static void runscanatpostrial(const Encodedsequence *encseq,
Encodedsequencescanstate *esr,
Readmode readmode,Seqpos startpos)
{
Seqpos pos, totallength;
GtUchar ccra, ccsr;
totallength = getencseqtotallength(encseq);
initEncodedsequencescanstate(esr,encseq,readmode,startpos);
for (pos=startpos; pos < totallength; pos++)
{
ccra = getencodedchar(encseq,pos,readmode); /* Random access */
ccsr = sequentialgetencodedchar(encseq,esr,pos,readmode);
if (ccra != ccsr)
{
fprintf(stderr,"startpos = " FormatSeqpos
" access=%s, mode=%s: position=" FormatSeqpos
": random access (correct) = %u != %u = "
" sequential read (wrong)\n",
PRINTSeqposcast(startpos),
encseqaccessname(encseq),
showreadmode(readmode),
PRINTSeqposcast(pos),
(unsigned int) ccra,
(unsigned int) ccsr);
exit(GT_EXIT_PROGRAMMING_ERROR);
}
}
}
static void showmatch(void *processinfo,const GtMatch *match)
{
Showmatchinfo *showmatchinfo = (Showmatchinfo *) processinfo;
unsigned long seqnum;
Seqpos relpos;
if (match->dbabsolute)
{
Seqinfo seqinfo;
seqnum = getencseqfrompos2seqnum(showmatchinfo->encseq,match->dbstartpos);
getencseqSeqinfo(&seqinfo,showmatchinfo->encseq,seqnum);
gt_assert(seqinfo.seqstartpos <= match->dbstartpos);
relpos = match->dbstartpos - seqinfo.seqstartpos;
} else
{
relpos = match->dbstartpos;
seqnum = match->dbseqnum;
}
printf("%lu\t" FormatSeqpos "\t",seqnum,PRINTSeqposcast(relpos));
printf(FormatSeqpos "\t",PRINTSeqposcast(match->dblen));
printf("\t" Formatuint64_t "\t%lu\t%lu\t%lu\n",
PRINTuint64_tcast(showmatchinfo->queryunit),
match->querystartpos,
match->querylen,
match->distance);
if (showmatchinfo->showalignment)
{
gt_alignment_show_with_mapped_chars(
(const GtAlignment *) match->alignment,
showmatchinfo->characters,
showmatchinfo->wildcardshow,
stdout);
}
}
static void produceshortoutput(const LTRboundaries *boundaries,Seqpos offset)
{
/* increase positions by 1 */
printf(FormatSeqpos " ",
PRINTSeqposcast( boundaries->leftLTR_5 -offset + 1));
printf(FormatSeqpos " ",
PRINTSeqposcast( boundaries->rightLTR_3 -offset + 1));
printf(FormatSeqpos " ",
PRINTSeqposcast( (boundaries->rightLTR_3
- boundaries->leftLTR_5 + 1)));
printf(FormatSeqpos " ",
PRINTSeqposcast( boundaries->leftLTR_5 -offset + 1));
printf(FormatSeqpos " ",
PRINTSeqposcast( boundaries->leftLTR_3 -offset + 1));
printf(FormatSeqpos " ",
PRINTSeqposcast( (boundaries->leftLTR_3
- boundaries->leftLTR_5 + 1)));
printf(FormatSeqpos " ",
PRINTSeqposcast( boundaries->rightLTR_5 -offset + 1));
printf(FormatSeqpos " ",
PRINTSeqposcast( boundaries->rightLTR_3 -offset + 1));
printf(FormatSeqpos " ",
PRINTSeqposcast( (boundaries->rightLTR_3
- boundaries->rightLTR_5 + 1)));
/* print similarity */
printf("%.2f ", boundaries->similarity);
/* print sequence number */
printf("%lu\n", boundaries->contignumber);
}
static void makeerrormsg(const Sequencerange *vala,const Sequencerange *valb,
const char *cmpflag)
{
fprintf(stderr,
"(" FormatSeqpos "," FormatSeqpos
") %s (" FormatSeqpos "," FormatSeqpos
")\n",
PRINTSeqposcast(vala->leftpos),
PRINTSeqposcast(vala->rightpos),
cmpflag,
PRINTSeqposcast(valb->leftpos),
PRINTSeqposcast(valb->rightpos));
}
static void showbucketspec2(const GtBucketspec2 *bucketspec2)
{
unsigned int idx1, idx2;
for (idx1 = 0; idx1 < bucketspec2->numofchars; idx1++)
{
for (idx2 = 0; idx2 < bucketspec2->numofchars; idx2++)
{
printf("subbucket[%u][%u]=" FormatSeqpos,idx1,idx2,
PRINTSeqposcast(bucketspec2->subbuckettab[idx1][idx2].bucketend));
if (bucketspec2->subbuckettab[idx1][idx2].sorted)
{
printf(" sorted\n");
} else
{
printf("\n");
}
}
printf("superbucket[%u]=" FormatSeqpos "\n",idx1,
PRINTSeqposcast(bucketspec2->superbuckettab[idx1].bucketend));
}
}
static void testscanatpos(const Encodedsequence *encseq,
Readmode readmode,
unsigned long scantrials)
{
Encodedsequencescanstate *esr = NULL;
Seqpos startpos, totallength;
unsigned long trial;
totallength = getencseqtotallength(encseq);
srand48(42349421);
esr = newEncodedsequencescanstate();
runscanatpostrial(encseq,esr,readmode,0);
runscanatpostrial(encseq,esr,readmode,totallength-1);
for (trial = 0; trial < scantrials; trial++)
{
startpos = (Seqpos) (drand48() * (double) totallength);
printf("trial %lu at " FormatSeqpos "\n",trial,PRINTSeqposcast(startpos));
runscanatpostrial(encseq,esr,readmode,startpos);
}
freeEncodedsequencescanstate(&esr);
}
void gt_copysortsuffixes(const GtBucketspec2 *bucketspec2,
Seqpos *suftab,
Verboseinfo *verboseinfo)
{
Seqpos hardwork = 0, **targetptr;
unsigned int idx, idxsource, source, second;
#ifdef WITHSUFFIXES
{
const Seqpos *ptr;
for (ptr = suftab; ptr < suftab + bucketspec2->partwidth; ptr++)
{
showsequenceatstartpos(stdout,
ISDIRREVERSE(readmode) ? false : true,
ISDIRCOMPLEMENT(readmode) ? true : false,
encseq,
*ptr);
}
}
#endif
targetptr = gt_malloc(sizeof(*targetptr) * bucketspec2->numofchars);
for (idxsource = 0; idxsource<bucketspec2->numofchars; idxsource++)
{
source = bucketspec2->order[idxsource];
for (second = 0; second < bucketspec2->numofchars; second++)
{
if (!bucketspec2->subbuckettab[source][second].sorted && source != second)
{
gt_assert(bucketspec2->subbuckettab[source][second].hardworktodo);
showverbose(verboseinfo,"hard work for %u %u",source,second);
hardwork += getendidx(bucketspec2,source,second) -
getstartidx(bucketspec2,source,second);
bucketspec2->subbuckettab[source][second].sorted = true;
} else
{
gt_assert(!bucketspec2->subbuckettab[source][second].hardworktodo);
}
}
if (getstartidx(bucketspec2,source,0) <
getstartidx(bucketspec2,source,source))
{
for (idx = 0; idx < bucketspec2->numofchars; idx++)
{
targetptr[idx] = suftab + getstartidx(bucketspec2,idx,source);
}
forwardderive(bucketspec2,
targetptr,
source,
suftab + getstartidx(bucketspec2,source,0));
}
if (getendidx(bucketspec2,source,source) <
getendidx(bucketspec2,source,bucketspec2->numofchars))
{
for (idx = 0; idx < bucketspec2->numofchars; idx++)
{
targetptr[idx] = suftab + getendidx(bucketspec2,idx,source) - 1;
}
backwardderive(bucketspec2,
targetptr,
source,
suftab +
getendidx(bucketspec2,source,bucketspec2->numofchars) - 1);
}
for (idx = 0; idx < bucketspec2->numofchars; idx++)
{
bucketspec2->subbuckettab[idx][source].sorted = true;
}
bucketspec2->superbuckettab[source].sorted = true;
}
gt_free(targetptr);
showverbose(verboseinfo,"hardwork = " FormatSeqpos " (%.2f)",
PRINTSeqposcast(hardwork),
(double) hardwork/getencseqtotallength(bucketspec2->encseq));
}
static void producelongutput(const LTRharvestoptions *lo,
const LTRboundaries *boundaries,
const Encodedsequence *encseq,
Seqpos offset)
{
const GtUchar *characters = getencseqAlphabetcharacters(encseq);
printf(FormatSeqpos " ",
PRINTSeqposcast(boundaries->leftLTR_5 -offset + 1));
printf(FormatSeqpos " ",
PRINTSeqposcast(boundaries->rightLTR_3 -offset + 1));
printf(FormatSeqpos " ",
PRINTSeqposcast((boundaries->rightLTR_3 - boundaries->leftLTR_5
+ 1)));
printf(FormatSeqpos " ",
PRINTSeqposcast(boundaries->leftLTR_5 -offset + 1));
printf(FormatSeqpos " ",
PRINTSeqposcast(boundaries->leftLTR_3 -offset + 1));
printf(FormatSeqpos " ",
PRINTSeqposcast((boundaries->leftLTR_3 - boundaries->leftLTR_5
+ 1)));
if (lo->minlengthTSD > 1U)
{
Seqpos j;
for (j = 0; j < boundaries->lenleftTSD; j++)
{
printf("%c",(char) characters[getencodedchar(encseq,
boundaries->leftLTR_5 -
boundaries->lenleftTSD + j,
Forwardmode)]);
}
printf(" " FormatSeqpos " ",
PRINTSeqposcast(boundaries->lenleftTSD));
}
if (lo->motif.allowedmismatches < 4U)
{
printf("%c%c..%c%c ",
(char) characters[getencodedchar(encseq,/* Random access */
boundaries->leftLTR_5,
Forwardmode)],
(char) characters[getencodedchar(encseq,/* Random access */
boundaries->leftLTR_5+1,
Forwardmode)],
(char) characters[getencodedchar(encseq,/* Random access */
boundaries->leftLTR_3-1,
Forwardmode)],
(char) characters[getencodedchar(encseq,/* Random access */
boundaries->leftLTR_3,
Forwardmode)] );
}
/* increase by 1 */
printf(FormatSeqpos " ",
PRINTSeqposcast(boundaries->rightLTR_5 -offset + 1));
/* increase by 1 */
printf(FormatSeqpos " ",PRINTSeqposcast(boundaries->rightLTR_3 -offset + 1));
printf(FormatSeqpos " ",PRINTSeqposcast(boundaries->rightLTR_3
- boundaries->rightLTR_5 + 1));
if (lo->minlengthTSD > 1U)
{
Seqpos j;
for (j = 0; j < boundaries->lenrightTSD; j++)
{
printf("%c", (char) characters[getencodedchar(encseq,
boundaries->rightLTR_3+j+1,
Forwardmode)]);
}
printf(" " FormatSeqpos " ",PRINTSeqposcast(boundaries->lenrightTSD));
}
if (lo->motif.allowedmismatches < 4U)
{
printf("%c%c..%c%c",
(char) characters[getencodedchar(encseq,/* Randomaccess */
boundaries->rightLTR_5,
Forwardmode)],
(char) characters[getencodedchar(encseq,/* Randomaccess */
boundaries->rightLTR_5+1,
Forwardmode)],
(char) characters[getencodedchar(encseq,/* Randomaccess */
boundaries->rightLTR_3-1,
Forwardmode)],
(char) characters[getencodedchar(encseq,/* Random access */
boundaries->rightLTR_3,/* Randomaccess */
Forwardmode)] );
}
/* print similarity */
printf(" %.2f", boundaries->similarity);
/* print sequence number */
printf(" %lu\n", boundaries->contignumber);
}
/*
The following function shows all options that are set by default or from
the user on stdout.
*/
void showuserdefinedoptionsandvalues(const LTRharvestoptions *lo)
{
printf("# user defined options and values:\n");
if (lo->verbosemode)
{
printf("# verbosemode: On\n");
}
else
{
printf("# verbosemode: Off\n");
}
printf("# indexname: %s\n", gt_str_get(lo->str_indexname));
if (lo->fastaoutput)
{
printf("# outputfile: %s\n", gt_str_get(lo->str_fastaoutputfilename));
}
if (lo->fastaoutputinnerregion)
{
printf("# outputfile inner region: %s\n",
gt_str_get(lo->str_fastaoutputfilenameinnerregion));
}
if (lo->gff3output)
{
printf("# outputfile gff3 format: %s\n",
gt_str_get(lo->str_gff3filename));
}
printf("# xdropbelowscore: %d\n", lo->xdropbelowscore);
printf("# similaritythreshold: %.2f\n", lo->similaritythreshold);
printf("# minseedlength: %lu\n", lo->minseedlength);
printf("# matchscore: %d\n", lo->arbitscores.mat);
printf("# mismatchscore: %d\n", lo->arbitscores.mis);
printf("# insertionscore: %d\n", lo->arbitscores.ins);
printf("# deletionscore: %d\n", lo->arbitscores.del);
printf("# minLTRlength: %lu\n", lo->repeatinfo.lmin);
printf("# maxLTRlength: %lu\n", lo->repeatinfo.lmax);
printf("# minLTRdistance: %lu\n", lo->repeatinfo.dmin);
printf("# maxLTRdistance: %lu\n", lo->repeatinfo.dmax);
if (lo->nooverlapallowed)
{
printf("# overlaps: no\n");
}
else
{
if (lo->bestofoverlap)
{
printf("# overlaps: best\n");
}
else
{
printf("# overlaps: all\n");
}
}
printf("# minTSDlength: %u\n", lo->minlengthTSD);
printf("# maxTSDlength: %u\n", lo->maxlengthTSD);
printf("# palindromic motif: %s\n", gt_str_get(lo->motif.str_motif));
printf("# motifmismatchesallowed: %u\n", lo->motif.allowedmismatches);
printf("# vicinity: " FormatSeqpos " nt\n",
PRINTSeqposcast(lo->vicinityforcorrectboundaries));
if (lo->repeatinfo.ltrsearchseqrange.start != 0 ||
lo->repeatinfo.ltrsearchseqrange.end != 0)
{
printf("# ltrsearchseqrange=(%lu,%lu)\n",
PRINTSeqposcast(lo->repeatinfo.ltrsearchseqrange.start),
PRINTSeqposcast(lo->repeatinfo.ltrsearchseqrange.end));
}
}
