
}

Eric Coissac committed
375

376
void printpairs (ppairtree_t pairs, poptions_t options,ecotaxonomy_t *taxonomy, pecodnadb_t seqdb)
377 378 379 380 381
{
   ppair_t* sortedpairs;
   ppair_t* index;
   ppairlist_t pl;
   size_t i,j;
Eric Coissac committed
382
   size_t count;
Eric Coissac committed
383 384
   char *taxon[]={"taxon","taxa"};
   ecotx_t    *current_taxon;
385 386
   //pairset pair_sets;
   pairset *pset = NULL;
Eric Coissac committed
387 388

   //printf("Index\tPrimer1\tPrimer2\tGB\tInexampleCount\tOutexampleCount\tYule\tIntaxaCount\tOuttaxaCount\tCoverage\tSpecificity\tMinAmplifiedLength\tMaxAmplifiedLength\tAvgAmplifiedLength\n");
389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404

   fprintf(stderr,"Total pair count : %d\n",pairs->count);

   sortedpairs = ECOMALLOC(pairs->count*sizeof(ppair_t),"Cannot Allocate ordered pairs");
   index=sortedpairs;
   pl=pairs->first;
   j=0;
   while(pl->next)
   {
	   for (i=0;i<pl->paircount;i++,j++)
		   sortedpairs[j]=pl->pairs+i;
	   pl=pl->next;
   }

   for (i=0;i<pl->paircount;i++,j++)
	   sortedpairs[j]=pl->pairs+i;
405
   
406
   count=filterandsortpairs(sortedpairs,pairs->count,options, seqdb);
407
   getThermoProperties(sortedpairs, count, options);
408

Eric Coissac committed
409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465
   fprintf(stderr,"Total good pair count : %u\n",(uint32_t)count);

	printf("#\n");
	printf("# ecoPrimer version %s\n",VERSION);
	printf("# Rank level optimisation : %s\n", options->taxonrank);
	printf("# max error count by oligonucleotide : %d\n",options->error_max);
	printf("#\n");

	if (options->r)
	{
		printf("# Restricted to %s:\n",taxon[(options->r>1) ? 1:0]);
		for(i=0;i<(uint32_t)options->r;i++)
		{
			current_taxon=eco_findtaxonbytaxid(taxonomy,options->restricted_taxid[i]);
			printf("#     %d : %s (%s)\n", current_taxon->taxid,
										   current_taxon->name,
										   taxonomy->ranks->label[current_taxon->rank]
										 );
		}
		printf("#\n");
	}
	if (options->g)
	{
		printf("# Ignore  %s:\n",taxon[(options->g>1) ? 1:0]);
		for(i=0;i<(uint32_t)options->r;i++)
		{
			current_taxon=eco_findtaxonbytaxid(taxonomy,options->ignored_taxid[i]);
			printf("#     %d : %s (%s)\n", current_taxon->taxid,
										   current_taxon->name,
										   taxonomy->ranks->label[current_taxon->rank]
										 );
		}
		printf("#\n");
	}
	printf("# strict primer quorum  : %3.2f\n",options->strict_quorum);
	printf("# example quorum        : %3.2f\n",options->sensitivity_quorum);
	if (options->g + options->r)
		printf("# counterexample quorum : %3.2f\n",options->false_positive_quorum);

	printf("#\n");
	printf("# database : %s\n",options->prefix);
    printf("# Database is constituted of %5d examples        corresponding to %5d %s\n",options->insamples,
    		options->intaxa,options->taxonrank);
    printf("#                        and %5d counterexamples corresponding to %5d %s\n",options->outsamples,
    		options->outtaxa,options->taxonrank);
	printf("#\n");

	if (options->lmin && options->lmax)
		printf("# amplifiat length between [%d,%d] bp\n",options->lmin,options->lmax);
	else if (options->lmin)
		printf("# amplifiat length larger than %d bp\n",options->lmin);
	else if (options->lmax)
		printf("# amplifiat length smaller than %d bp\n",options->lmax);
	if (options->circular)
		printf("# DB sequences are considered as circular\n");
	else
		printf("# DB sequences are considered as linear\n");
466
	printf("# Pairs having specificity less than %0.2f will be ignored\n", options->specificity_threshold);
Eric Coissac committed
467
	printf("#\n");
468
	
Eric Coissac committed
469

470 471
   for (i=0;i < count;i++)
	   printapair(i,sortedpairs[i],options);
472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489

   if (options->filter_on_links)
   {
		fprintf (stderr, "Old size: %d, ", count);
		count = primers_changeSortedArray (&sortedpairs, count, options);
		//count = primers_filterWithGivenLinks (&sortedpairs, count, options);
		fprintf (stderr, "New size: %d\n", count);

		if (count == 0)
		{
			fprintf (stderr, "No pairs passed the links constraints.\n");
			printf ("No pairs passed the links constraints.\n");
			return;
		}

		for (i=0;i < count;i++)
			printapair(i,sortedpairs[i],options);
   }
490 491 492
   
   if (options->print_sets_of_primers == TRUE)
   {
493 494
	   /*pair_sets = build_primers_set (sortedpairs, count, seqdb, options);
	   printf("Results from Greedy Algorithm and some other possibilities:\n");
495
	   some_other_set_possibilities (&pair_sets, sortedpairs, count, seqdb, options);
496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514
	   printf("Results from simulated Anealing:\n");
	   sets_by_SimulatedAnealing (&pair_sets, sortedpairs, count, seqdb, options);
	   printf("Results from Tabu Search:\n");
	   sets_by_TabuSearch (&pair_sets, sortedpairs, count, seqdb, options);*/
	   //pset = sets_by_BruteForce (sortedpairs, count, seqdb, options);
	   //if (pset)
	   /*/{
		   printf("Results from simulated Anealing:\n");
		   sets_by_SimulatedAnealing (pset, sortedpairs, count, seqdb, options);
		   printf("Results from Tabu Search:\n");
		   sets_by_TabuSearch (pset, sortedpairs, count, seqdb, options);
		   
		   if (pset)
		   {
			   ECOFREE (pset->set_wellIdentifiedTaxa, "Could not free memory for pair set wi");
			   ECOFREE (pset, "Could not free memory for pair");
		   }
	   }*/
	   build_and_print_sets (sortedpairs, count, seqdb, options);
515
   }
516
   //primers_graph_graphviz (sortedpairs, count, options);
517 518 519
}


520 521
/*updateseqparams: This function counts the insample and outsample sequences
 *  and with each sequences adds a tag of the taxon to which the sequence beongs*/
522 523

void updateseqparams (pecodnadb_t seqdb, uint32_t seqdbsize, ecotaxonomy_t *taxonomy,
524 525 526 527 528 529 530
		poptions_t options, int32_t *insamples, int32_t *outsamples)
{
	uint32_t i;
	int32_t taxid;
	ecotx_t  *tmptaxon;

    for (i=0;i<seqdbsize;i++)
531
    {
532
    	seqdb[i]->isexample=isExampleTaxon(taxonomy,seqdb[i]->taxid,options);
533 534 535 536 537 538
    	if (seqdb[i]->isexample)
    		(*insamples)++;
    	else
    		(*outsamples)++;

    	taxid = taxonomy->taxons->taxon[seqdb[i]->taxid].taxid;
539
		tmptaxon = eco_findtaxonbytaxid(taxonomy, taxid);
540 541 542 543 544 545 546 547 548 549 550 551 552 553
		if (tmptaxon)
			tmptaxon = eco_findtaxonatrank(tmptaxon, options->taxonrankidx);
		if (tmptaxon)
			seqdb[i]->ranktaxonid = tmptaxon->taxid;
    }
}

void setresulttaxonrank (ecotaxonomy_t *taxonomy, poptions_t options)
{
	int32_t i;

    /*set taxon rank for which result is to be given*/
    for (i = 0; i < taxonomy->ranks->count; i++)
    {
554
    	if (strcmp(taxonomy->ranks->label[i], options->taxonrank) == 0)
555 556 557 558 559 560 561 562 563 564 565 566 567 568
    	{
    		options->taxonrankidx = i;
    		break;
    	}
    }

    if (i == taxonomy->ranks->count)
    {
    	fprintf(stderr,"\nUnknown taxon level: '%s'\n", options->taxonrank);
    	exit(0);
    }
}
/* to get db stats, totals of species, genus etc....*/

569 570 571 572 573
static void printAC(pecodnadb_t seqdb,uint32_t seqdbsize)
{
	uint32_t i;

	for (i=0; i< seqdbsize;i++)
574
		printf("%15s (%8d bp ): %s\n",seqdb[i]->AC,seqdb[i]->SQ_length,seqdb[i]->DE);
575
}
576

577 578 579 580 581 582 583 584 585 586 587 588 589
int main(int argc, char **argv)
{
	pecodnadb_t   seqdb; /* of type ecoseq_t */
	uint32_t        seqdbsize=0;
	ecotaxonomy_t *taxonomy;

	options_t     options;
	int           carg;
	int32_t       errflag=0;

	int32_t       insamples=0;
	int32_t       outsamples=0;
	uint32_t        i;
590 591

	pwordcount_t    words;
592
//	pwordcount_t    words2;
593 594 595
	pprimercount_t  primers;
	ppairtree_t		pairs;

596 597
	int32_t		  rankdbstats = 0;

598
	CNNParams nnparams;
599

600 601
	initoptions(&options);

602
    while ((carg = getopt(argc, argv, "hAfvcUDSpbE:d:l:L:e:i:r:R:q:3:s:x:t:O:m:a:T:k:M:")) != -1) {
603 604

     switch (carg) {
605 606 607 608 609 610
								 /* ---------------------------- */
		 case 'v':               /* set in single strand mode    */
								 /* ---------------------------- */
			options.statistics=TRUE;
			break;

611 612 613 614
			 /* ---------------------------- */
		 case 'f':               /* set in single strand mode    */
			 /* ---------------------------- */
			 options.filtering=FALSE;
615 616 617 618 619 620
			 break;

			 /* ---------------------------- */
		 case 'A':               /* set in single strand mode    */
			 /* ---------------------------- */
			 options.printAC=TRUE;
621 622
			 break;

623 624 625 626 627 628 629 630 631 632 633
                                /* -------------------- */
        case 'd':               /* database name        */
                                /* -------------------- */
        	options.prefix = ECOMALLOC(strlen(optarg)+1,
                              "Error on prefix allocation");
           strcpy(options.prefix,optarg);
           break;

                                /* -------------------- */
        case 'h':               /* help                 */
                                /* -------------------- */
634 635 636
        	PrintHelp();
        	 exit(0);
        	 break;
637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655

                                /* ------------------------- */
        case 'l':               /* min amplification lenght  */
                                /* ------------------------- */
           sscanf(optarg,"%d",&(options.lmin));
           break;

                                /* -------------------------- */
        case 'L':               /* max amplification lenght   */
                                /* -------------------------- */
          sscanf(optarg,"%d",&(options.lmax));
          break;

                                /* -------------------- */
        case 'e':               /* error max            */
                                /* -------------------- */
          sscanf(optarg,"%d",&(options.error_max));
          break;

656 657 658 659 660 661

								/* ------------------------ */
		case '3':               /* three prime strict match */
                                /* ------------------------ */
			sscanf(optarg,"%d",&(options.strict_three_prime));
			break;
662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680

                                /* -------------------- */
        case 'q':               /* strict matching quorum           */
                                /* -------------------- */
          sscanf(optarg,"%f",&(options.strict_quorum));
          break;

								/* -------------------- */
        case 's':               /* strict matching quorum      */
								/* -------------------- */
        	sscanf(optarg,"%f",&(options.sensitivity_quorum));
        	break;

        						/* -------------------- */
        case 't':               /* required taxon level for results           */
        						/* -------------------- */
        	strncpy(options.taxonrank, optarg, 19);
        	options.taxonrank[19] = 0;
            break;
681

682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712
        						/* -------------------- */
        case 'x':               /* strict matching quorum           */
								/* -------------------- */
        	sscanf(optarg,"%f",&(options.false_positive_quorum));
        	break;

							    /* ---------------------------- */
		case 'D':               /* set in double strand mode    */
							    /* ---------------------------- */
			options.doublestrand=1;
			break;

								/* ---------------------------- */
		case 'S':               /* set in single strand mode    */
								/* ---------------------------- */
			options.doublestrand=0;
			break;

			                    /* ---------------------------- */
        case 'U':               /* set in single strand mode    */
			                    /* ---------------------------- */
            options.no_multi_match=TRUE;
            break;

         						/* ------------------------------------------ */
        case 'r':               /* stores the restricting search taxonomic id */
        						/* ------------------------------------------ */
          options.restricted_taxid = ECOREALLOC(options.restricted_taxid,sizeof(int32_t)*(options.r+1),
          									"Error on restricted_taxid reallocation");
          sscanf(optarg,"%d",&(options.restricted_taxid[options.r]));
          options.r++;
713 714
          break;

715 716 717 718 719 720 721 722 723
								/* ------------------------------------------ */
        case 'E':               /* stores the restricting search taxonomic id */
								/* ------------------------------------------ */
        	options.exception_taxid = ECOREALLOC(options.exception_taxid,sizeof(int32_t)*(options.e+1),
						            "Error on exception_taxid reallocation");
			sscanf(optarg,"%d",&(options.exception_taxid[options.e]));
			options.e++;
			break;

724 725 726 727 728 729
								/* -------------------- */
        case 'R':               /* reference sequence   */
                                /* -------------------- */
          options.reference = ECOMALLOC(strlen(optarg)+1,
                                     "Error on prefix allocation");
          strcpy(options.reference,optarg);
730 731 732 733 734 735 736 737 738 739 740
          break;

         						/* --------------------------------- */
        case 'i':               /* stores the taxonomic id to ignore */
        						/* --------------------------------- */
          options.ignored_taxid = ECOREALLOC(options.ignored_taxid,sizeof(int32_t)*(options.g+1),
          									"Error on excluded_taxid reallocation");
          sscanf(optarg,"%d",&(options.ignored_taxid[options.g]));
          options.g++;
          break;

Eric Coissac committed
741 742 743 744 745 746
			                    /* --------------------------------- */
        case 'O':               /* set primer size                   */
        						/* --------------------------------- */
			sscanf(optarg,"%d",&(options.primer_length));
			break;

747 748 749 750 751 752 753 754 755 756 757 758 759
								/* --------------------------------- */
        case 'm':               /* set salt method                   */
								/* --------------------------------- */
        	sscanf(optarg,"%d",&(options.saltmethod));
        	break;

        						/* --------------------------------- */
        case 'a':               /* set salt 	                     */
								/* --------------------------------- */
        	sscanf(optarg,"%f",&(options.salt));
        	break;

        						/* -------------------- */
760 761 762 763 764
        case 'c':               /* sequences are circular */
        						/* --------------------------------- */
		  options.circular = 1;
          break;

765 766 767
								/* -------------------- */
        case 'p':               /* print sets of primers */
								/* --------------------------------- */
768
        	//options.print_sets_of_primers = TRUE;
769
        break;
770 771 772 773 774 775 776 777 778 779 780 781 782
		
        						/* --------------------------------- */
        case 'T':               /* Ignore pairs having specificity below this Threshold     */
								/* --------------------------------- */
        	sscanf(optarg,"%f",&(options.specificity_threshold));
        	break;

								/* --------------------------------- */
		case 'M':               /* Max link percentage for graph     */
								/* --------------------------------- */
			sscanf(optarg,"%f",&(options.max_links_percent));
		break;

783
								/* --------------------------------- */
784
		case 'k':               /* links count                   */
785
								/* --------------------------------- */
786 787 788 789 790
			sscanf(optarg,"%d",&(options.links_cnt));
		break;

		case 'b':
			options.filter_on_links = TRUE;
791
		break;
792

793 794 795 796 797 798
        case '?':               /* bad option           */
                                /* -------------------- */
            errflag++;
   		}

	}
799
    options.pnparm = &nnparams;
800 801
    if (options.saltmethod != 2) //if not SALT_METHOD_OWCZARZY
    	options.saltmethod = SALT_METHOD_SANTALUCIA; //then force SALT_METHOD_SANTALUCIA
802

803 804
    if (options.salt < 0.01 || options.salt > 0.3) //if salt value out of literature values
    	options.salt = DEF_SALT; //set to default
805

806 807
	nparam_InitParams(&nnparams, DEF_CONC_PRIMERS,DEF_CONC_SEQUENCES,options.salt,options.saltmethod);

808 809 810
    fprintf(stderr,"Reading taxonomy database ...");
    taxonomy = read_taxonomy(options.prefix,0);
    fprintf(stderr,"Ok\n");
811

812
    setresulttaxonrank(taxonomy, &options); /*TR: set rank level for statistics*/
813

814 815
    fprintf(stderr,"Reading sequence database ...\n");

816
    seqdb = readdnadb(options.prefix,taxonomy,&seqdbsize, &options);
817

818 819 820 821 822
	if (options.printAC)
	{
		printAC(seqdb,seqdbsize);
		exit(0);
	}
823 824 825 826 827 828 829 830 831
    if (options.reference)
    	for (i=0; i < seqdbsize;i++)
    		if (strcmp(seqdb[i]->AC,options.reference)==0)
    		{
    			options.refseq=seqdb[i];
    			options.refseqid=i;
    			fprintf(stderr,"Reference sequence %s identified\n",options.reference);
    		}

832 833
    fprintf(stderr,"Ok\n");
    fprintf(stderr,"Sequence read : %d\n",(int32_t)seqdbsize);
834

835
    updateseqparams(seqdb, seqdbsize, taxonomy, &options, &insamples , &outsamples);
836 837 838
    options.dbsize=seqdbsize;
    options.insamples=insamples;
    options.outsamples=outsamples;
839 840 841

    rankdbstats = getrankdbstats(seqdb, seqdbsize, taxonomy, &options);

842 843 844 845
    fprintf(stderr,"Database is constituted of %5d examples        corresponding to %5d %s\n",insamples,
    		options.intaxa,options.taxonrank);
    fprintf(stderr,"                       and %5d counterexamples corresponding to %5d %s\n",outsamples,
    		options.outtaxa,options.taxonrank);
846 847 848 849 850 851
    fprintf(stderr,"Total distinct %s count %d\n",options.taxonrank, rankdbstats);

    fprintf(stderr,"\nIndexing words in sequences\n");

    words = lookforStrictPrimer(seqdb,seqdbsize,insamples,&options);
    fprintf(stderr,"\n  Strict primer count : %d\n",words->size);
852 853 854 855 856
    
    /*/TR Testing
    fprintf(stderr,"\nReducing for debugging\n");
    words = reduce_words_to_debug (words, &options);
    ///*/
857 858 859 860 861 862 863 864 865 866 867
//    options.filtering=FALSE;
//    words2= lookforStrictPrimer(seqdb,seqdbsize,insamples,&options);
//    fprintf(stderr,"\n  Strict primer count : %d\n",words2->size);
//
//    fprintf(stderr,"\n\n  Primer sample : \n");
//    for (i=0; i<words->size; i++)
//    	fprintf(stderr,"  + Primer : %s   sequence count : %d\n",ecoUnhashWord(words->words[i],options.primer_length),words->strictcount[i]);
//    fprintf(stderr,"\n\n  Primer sample : \n");
//    for (i=0; i<words2->size; i++)
//    	fprintf(stderr,"  + Primer : %s   sequence count : %d\n",ecoUnhashWord(words2->words[i],options.primer_length),words2->strictcount[i]);

868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887
    if (options.no_multi_match)
    {
    	(void)filterMultiStrictPrimer(words);
    	fprintf(stderr,"\n  Strict primer with single match count : %d\n",words->size);
    }


    fprintf(stderr,"\n\n  Primer sample : \n");
    for (i=0; i<MINI(10,words->size); i++)
    	fprintf(stderr,"  + Primer : %s   sequence count : %d\n",ecoUnhashWord(words->words[i],options.primer_length),words->strictcount[i]);

    fprintf(stderr,"\nEncoding sequences for fuzzy pattern matching...\n");
    for (i=0;i<seqdbsize;i++)
    {
    	encodeSequence(seqdb[i]);
    	fprintf(stderr,"  Encoded sequences %5d/%5d          \r",(int32_t)i+1,(int32_t)seqdbsize);
    }

    ECOFREE(words->strictcount,"Free strict primer count table");

888 889 890 891 892 893 894
    if (options.error_max == 0)//aho, if(options.error_max == 0 && 0) old
    	primers = ahoc_lookforStrictPrimers (seqdb,seqdbsize,insamples,words,&options);
    else
    	primers = lookforAproxPrimer(seqdb,seqdbsize,insamples,words,&options);
    
    //for (i=0; i<primers->size; i++)
    //	print_wordwith_positions (primers->primers[i], seqdbsize, &options);
895 896 897 898 899 900 901 902 903 904 905 906 907 908

    ECOFREE(words->words,"Free strict primer table");
    ECOFREE(words,"Free strict primer structure");
    fprintf(stderr,"\n\n  Approximate repeats :%d \n", primers->size);

    fprintf(stderr,"\n\n  Primer sample : \n");
    for (i=0; i<MINI(10,primers->size); i++)
    	fprintf(stderr,"  + Primer : %s   example sequence count : %5d counterexample sequence count : %5d  status : %s\n",ecoUnhashWord(primers->primers[i].word,options.primer_length),
                       primers->primers[i].inexample,
                       primers->primers[i].outexample,
                       primers->primers[i].good ? "good":"bad");

    fprintf(stderr,"\n");

909 910 911 912
   
    	pairs = buildPrimerPairs(seqdb, seqdbsize, primers, &options);
    	printpairs (pairs, &options,taxonomy, seqdb);
    
913 914
    return 0;
}
915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019

#define DEBUG_WORDS_CNT 14
pwordcount_t reduce_words_to_debug (pwordcount_t words, poptions_t options)
{
	uint32_t i, k;
	pwordcount_t new_words;
	char *rwrd;
	char dwrd[20];
	/*char *strict_words[DEBUG_WORDS_CNT] = {"GAGTCTCTGCACCTATCC", "GCAATCCTGAGCCAAATC", "ACCCCTAACCACAACTCA", 
			"TCCGAACCGACTGATGTT", "GAAGCTTGGGTGAAACTA", "GGAGAACCAGCTAGCTCT", "GCTGGTTCTCCCCGAAAT",
			"TCGATTTGGTACCGCTCT", "AAAGGAGAGAGAGGGATT", "GGATTGCTAATCCGTTGT", "CCCCCATCGTCTCACTGG",
			"TGAGGCGCAGCAGTTGAC", "GCGCTACGGCGCTGAAGT", "TTTCCTGGGAGTATGGCA"};*/
	char *strict_words[DEBUG_WORDS_CNT] = {"CTCCGGTCTGAACTCAGA", "TGTTGGATCAGGACATCC", "TAGATAGAAACCGACCTG", 
			"TGGTGCAGCCGCTATTAA", "AGATAGAAACTGACCTGG", "TGGTGCAGCCGCTATTAA", "CTAATGGTGCAGCCGCTA",
			"TAGAAACTGACCTGGATT", "AGATAGAAACCGACCTGG", "ATGGTGCAGCCGCTATTA", "ATAGATAGAAACCGACCT",
			"GCCGCTATTAAGGGTTCG", "GGTGCAGCCGCTATTAAG", "TAGAAACTGACCTGGATT"};
	int word_seen[DEBUG_WORDS_CNT];
	                   
	
	new_words = ECOMALLOC(sizeof(wordcount_t),"Cannot allocate memory for word count structure");
	new_words->inseqcount = words->inseqcount;
	new_words->outseqcount = words->outseqcount;
	new_words->size = DEBUG_WORDS_CNT;
	new_words->strictcount = ECOMALLOC((new_words->size*sizeof(uint32_t)), "Cannot allocate memory for word count table");
	new_words->words = ECOMALLOC(new_words->size*sizeof(word_t), "I cannot allocate memory for debug words");
			
	for (k = 0; k < DEBUG_WORDS_CNT; k++)
		word_seen[k] = 0;
	
	for (i=0; i < words->size; i++)
	{
		rwrd = ecoUnhashWord(words->words[i],options->primer_length);
		strcpy (dwrd, rwrd);
		rwrd = ecoUnhashWord(ecoComplementWord(words->words[i],options->primer_length),options->primer_length);
		for (k = 0; k < DEBUG_WORDS_CNT; k++)
		{
			if (strcmp (dwrd, strict_words[k]) == 0) break;
			if (strcmp (rwrd, strict_words[k]) == 0) break;
		}
		
		if (k < DEBUG_WORDS_CNT)
		{
			if (word_seen[k] == 0)
			{
				new_words->words[k] = words->words[i];
				new_words->strictcount[k] = words->strictcount[i];
			}
			word_seen[k]++;
		}
	}
	
	fprintf (stderr, "Debug Words Info:\n");
	for (k = 0; k < DEBUG_WORDS_CNT; k++)
		fprintf (stderr, "%s:%d\n", strict_words[k], word_seen[k]);
	
	
	//clean input wods;
	ECOFREE(words->words,"Clean  word table");
	ECOFREE(words->strictcount,"Clean  word count table");
	ECOFREE(words,"Clean word structure");
	
	return new_words;
}

void print_wordwith_positions (primer_t prm, uint32_t seqdbsize, poptions_t options)
{
	char *wrd;
	uint32_t i, j;
	char *twrd = "GCCTGTTTACCAAAAACA";

	wrd = ecoUnhashWord(prm.word,options->primer_length);

	if (strcmp (twrd, wrd) == 0)
	{
		printf ("Positions for Word: %s\n", wrd);
		for (i=0; i<seqdbsize; i++)
		{
			if (prm.directCount[i] > 0)
			{
				printf ("%d:", i);
				if (prm.directCount[i] == 1)
					printf ("%d", prm.directPos[i].value);
				else
					for (j=0; j<prm.directCount[i]; j++)
						printf ("%d,", prm.directPos[i].pointer[j]);
				printf (" ");
			}
		}
		printf ("\n");
		for (i=0; i<seqdbsize; i++)
		{
			if (prm.reverseCount[i] > 0)
			{
				printf ("%d:", i);
				if (prm.reverseCount[i] == 1)
					printf ("%d", prm.reversePos[i].value);
				else
					for (j=0; j<prm.reverseCount[i]; j++)
						printf ("%d,", prm.reversePos[i].pointer[j]);
				printf (" ");
			}
		}
		printf ("\n");
	}
}