static void tranalign_AddGaps(AjPSeq newseq,
const AjPSeq nseq, const AjPSeq pseq,
ajlong npos)
{
AjPStr newstr = NULL;
ajuint ppos = 0;
newstr = ajStrNew();
for(; ppos<ajSeqGetLen(pseq); ppos++)
if(ajSeqGetSeqC(pseq)[ppos] == '-')
ajStrAppendC(&newstr, "---");
else
{
ajStrAppendSubS(&newstr, ajSeqGetSeqS(nseq), npos, npos+2);
npos+=3;
}
ajDebug("aligned seq=%S\n", newstr);
ajSeqAssignSeqS(newseq, newstr);
ajStrDel(&newstr);
return;
}
AjPStr embPropProt1to3(AjPSeq seq, ajint pad)
{
const char *p;
const char *p3;
AjPStr temp;
ajint i;
temp = ajStrNewRes(ajSeqGetLen(seq)*3 + pad+1);
/* put any required padding spaces at the start */
for(i=0; i<pad; i++)
ajStrAppendC(&temp, " ");
for(p=ajSeqGetSeqC(seq); *p; p++)
{
if(*p == '*')
ajStrAppendC(&temp, "***");
else if(*p == '.')
ajStrAppendC(&temp, "...");
else if(*p == '-')
ajStrAppendC(&temp, "---");
else if(!isalpha((ajint)*p))
ajStrAppendC(&temp, "???");
else
{
p3 = embPropCharToThree(*p);
ajStrAppendK(&temp, *p3);
ajStrAppendK(&temp, *(p3+1));
ajStrAppendK(&temp, *(p3+2));
}
}
return temp;
}
void biolib_getorf(AjPSeq seq, AjPTrn table, unsigned int minsize) {
ORFrec record;
/* ORF number to append to name of sequence to create unique name */
ajint orf_no;
AjBool sense; /* ajTrue = forward sense */
ajint len;
orf_no = 1; /* number of the next ORF */
sense = ajTrue; /* forward sense initially */
/* get the length of the sequence */
len = ajSeqGetLen(seq);
/* find the ORFs */
getorf_FindORFs(seq, len, trnTable, minsize/3, 10000, seqout, sense,
circular, find, &orf_no, methionine, around, &record);
/* now reverse complement the sequence and do it again */
sense = ajFalse;
ajSeqReverseForce(seq);
// getorf_FindORFs(seq, len, trnTable, minsize, maxsize, seqout, sense,
// circular, find, &orf_no, methionine,
// around);
}
static void supermatcher_findendpoints(const EmbPWordMatch max,
const AjPSeq trgseq, const AjPSeq qryseq,
ajint *trgstart, ajint *qrystart,
ajint *trgend, ajint *qryend)
{
ajint amax;
ajint bmax;
ajint offset;
*trgstart = max->seq1start;
*qrystart = max->seq2start;
offset = *trgstart - *qrystart;
if(offset > 0)
{
*trgstart = offset;
*qrystart = 0;
}
else
{
*qrystart = 0-offset;
*trgstart = 0;
}
amax = ajSeqGetLen(trgseq)-1;
bmax = ajSeqGetLen(qryseq)-1;
*trgend = *trgstart;
*qryend = *qrystart;
ajDebug("++ end1 %d -> %d end2 %d -> %d\n", *trgend, amax, *qryend, bmax);
while(*trgend<amax && *qryend<bmax)
{
(*trgend)++;
(*qryend)++;
}
ajDebug("++ end1 %d end2 %d\n", *trgend, *qryend);
ajDebug("supermatcher_findendpoints: %d..%d [%d] %d..%d [%d]\n",
trgstart, *trgend, ajSeqGetLen(trgseq), qrystart, *qryend,
ajSeqGetLen(qryseq));
return;
}
AjPStr embPropProt1to3Rev(AjPSeq seq, ajint pad)
{
const char *p;
const char *p3;
AjPStr temp;
ajint i=0;
temp = ajStrNewRes(ajSeqGetLen(seq)*3 + pad+1);
for(p=ajSeqGetSeqC(seq); *p; p++)
{
if(*p == '*')
ajStrAppendC(&temp, "***");
else if(*p == '.')
ajStrAppendC(&temp, "...");
else if(*p == '-')
ajStrAppendC(&temp, "---");
else if(!isalpha((ajint)*p))
ajStrAppendC(&temp, "???");
else
{
p3 = embPropCharToThree(*p);
if(i++)
{
ajStrAppendK(&temp, *(p3+2));
ajStrAppendK(&temp, *(p3+1));
ajStrAppendK(&temp, *p3);
}
else
{
if(pad >= 2)
ajStrAppendK(&temp, *(p3+2));
if(pad >= 1)
ajStrAppendK(&temp, *(p3+1));
ajStrAppendK(&temp, *p3);
}
}
}
return temp;
}
int main(int argc, char **argv)
{
AjPSeq seq = NULL;
AjPFile outf = NULL;
embInitPV("myseq", argc, argv, "myemboss",VERSION);
seq = ajAcdGetSeq("sequence");
outf = ajAcdGetOutfile("outfile");
ajFmtPrintF(outf, "Sequence properties\n");
ajFmtPrintF(outf, "===================\n");
ajFmtPrintF(outf, "Name: %S\n", ajSeqGetNameS(seq));
ajFmtPrintF(outf, "Usa: %S\n", ajSeqGetUsaS(seq));
ajFmtPrintF(outf, "Length: %d\n", ajSeqGetLen(seq));
ajFileClose(&outf);
embExit();
return 0;
}
AjPStr embPropProtGaps(AjPSeq seq, ajint pad)
{
const char *p;
AjPStr temp;
ajint i;
temp = ajStrNewRes(ajSeqGetLen(seq)*3 + pad+1);
/* put any required padding spaces at the start */
for(i=0; i<pad; i++)
ajStrAppendC(&temp, " ");
for(p=ajSeqGetSeqC(seq); *p; p++)
{
ajStrAppendK(&temp, *p);
ajStrAppendC(&temp, " ");
}
return temp;
}
int main(int argc, char **argv)
{
AjPSeqset seqset;
AjPSeqall seqall;
AjPSeq queryseq;
const AjPSeq targetseq;
ajint wordlen;
AjPTable wordsTable = NULL;
AjPList* matchlist = NULL;
AjPFile logfile;
AjPFeattable* seqsetftables = NULL;
AjPFeattable seqallseqftable = NULL;
AjPFeattabOut ftoutforseqsetseq = NULL;
AjPFeattabOut ftoutforseqallseq = NULL;
AjPAlign align = NULL;
AjIList iter = NULL;
ajint targetstart;
ajint querystart;
ajint len;
ajuint i, j;
ajulong nAllMatches = 0;
ajulong sumAllScore = 0;
AjBool dumpAlign = ajTrue;
AjBool dumpFeature = ajTrue;
AjBool checkmode = ajFalse;
EmbPWordRK* wordsw = NULL;
ajuint npatterns = 0;
ajuint seqsetsize;
ajuint nmatches;
ajuint* nmatchesseqset;
ajuint* lastlocation; /* Cursors for Rabin-Karp search. */
/* Shows until what point the query sequence was
* scanned for a pattern sequences in the seqset.
*/
char* paddedheader = NULL;
const char* header;
AjPStr padding;
header = "Pattern %S #pat-sequences #all-matches avg-match-length\n";
padding = ajStrNew();
embInit("wordmatch", argc, argv);
wordlen = ajAcdGetInt("wordsize");
seqset = ajAcdGetSeqset("asequence");
seqall = ajAcdGetSeqall("bsequence");
logfile = ajAcdGetOutfile("logfile");
dumpAlign = ajAcdGetToggle("dumpalign");
dumpFeature = ajAcdGetToggle("dumpfeat");
if(dumpAlign)
{
align = ajAcdGetAlign("outfile");
ajAlignSetExternal(align, ajTrue);
}
seqsetsize = ajSeqsetGetSize(seqset);
ajSeqsetTrim(seqset);
AJCNEW0(matchlist, seqsetsize);
AJCNEW0(seqsetftables, seqsetsize);
AJCNEW0(nmatchesseqset, seqsetsize);
if (dumpFeature)
{
ftoutforseqsetseq = ajAcdGetFeatout("aoutfeat");
ftoutforseqallseq = ajAcdGetFeatout("boutfeat");
}
checkmode = !dumpFeature && !dumpAlign;
embWordLength(wordlen);
ajFmtPrintF(logfile, "Small sequence/file for constructing"
" target patterns: %S\n", ajSeqsetGetUsa(seqset));
ajFmtPrintF(logfile, "Large sequence/file to be scanned"
" for patterns: %S\n", ajSeqallGetUsa(seqall));
ajFmtPrintF(logfile, "Number of sequences in the patterns file: %u\n",
seqsetsize);
ajFmtPrintF(logfile, "Pattern/word length: %u\n", wordlen);
for(i=0;i<seqsetsize;i++)
{
targetseq = ajSeqsetGetseqSeq(seqset, i);
embWordGetTable(&wordsTable, targetseq);
}
AJCNEW0(lastlocation, seqsetsize);
if(ajTableGetLength(wordsTable)>0)
{
npatterns = embWordRabinKarpInit(wordsTable,
&wordsw, wordlen, seqset);
ajFmtPrintF(logfile, "Number of patterns/words found: %u\n", npatterns);
while(ajSeqallNext(seqall,&queryseq))
{
for(i=0;i<seqsetsize;i++)
{
lastlocation[i]=0;
if (!checkmode)
matchlist[i] = ajListstrNew();
}
nmatches = embWordRabinKarpSearch(
ajSeqGetSeqS(queryseq), seqset,
(EmbPWordRK const *)wordsw, wordlen, npatterns,
matchlist, lastlocation, checkmode);
nAllMatches += nmatches;
if (checkmode)
continue;
for(i=0;i<seqsetsize;i++)
{
if(ajListGetLength(matchlist[i])>0)
{
iter = ajListIterNewread(matchlist[i]) ;
while(embWordMatchIter(iter, &targetstart, &querystart, &len,
&targetseq))
{
if(dumpAlign)
{
ajAlignDefineSS(align, targetseq, queryseq);
ajAlignSetScoreI(align, len);
/* ungapped alignment means same length
* for both sequences
*/
ajAlignSetSubRange(align, targetstart, 1, len,
ajSeqIsReversed(targetseq),
ajSeqGetLen(targetseq),
querystart, 1, len,
ajSeqIsReversed(queryseq),
ajSeqGetLen(queryseq));
}
}
if(dumpAlign)
{
ajAlignWrite(align);
ajAlignReset(align);
}
if(ajListGetLength(matchlist[i])>0 && dumpFeature)
{
embWordMatchListConvToFeat(matchlist[i],
&seqsetftables[i],
&seqallseqftable,
targetseq, queryseq);
ajFeattableWrite(ftoutforseqallseq, seqallseqftable);
ajFeattableDel(&seqallseqftable);
}
ajListIterDel(&iter);
}
embWordMatchListDelete(&matchlist[i]);
}
}
/* search completed, now report statistics */
for(i=0;i<npatterns;i++)
{
sumAllScore += wordsw[i]->lenMatches;
for(j=0;j<wordsw[i]->nseqs;j++)
nmatchesseqset[wordsw[i]->seqindxs[j]] +=
wordsw[i]->nSeqMatches[j];
}
ajFmtPrintF(logfile, "Number of sequences in the file scanned "
"for patterns: %u\n", ajSeqallGetCount(seqall));
ajFmtPrintF(logfile, "Number of all matches: %Lu"
" (wordmatch finds exact matches only)\n", nAllMatches);
if(nAllMatches>0)
{
ajFmtPrintF(logfile, "Sum of match lengths: %Lu\n", sumAllScore);
ajFmtPrintF(logfile, "Average match length: %.2f\n",
sumAllScore*1.0/nAllMatches);
ajFmtPrintF(logfile, "\nDistribution of the matches among pattern"
" sequences:\n");
ajFmtPrintF(logfile, "-----------------------------------------"
"-----------\n");
for(i=0;i<ajSeqsetGetSize(seqset);i++)
{
if (nmatchesseqset[i]>0)
ajFmtPrintF(logfile, "%-42s: %8u\n",
ajSeqGetNameC(ajSeqsetGetseqSeq(seqset, i)),
nmatchesseqset[i]);
ajFeattableWrite(ftoutforseqsetseq, seqsetftables[i]);
ajFeattableDel(&seqsetftables[i]);
}
ajFmtPrintF(logfile, "\nPattern statistics:\n");
ajFmtPrintF(logfile, "-------------------\n");
if(wordlen>7)
ajStrAppendCountK(&padding, ' ', wordlen-7);
paddedheader = ajFmtString(header,padding);
ajFmtPrintF(logfile, paddedheader);
for(i=0;i<npatterns;i++)
if (wordsw[i]->nMatches>0)
ajFmtPrintF(logfile, "%-7s: %12u %12u %17.2f\n",
wordsw[i]->word->fword, wordsw[i]->nseqs,
wordsw[i]->nMatches,
wordsw[i]->lenMatches*1.0/wordsw[i]->nMatches);
}
}
for(i=0;i<npatterns;i++)
{
for(j=0;j<wordsw[i]->nseqs;j++)
AJFREE(wordsw[i]->locs[j]);
AJFREE(wordsw[i]->locs);
AJFREE(wordsw[i]->seqindxs);
AJFREE(wordsw[i]->nnseqlocs);
AJFREE(wordsw[i]->nSeqMatches);
AJFREE(wordsw[i]);
}
embWordFreeTable(&wordsTable);
AJFREE(wordsw);
AJFREE(matchlist);
AJFREE(lastlocation);
AJFREE(nmatchesseqset);
AJFREE(seqsetftables);
if(dumpAlign)
{
ajAlignClose(align);
ajAlignDel(&align);
}
if(dumpFeature)
{
ajFeattabOutDel(&ftoutforseqsetseq);
ajFeattabOutDel(&ftoutforseqallseq);
}
ajFileClose(&logfile);
ajSeqallDel(&seqall);
ajSeqsetDel(&seqset);
ajSeqDel(&queryseq);
ajStrDel(&padding);
AJFREE(paddedheader);
embExit();
return 0;
}
int main(int argc, char **argv)
{
AjPSeqset seq1;
AjPSeqset seq2;
AjPFile list;
ajint n1;
ajint n2;
ajint *lengths1;
ajint *lengths2;
ajuint *order1;
ajuint *order2;
ajint *hits1;
ajint *hits2;
ajint curr1;
ajint curr2;
ajint tmp1;
ajint tmp2 = 0;
ajint i;
AjPStr operator;
ajint OperatorCode=0;
embInit("listor", argc, argv);
seq1 = ajAcdGetSeqset("firstsequences");
seq2 = ajAcdGetSeqset("secondsequences");
list = ajAcdGetOutfile("outfile");
operator = ajAcdGetListSingle("operator");
/* get the operator value */
switch(ajStrGetCharFirst(operator))
{
case 'O':
OperatorCode = L_OR;
break;
case 'A':
OperatorCode = L_AND;
break;
case 'X':
OperatorCode = L_XOR;
break;
case 'N':
OperatorCode = L_NOT;
break;
default:
ajFatal("Invalid operator type: %S", operator);
embExitBad();
}
/* get the order of seqset 1 by length */
n1 = ajSeqsetGetSize(seq1);
/* lengths of seq1 entries */
lengths1 = AJCALLOC0(n1, sizeof(ajint));
/* seq1 entries which match seq2 */
hits1 = AJCALLOC0(n1, sizeof(ajint));
/* seq1 entries in length order */
order1 = AJCALLOC0(n1, sizeof(ajint));
for(i=0; i<n1; i++)
{
lengths1[i] = ajSeqGetLen(ajSeqsetGetseqSeq(seq1, i));
order1[i] = i;
hits1[i] = -1;
}
ajSortIntIncI(lengths1, order1, n1);
/* get the order of seqset 2 by length */
n2 = ajSeqsetGetSize(seq2);
lengths2 = AJCALLOC0(n2, sizeof(ajint));
hits2 = AJCALLOC0(n2, sizeof(ajint));
order2 = AJCALLOC0(n2, sizeof(ajint));
for(i=0; i<n2; i++)
{
lengths2[i] = ajSeqGetLen(ajSeqsetGetseqSeq(seq2, i));
order2[i] = i;
hits2[i] = -1;
}
ajSortIntIncI(lengths2, order2, n2);
/*
** go down the two sequence sets, by size order, looking for identical
**lengths
*/
curr1 = 0;
curr2 = 0;
while(curr1 < n1 && curr2 < n2)
{
if(lengths1[order1[curr1]] < lengths2[order2[curr2]])
/* seq1 is shorter - increment curr1 index */
curr1++;
else if(lengths1[order1[curr1]] > lengths2[order2[curr2]])
/* seq2 is shorter - increment curr2 index */
curr2++;
else
{
/* identical lengths - check all seq1/seq2 entries of this len */
for(tmp1=curr1; tmp1<n1
&& lengths1[order1[tmp1]] == lengths2[order2[curr2]]; tmp1++)
for(tmp2=curr2; tmp2<n2 && lengths2[order2[tmp2]] ==
lengths2[order2[curr2]]; tmp2++)
/* check to see if the sequences are identical */
if(!ajStrCmpCaseS(ajSeqGetSeqS(ajSeqsetGetseqSeq(seq1,
order1[tmp1])),
ajSeqGetSeqS(ajSeqsetGetseqSeq(seq2,
order2[tmp2]))))
{
hits1[order1[tmp1]] = order2[tmp2];
hits2[order2[tmp2]] = order1[tmp1];
}
curr1 = tmp1;
curr2 = tmp2;
}
}
/* output the required entries to the list file */
listor_Output(list, OperatorCode, seq1, seq2, hits1, hits2, n1, n2);
AJFREE(lengths1);
AJFREE(lengths2);
AJFREE(order1);
AJFREE(order2);
AJFREE(hits1);
AJFREE(hits2);
ajFileClose(&list);
ajStrDel(&operator);
ajSeqsetDel(&seq1);
ajSeqsetDel(&seq2);
embExit();
return 0;
}
int main(int argc, char **argv)
{
AjPDasServer server = NULL;
AjPDasSource source = NULL;
AjPDasSegment segment = NULL;
AjPStr host = NULL;
AjPStr path = NULL;
AjPFile outf = NULL;
ajint port = 80;
AjBool sequencesourcesonly;
AjBool entrypoints;
AjBool showtestqueries;
AjBool runtestqueries = ajTrue;
AjBool quickexit = ajFalse;
ajint itest=0;
ajint j=0;
ajint maxtests=0;
ajint maxfeatures=0;
ajint maxsegments=0;
AjIList iter = NULL;
AjIList coordsi = NULL;
AjIList itereps = NULL;
AjPFilebuff buff = NULL;
AjPUrlref uo = NULL;
AjPList segments = NULL;
AjPStr ffname = NULL;
AjPStr url = NULL;
AjPStr dbhttpver = ajStrNew();
AjPStr dbname = ajStrNew();
AjPStr dbproxy = ajStrNew();
AjPStr servername = NULL;
AjPTable titlecount = NULL;
const ajuint* count;
int k=0;
embInit("dastest", argc, argv);
host = ajAcdGetString("host");
path = ajAcdGetString("path");
port = ajAcdGetInt("port");
sequencesourcesonly = ajAcdGetBoolean("sequencesourcesonly");
entrypoints = ajAcdGetBoolean("entrypoints");
showtestqueries = ajAcdGetBoolean("showtestqueries");
runtestqueries = ajAcdGetBoolean("runtestqueries");
servername = ajAcdGetString("servername");
outf = ajAcdGetOutfile("outfile");
maxtests = ajAcdGetInt("maxtests");
maxfeatures = ajAcdGetInt("maxfeatures");
maxsegments = ajAcdGetInt("maxsegments");
server = ajDasServerNew();
if(runtestqueries)
{
url = ajStrNew();
if(!ajNamServer(servername))
{
ajWarn("following das server is required to be defined "
"for test queries...");
ajWarn("\nSERVER %S [\n"
" type: \"sequence\"\n"
" method: \"das\"\n"
" url: \"http://%S%S\"\n"
"]\n",servername, host,path);
ajWarn("ignoring -runtestqueries option...");
runtestqueries = ajFalse;
}
else
{
ajNamSvrGetUrl(servername, &url);
ajHttpUrlDeconstruct(url, &port, &host, &path);
ajStrDel(&url);
}
}
ajDasServerSethostS(server,host);
ajDasServerSetport(server,port);
if(ajStrGetCharLast(path)!='/')
ajStrAppendK(&path,'/');
ajStrAppendC(&path,"sources");
ajDasServerSetpathS(server,path);
ajFmtPrintF(outf,"host = %S\npath = %S\nport = %d\n",
server->host, server->path, server->port);
/*
* TODO: stop using http-read but instead use
* ajNamSvrListListDatabases(svrname, dbnames);
*/
buff = ajHttpRead(dbhttpver, dbname, dbproxy, host, port, path);
if(!buff)
ajExitAbort();
ajFilebuffHtmlNoheader(buff);
ajDasParseRegistry(buff, server->sources);
ajFmtPrintF(outf,"DAS sources and descriptions\n\n");
titlecount = dastestGetTitleCount(server);
iter = ajListIterNew(server->sources);
while(!ajListIterDone(iter) && !quickexit)
{
source = ajListIterGet(iter);
if ((sequencesourcesonly && !source->sequence)
|| ajStrMatchC(source->title,"cath") || k++ <50)
continue;
ajFmtPrintF(outf,"%-30S %-50S\n%S\n",source->uri,source->title,
source->description);
if(entrypoints && source->entry_points)
{
uo = ajHttpUrlrefNew();
ajHttpUrlrefParseC(&uo, ajStrGetPtr(source->entry_points_uri));
if(ajStrGetLen(uo->Port))
ajStrToInt(uo->Port, &port);
else
port = 80;
ajFilebuffDel(&buff);
buff = ajHttpRead(dbhttpver, dbname, dbproxy,
uo->Host, port, uo->Absolute);
ajHttpUrlrefDel(&uo);
if(!buff)
continue;
ajFilebuffHtmlNoheader(buff);
segments = ajListNew();
ajDasParseEntrypoints(buff, segments);
itereps = ajListIterNew(segments);
ajFmtPrintF(outf, "Number of entry points %d\n",
ajListGetLength(segments));
j=0;
while(!ajListIterDone(itereps))
{
segment = ajListIterGet(itereps);
if (j++ < maxsegments)
ajFmtPrintF(outf,
"segment id:%S orientation:%S start:%d stop:%d\n",
segment->id,
segment->orientation,
segment->start, segment->stop);
ajDasSegmentDel(&segment);
}
ajListIterDel(&itereps);
ajListFree(&segments);
}
if(showtestqueries || runtestqueries)
{
AjPDasCoordinate coord;
coordsi = ajListIterNew(source->coordinates);
while(!ajListIterDone(coordsi) && !quickexit)
{
coord = ajListIterGet(coordsi);
ajDebug("coordinate uri:%S taxid:%S source:%S test_range:%S\n",
coord->uri,
coord->taxid,
coord->source,
coord->test_range);
if(showtestqueries)
{
if(source->sequence)
ajFmtPrintF(outf,
"example/test entry = '%S?segment=%S'\n",
source->sequence_query_uri,coord->test_range);
if(source->features)
ajFmtPrintF(outf,
"example/test entry = '%S?segment=%S'\n",
source->features_query_uri,coord->test_range);
}
if(runtestqueries)
{
AjPStr idqry = ajStrNew();
AjPStr entry = NULL;
AjPSeq seq = NULL;
ajint ibegin = 0;
ajint iend = 0;
AjPStr example = NULL;
example = ajDasTestrangeParse(coord->test_range,
&entry, &ibegin, &iend);
if(ajStrGetLen(entry))
{
count = ajTableFetchS(titlecount, source->title);
dbname = ajDasSourceGetDBname(source, *count>1);
if (source->features)
{
AjPStr qpath=NULL;
uo = ajHttpUrlrefNew();
ajFmtPrintS(&idqry,"dasgff::%S:%S:%S",
servername,
dbname,
entry);
ajFmtPrintF(outf,
"feature query: %S start:%d end:%d\n",
idqry,
ibegin, iend);
ajHttpUrlrefParseC(&uo,
ajStrGetPtr(source->features_query_uri));
ajHttpUrlrefSplitPort(uo);
ajFmtPrintS(&qpath,"%S?segment=%S",
uo->Absolute,entry);
if(iend>0)
ajFmtPrintAppS(&qpath,":%d,%d",ibegin, iend);
if(ajStrGetLen(uo->Port))
ajStrToInt(uo->Port, &port);
else
port = 80;
ajDebug("calling ajHttpRead to get the raw"
" output; host:%S port:%d path:%S\n",
uo->Host, port, qpath);
ajFilebuffDel(&buff);
buff = ajHttpRead(dbhttpver, dbname, dbproxy,
uo->Host, port, qpath);
if(buff)
{
AjPFeattable ft;
ajFmtPrintS(&ffname, "%S.%S", source->uri,
entry);
ajFilebuffHtmlNoheader(buff);
dastestSaveRawFeatures(buff, ffname);
ajDebug("now using EMBOSS feature queries\n");
ft = dastestFeatureQuery(idqry,
ibegin, iend);
dastestSaveMappedFeatures(ft, ffname,
outf, maxfeatures);
ajStrDel(&ffname);
ajFeattableDel(&ft);
}
ajHttpUrlrefDel(&uo);
ajStrDel(&qpath);
if(++itest>=maxtests)
quickexit = ajTrue;
}
else if(source->sequence)
{
seq = ajSeqNewRes(iend-ibegin+1);
ajFmtPrintS(&idqry,"%S:%S:%S",
servername, dbname, entry);
ajFmtPrintF(outf,
"sequence query: %S start:%d end:%d\n",
idqry,
ibegin, iend);
ajSeqGetFromUsaRange(idqry, ajFalse, ibegin, iend,
seq);
ajFmtPrintF(outf,
"length of sequence returned: %d\n",
ajSeqGetLen(seq));
if(ajSeqGetLen(seq)>0)
ajFmtPrintF(outf,
"sequence returned (first 100 bases):"
" %-100.100s\n",
ajSeqGetSeqC(seq));
ajSeqDel(&seq);
}
ajStrDel(&dbname);
}
ajStrDel(&entry);
ajStrDel(&idqry);
ajStrDel(&example);
}
}
ajListIterDel(&coordsi);
}
}
ajListIterDel(&iter);
ajDasServerDel(&server);
ajFilebuffDel(&buff);
ajStrDel(&host);
ajStrDel(&path);
ajStrDel(&servername);
ajStrDel(&dbhttpver);
ajStrDel(&dbname);
ajStrDel(&dbproxy);
ajFileClose(&outf);
ajTableDelValdel(&titlecount, ajMemFree);
embExit();
return 0;
}
int main(int argc, char **argv)
{
AjPSeqall seq1;
AjPSeqset seq2;
AjPSeq a;
const AjPSeq b;
AjPStr m = 0;
AjPStr n = 0;
AjPFile errorf;
AjBool show = ajFalse;
ajint lena = 0;
ajint lenb = 0;
const char *p;
const char *q;
AjPMatrixf matrix;
AjPSeqCvt cvt = 0;
float **sub;
ajint *compass = NULL;
float *path = NULL;
float gapopen;
float gapextend;
float score;
ajint begina;
ajint i;
ajuint k;
ajint beginb;
ajint start1 = 0;
ajint start2 = 0;
ajint end1 = 0;
ajint end2 = 0;
ajint width = 0;
AjPTable seq1MatchTable = 0;
ajint wordlen = 6;
ajint oldmax = 0;
AjPAlign align = NULL;
embInit("supermatcher", argc, argv);
matrix = ajAcdGetMatrixf("datafile");
seq1 = ajAcdGetSeqall("asequence");
seq2 = ajAcdGetSeqset("bsequence");
gapopen = ajAcdGetFloat("gapopen");
gapextend = ajAcdGetFloat("gapextend");
wordlen = ajAcdGetInt("wordlen");
align = ajAcdGetAlign("outfile");
errorf = ajAcdGetOutfile("errorfile");
width = ajAcdGetInt("width"); /* not the same as awidth */
gapopen = ajRoundFloat(gapopen, 8);
gapextend = ajRoundFloat(gapextend, 8);
sub = ajMatrixfGetMatrix(matrix);
cvt = ajMatrixfGetCvt(matrix);
embWordLength(wordlen);
ajSeqsetTrim(seq2);
while(ajSeqallNext(seq1,&a))
{
ajSeqTrim(a);
begina = 1 + ajSeqGetOffset(a);
m = ajStrNewRes(1+ajSeqGetLen(a));
lena = ajSeqGetLen(a);
ajDebug("Read '%S'\n", ajSeqGetNameS(a));
if(!embWordGetTable(&seq1MatchTable, a)) /* get table of words */
ajErr("Could not generate table for %s\n",
ajSeqGetNameC(a));
for(k=0;k<ajSeqsetGetSize(seq2);k++)
{
b = ajSeqsetGetseqSeq(seq2, k);
lenb = ajSeqGetLen(b);
beginb = 1 + ajSeqGetOffset(b);
ajDebug("Processing '%S'\n", ajSeqGetNameS(b));
p = ajSeqGetSeqC(a);
q = ajSeqGetSeqC(b);
if(!supermatcher_findstartpoints(seq1MatchTable,b,a,
&start1, &start2,
&end1, &end2))
{
ajFmtPrintF(errorf,
"No wordmatch start points for "
"%s vs %s. No alignment\n",
ajSeqGetNameC(a),ajSeqGetNameC(b));
continue;
}
n=ajStrNewRes(1+ajSeqGetLen(b));
ajStrAssignC(&m,"");
ajStrAssignC(&n,"");
ajDebug("++ %S v %S start:%d %d end:%d %d\n",
ajSeqGetNameS(a), ajSeqGetNameS(b),
start1, start2, end1, end2);
if(end1-start1+1 > oldmax)
{
oldmax = ((end1-start1)+1);
AJRESIZE(path,oldmax*width*sizeof(float));
AJRESIZE(compass,oldmax*width*sizeof(ajint));
ajDebug("++ resize to oldmax: %d\n", oldmax);
}
for(i=0;i<((end1-start1)+1)*width;i++)
path[i] = 0.0;
ajDebug("Calling embAlignPathCalcFast "
"%d..%d [%d/%d] %d..%d [%d/%d]\n",
start1, end1, (end1 - start1 + 1), lena,
start2, end2, (end2 - start2 + 1), lenb);
score = embAlignPathCalcSWFast(&p[start1],&q[start2],
end1-start1+1,end2-start2+1,
0,width,
gapopen,gapextend,
path,sub,cvt,
compass,show);
embAlignWalkSWMatrixFast(path,compass,gapopen,gapextend,a,b,
&m,&n,end1-start1+1,end2-start2+1,
0,width,
&start1,&start2);
if(!ajAlignFormatShowsSequences(align))
{
ajAlignDefineCC(align, ajStrGetPtr(m),
ajStrGetPtr(n), ajSeqGetNameC(a),
ajSeqGetNameC(b));
ajAlignSetScoreR(align, score);
}
else
{
embAlignReportLocal(align, a, b,
m,n,start1,start2,
gapopen, gapextend,
score,matrix, begina, beginb);
}
ajAlignWrite(align);
ajAlignReset(align);
ajStrDel(&n);
}
embWordFreeTable(&seq1MatchTable); /* free table of words */
seq1MatchTable=0;
ajStrDel(&m);
}
if(!ajAlignFormatShowsSequences(align))
{
ajMatrixfDel(&matrix);
}
AJFREE(path);
AJFREE(compass);
ajAlignClose(align);
ajAlignDel(&align);
ajSeqallDel(&seq1);
ajSeqDel(&a);
ajSeqsetDel(&seq2);
ajFileClose(&errorf);
embExit();
return 0;
}
int main(int argc, char **argv)
{
AjPAlign align;
AjPSeqall seqall;
AjPSeq a;
AjPSeq b;
AjPStr alga;
AjPStr algb;
AjPStr ss;
ajuint lena;
ajuint lenb;
const char *p;
const char *q;
ajint start1 = 0;
ajint start2 = 0;
float *path;
ajint *compass;
float* ix;
float* iy;
float* m;
AjPMatrixf matrix;
AjPSeqCvt cvt = 0;
float **sub;
float gapopen;
float gapextend;
float endgapopen;
float endgapextend;
ajulong maxarr = 1000; /* arbitrary. realloc'd if needed */
ajulong len;
float score;
AjBool dobrief = ajTrue;
AjBool endweight = ajFalse; /* whether end gap penalties should be applied */
float id = 0.;
float sim = 0.;
float idx = 0.;
float simx = 0.;
AjPStr tmpstr = NULL;
size_t stlen;
embInit("needle", argc, argv);
matrix = ajAcdGetMatrixf("datafile");
a = ajAcdGetSeq("asequence");
ajSeqTrim(a);
seqall = ajAcdGetSeqall("bsequence");
gapopen = ajAcdGetFloat("gapopen");
gapextend = ajAcdGetFloat("gapextend");
endgapopen = ajAcdGetFloat("endopen");
endgapextend = ajAcdGetFloat("endextend");
dobrief = ajAcdGetBoolean("brief");
endweight = ajAcdGetBoolean("endweight");
align = ajAcdGetAlign("outfile");
gapopen = ajRoundFloat(gapopen, 8);
gapextend = ajRoundFloat(gapextend, 8);
AJCNEW(path, maxarr);
AJCNEW(compass, maxarr);
AJCNEW(m, maxarr);
AJCNEW(ix, maxarr);
AJCNEW(iy, maxarr);
alga = ajStrNew();
algb = ajStrNew();
ss = ajStrNew();
sub = ajMatrixfGetMatrix(matrix);
cvt = ajMatrixfGetCvt(matrix);
lena = ajSeqGetLen(a);
while(ajSeqallNext(seqall,&b))
{
ajSeqTrim(b);
lenb = ajSeqGetLen(b);
if(lenb > (ULONG_MAX/(ajulong)(lena+1)))
ajFatal("Sequences too big. Try 'stretcher' or 'supermatcher'");
len = lena*lenb;
if(len>maxarr)
{
stlen = (size_t) len;
AJCRESIZETRY(path,stlen);
if(!path)
ajDie("Sequences too big. Try 'stretcher'");
AJCRESIZETRY(compass,stlen);
if(!compass)
ajDie("Sequences too big. Try 'stretcher'");
AJCRESIZETRY(m,stlen);
if(!m)
ajDie("Sequences too big. Try 'stretcher'");
AJCRESIZETRY(ix,stlen);
if(!ix)
ajDie("Sequences too big. Try 'stretcher'");
AJCRESIZETRY(iy,stlen);
if(!iy)
ajDie("Sequences too big. Try 'stretcher'");
maxarr=len;
}
p = ajSeqGetSeqC(a);
q = ajSeqGetSeqC(b);
ajStrAssignC(&alga,"");
ajStrAssignC(&algb,"");
score = embAlignPathCalcWithEndGapPenalties(p, q, lena, lenb,
gapopen, gapextend, endgapopen, endgapextend,
&start1, &start2, path, sub, cvt,
m, ix, iy, compass, ajTrue, endweight);
embAlignWalkNWMatrixUsingCompass(p, q, &alga, &algb,
lena, lenb, &start1, &start2,
compass);
embAlignReportGlobal(align, a, b, alga, algb,
start1, start2,
gapopen, gapextend,
score, matrix,
ajSeqGetOffset(a), ajSeqGetOffset(b));
if(!dobrief)
{
embAlignCalcSimilarity(alga,algb,sub,cvt,lena,lenb,&id,&sim,&idx,
&simx);
ajFmtPrintS(&tmpstr,"Longest_Identity = %5.2f%%\n",
id);
ajFmtPrintAppS(&tmpstr,"Longest_Similarity = %5.2f%%\n",
sim);
ajFmtPrintAppS(&tmpstr,"Shortest_Identity = %5.2f%%\n",
idx);
ajFmtPrintAppS(&tmpstr,"Shortest_Similarity = %5.2f%%",
simx);
ajAlignSetSubHeaderApp(align, tmpstr);
}
ajAlignWrite(align);
ajAlignReset(align);
}
ajAlignClose(align);
ajAlignDel(&align);
ajSeqallDel(&seqall);
ajSeqDel(&a);
ajSeqDel(&b);
AJFREE(compass);
AJFREE(path);
AJFREE(ix);
AJFREE(iy);
AJFREE(m);
ajStrDel(&alga);
ajStrDel(&algb);
ajStrDel(&ss);
ajStrDel(&tmpstr);
embExit();
return 0;
}
int main(int argc, char *argv[])
{
char *string;
char *structure=NULL;
char *cstruc=NULL;
char *ns_bases=NULL;
char *c;
int n_seq;
int i;
int length;
int sym;
int endgaps = 0;
int mis = 0;
double min_en;
double real_en;
double sfact = 1.07;
int pf = 0;
int istty;
char *AS[MAX_NUM_NAMES]; /* aligned sequences */
char *names[MAX_NUM_NAMES]; /* sequence names */
AjPSeqset seq = NULL;
AjPFile confile = NULL;
AjPFile alifile = NULL;
AjPFile paramfile = NULL;
AjPFile outf = NULL;
AjPFile essfile = NULL;
AjPFile dotfile = NULL;
AjPStr constring = NULL;
float eT = 0.;
AjBool eGU;
AjBool eclose;
AjBool lonely;
AjPStr ensbases = NULL;
AjBool etloop;
AjPStr eenergy = NULL;
char ewt = '\0';
float escale = 0.;
AjPStr edangles = NULL;
char edangle = '\0';
ajint len;
AjPSeq tseq = NULL;
AjPStr tname = NULL;
int circ = 0;
int doAlnPS = 0;
int doColor = 0;
embInitPV("vrnaalifoldpf",argc,argv,"VIENNA",VERSION);
constring = ajStrNew();
seq = ajAcdGetSeqset("sequence");
confile = ajAcdGetInfile("constraintfile");
paramfile = ajAcdGetInfile("paramfile");
eT = ajAcdGetFloat("temperature");
eGU = ajAcdGetBoolean("gu");
eclose = ajAcdGetBoolean("closegu");
lonely = ajAcdGetBoolean("lp");
ensbases = ajAcdGetString("nsbases");
etloop = ajAcdGetBoolean("tetraloop");
eenergy = ajAcdGetListSingle("energy");
escale = ajAcdGetFloat("scale");
edangles = ajAcdGetListSingle("dangles");
mis = !!ajAcdGetBoolean("most");
endgaps = !!ajAcdGetBoolean("endgaps");
nc_fact = (double) ajAcdGetFloat("nspenalty");
cv_fact = (double) ajAcdGetFloat("covariance");
outf = ajAcdGetOutfile("outfile");
essfile = ajAcdGetOutfile("ssoutfile");
alifile = ajAcdGetOutfile("alignoutfile");
circ = !!ajAcdGetBoolean("circular");
doColor = !!ajAcdGetBoolean("colour");
dotfile = ajAcdGetOutfile("dotoutfile");
do_backtrack = 1;
pf = 1;
string = NULL;
istty = 0;
dangles = 2;
temperature = (double) eT;
noGU = (eGU) ? 0 : 1;
no_closingGU = (eclose) ? 0 : 1;
noLonelyPairs = (lonely) ? 0 : 1;
ns_bases = (ajStrGetLen(ensbases)) ? MAJSTRGETPTR(ensbases) : NULL;
tetra_loop = !!etloop;
ewt = *ajStrGetPtr(eenergy);
if(ewt == '0')
energy_set = 0;
else if(ewt == '1')
energy_set = 1;
else if(ewt == '2')
energy_set = 2;
sfact = (double) escale;
edangle = *ajStrGetPtr(edangles);
if(edangle == '0')
dangles = 0;
else if(edangle == '1')
dangles = 1;
else if(edangle == '2')
dangles = 2;
else if(edangle == '3')
dangles = 3;
if(paramfile)
read_parameter_file(paramfile);
if (ns_bases != NULL)
{
nonstandards = space(33);
c=ns_bases;
i=sym=0;
if (*c=='-')
{
sym=1;
c++;
}
while (*c!='\0')
{
if (*c!=',')
{
nonstandards[i++]=*c++;
nonstandards[i++]=*c;
if ((sym)&&(*c!=*(c-1)))
{
nonstandards[i++]=*c;
nonstandards[i++]=*(c-1);
}
}
c++;
}
}
if(alifile)
doAlnPS = 1;
if(confile)
vienna_GetConstraints(confile,&constring);
n_seq = ajSeqsetGetSize(seq);
if(n_seq > MAX_NUM_NAMES - 1)
ajFatal("[e]RNAalifold is restricted to %d sequences\n",
MAX_NUM_NAMES - 1);
if (n_seq==0)
ajFatal("No sequences found");
for(i=0;i<n_seq;++i)
{
tseq = (AjPSeq) ajSeqsetGetseqSeq(seq,i);
ajSeqGapStandard(tseq, '-');
tname = (AjPStr) ajSeqsetGetseqNameS(seq,i);
len = ajSeqGetLen(tseq);
AS[i] = (char *) space(len+1);
names[i] = (char *) space(ajStrGetLen(tname)+1);
strcpy(AS[i],ajSeqGetSeqC(tseq));
strcpy(names[i],ajStrGetPtr(tname));
}
AS[n_seq] = NULL;
names[n_seq] = NULL;
if (endgaps)
for (i=0; i<n_seq; i++)
mark_endgaps(AS[i], '~');
length = (int) strlen(AS[0]);
structure = (char *) space((unsigned) length+1);
if(confile)
{
fold_constrained = 1;
strcpy(structure,ajStrGetPtr(constring));
}
if (circ && noLonelyPairs)
ajWarn(
"warning, depending on the origin of the circular sequence, "
"some structures may be missed when using -noLP\n"
"Try rotating your sequence a few times\n");
if (circ)
min_en = circalifold((const char **)AS, structure);
else
min_en = alifold(AS, structure);
{
int i;
double s=0;
extern int eos_debug;
eos_debug=-1; /* shut off warnings about nonstandard pairs */
for (i=0; AS[i]!=NULL; i++)
if (circ)
s += energy_of_circ_struct(AS[i], structure);
else
s += energy_of_struct(AS[i], structure);
real_en = s/i;
}
string = (mis) ?
consens_mis((const char **) AS) : consensus((const char **) AS);
ajFmtPrintF(outf,"%s\n%s", string, structure);
ajFmtPrintF(outf," (%6.2f = %6.2f + %6.2f) \n", min_en, real_en,
min_en-real_en );
if (length<=2500) {
char **A;
A = annote(structure, (const char**) AS);
if (doColor)
(void) PS_rna_plot_a(string, structure, essfile, A[0], A[1]);
else
(void) PS_rna_plot_a(string, structure, essfile, NULL, A[1]);
free(A[0]); free(A[1]);free(A);
} else
ajWarn("INFO: structure too long, not doing xy_plot\n");
if (doAlnPS)
PS_color_aln(structure, alifile, AS, names);
{ /* free mfe arrays but preserve base_pair for PS_dot_plot */
struct bond *bp;
bp = base_pair; base_pair = space(16);
free_alifold_arrays(); /* free's base_pair */
free_alipf_arrays();
base_pair = bp;
}
if (pf) {
double energy, kT;
pair_info *pi;
char * mfe_struc;
mfe_struc = strdup(structure);
kT = (temperature+273.15)*1.98717/1000.; /* in Kcal */
pf_scale = exp(-(sfact*min_en)/kT/length);
if (length>2000)
ajWarn("scaling factor %f\n", pf_scale);
/* init_alipf_fold(length); */
if (confile)
strncpy(structure, ajStrGetPtr(constring), length+1);
energy = (circ) ? alipf_circ_fold(AS, structure, &pi) : alipf_fold(AS, structure, &pi);
if (do_backtrack) {
ajFmtPrintF(outf,"%s", structure);
ajFmtPrintF(outf," [%6.2f]\n", energy);
}
if ((istty)||(!do_backtrack))
ajFmtPrintF(outf," free energy of ensemble = %6.2f kcal/mol\n", energy);
ajFmtPrintF(outf," frequency of mfe structure in ensemble %g\n",
exp((energy-min_en)/kT));
if (do_backtrack) {
FILE *aliout;
cpair *cp;
short *ptable; int k;
ptable = make_pair_table(mfe_struc);
ajFmtPrintF(outf,"\n# Alignment section\n\n");
aliout = ajFileGetFileptr(outf);
fprintf(aliout, "%d sequences; length of alignment %d\n",
n_seq, length);
fprintf(aliout, "alifold output\n");
for (k=0; pi[k].i>0; k++) {
pi[k].comp = (ptable[pi[k].i] == pi[k].j) ? 1:0;
print_pi(pi[k], aliout);
}
fprintf(aliout, "%s\n", structure);
free(ptable);
cp = make_color_pinfo(pi);
(void) PS_color_dot_plot(string, cp, dotfile);
free(cp);
free(mfe_struc);
free(pi);
}
}
if (cstruc!=NULL) free(cstruc);
free(base_pair);
(void) fflush(stdout);
free(string);
free(structure);
for (i=0; AS[i]; i++) {
free(AS[i]); free(names[i]);
}
ajSeqsetDel(&seq);
ajStrDel(&constring);
ajStrDel(&eenergy);
ajStrDel(&edangles);
ajStrDel(&ensbases);
ajFileClose(&confile);
ajFileClose(¶mfile);
ajFileClose(&outf);
ajFileClose(&essfile);
ajFileClose(&alifile);
ajFileClose(&dotfile);
embExit();
return 0;
}
int main(int argc, char **argv)
{
AjPSeqall queryseqs;
AjPSeqset targetseqs;
AjPSeq queryseq;
const AjPSeq targetseq;
AjPStr queryaln = 0;
AjPStr targetaln = 0;
AjPFile errorf;
AjBool show = ajFalse;
const char *queryseqc;
const char *targetseqc;
AjPMatrixf matrix;
AjPSeqCvt cvt = 0;
float **sub;
ajint *compass = NULL;
float *path = NULL;
float gapopen;
float gapextend;
float score;
float minscore;
ajuint j, k;
ajint querystart = 0;
ajint targetstart = 0;
ajint queryend = 0;
ajint targetend = 0;
ajint width = 0;
AjPTable kmers = 0;
ajint wordlen = 6;
ajint oldmax = 0;
ajint newmax = 0;
ajuint ntargetseqs;
ajuint nkmers;
AjPAlign align = NULL;
EmbPWordMatch maxmatch; /* match with maximum score */
/* Cursors for the current sequence being scanned,
** i.e., until which location it was scanned.
** Separate cursor/location entries for each sequence in the seqset.
*/
ajuint* lastlocation;
EmbPWordRK* wordsw = NULL;
AjPList* matchlist = NULL;
embInit("supermatcher", argc, argv);
matrix = ajAcdGetMatrixf("datafile");
queryseqs = ajAcdGetSeqall("asequence");
targetseqs= ajAcdGetSeqset("bsequence");
gapopen = ajAcdGetFloat("gapopen");
gapextend = ajAcdGetFloat("gapextend");
wordlen = ajAcdGetInt("wordlen");
align = ajAcdGetAlign("outfile");
errorf = ajAcdGetOutfile("errorfile");
width = ajAcdGetInt("width"); /* width for banded Smith-Waterman */
minscore = ajAcdGetFloat("minscore");
gapopen = ajRoundFloat(gapopen, 8);
gapextend = ajRoundFloat(gapextend, 8);
sub = ajMatrixfGetMatrix(matrix);
cvt = ajMatrixfGetCvt(matrix);
embWordLength(wordlen);
/* seqset sequence is the reference sequence for SAM format */
ajAlignSetRefSeqIndx(align, 1);
ajSeqsetTrim(targetseqs);
ntargetseqs = ajSeqsetGetSize(targetseqs);
AJCNEW0(matchlist, ntargetseqs);
/* get tables of words */
for(k=0;k<ntargetseqs;k++)
{
targetseq = ajSeqsetGetseqSeq(targetseqs, k);
embWordGetTable(&kmers, targetseq);
ajDebug("Number of distinct kmers found so far: %d\n",
ajTableGetLength(kmers));
}
AJCNEW0(lastlocation, ntargetseqs);
if(ajTableGetLength(kmers)<1)
ajErr("no kmers found");
nkmers = embWordRabinKarpInit(kmers, &wordsw, wordlen, targetseqs);
while(ajSeqallNext(queryseqs,&queryseq))
{
ajSeqTrim(queryseq);
queryaln = ajStrNewRes(1+ajSeqGetLen(queryseq));
ajDebug("Read '%S'\n", ajSeqGetNameS(queryseq));
for(k=0;k<ntargetseqs;k++)
{
lastlocation[k]=0;
matchlist[k] = ajListstrNew();
}
embWordRabinKarpSearch(ajSeqGetSeqS(queryseq), targetseqs,
(const EmbPWordRK*)wordsw, wordlen, nkmers,
matchlist, lastlocation, ajFalse);
for(k=0;k<ajSeqsetGetSize(targetseqs);k++)
{
targetseq = ajSeqsetGetseqSeq(targetseqs, k);
ajDebug("Processing '%S'\n", ajSeqGetNameS(targetseq));
if(ajListGetLength(matchlist[k])==0)
{
ajFmtPrintF(errorf,
"No wordmatch start points for "
"%s vs %s. No alignment\n",
ajSeqGetNameC(queryseq),ajSeqGetNameC(targetseq));
embWordMatchListDelete(&matchlist[k]);
continue;
}
/* only the maximum match is used as seed
* (if there is more than one location with the maximum match
* only the first one is used)
* TODO: we should add a new option to make above limit optional
*/
maxmatch = embWordMatchFirstMax(matchlist[k]);
supermatcher_findendpoints(maxmatch,targetseq, queryseq,
&targetstart, &querystart,
&targetend, &queryend);
targetaln=ajStrNewRes(1+ajSeqGetLen(targetseq));
queryseqc = ajSeqGetSeqC(queryseq);
targetseqc = ajSeqGetSeqC(targetseq);
ajStrAssignC(&queryaln,"");
ajStrAssignC(&targetaln,"");
ajDebug("++ %S v %S start:%d %d end:%d %d\n",
ajSeqGetNameS(targetseq), ajSeqGetNameS(queryseq),
targetstart, querystart, targetend, queryend);
newmax = (targetend-targetstart+2)*width;
if(newmax > oldmax)
{
AJCRESIZE0(path,oldmax,newmax);
AJCRESIZE0(compass,oldmax,newmax);
oldmax=newmax;
ajDebug("++ memory re/allocation for path/compass arrays"
" to size: %d\n", newmax);
}
else
{
AJCSET0(path,newmax);
AJCSET0(compass,newmax);
}
ajDebug("Calling embAlignPathCalcSWFast "
"%d..%d [%d/%d] %d..%d [%d/%d] width:%d\n",
querystart, queryend, (queryend - querystart + 1),
ajSeqGetLen(queryseq),
targetstart, targetend, (targetend - targetstart + 1),
ajSeqGetLen(targetseq),
width);
score = embAlignPathCalcSWFast(&targetseqc[targetstart],
&queryseqc[querystart],
targetend-targetstart+1,
queryend-querystart+1,
0,width,
gapopen,gapextend,
path,sub,cvt,
compass,show);
if(score>minscore)
{
embAlignWalkSWMatrixFast(path,compass,gapopen,gapextend,
targetseq,queryseq,
&targetaln,&queryaln,
targetend-targetstart+1,
queryend-querystart+1,
0,width,
&targetstart,&querystart);
if(!ajAlignFormatShowsSequences(align))
{
ajAlignDefineCC(align, ajStrGetPtr(targetaln),
ajStrGetPtr(queryaln),
ajSeqGetNameC(targetseq),
ajSeqGetNameC(queryseq));
ajAlignSetScoreR(align, score);
}
else
{
ajDebug(" queryaln:%S \ntargetaln:%S\n",
queryaln,targetaln);
embAlignReportLocal(align,
queryseq, targetseq,
queryaln, targetaln,
querystart, targetstart,
gapopen, gapextend,
score, matrix,
1 + ajSeqGetOffset(queryseq),
1 + ajSeqGetOffset(targetseq)
);
}
ajAlignWrite(align);
ajAlignReset(align);
}
ajStrDel(&targetaln);
embWordMatchListDelete(&matchlist[k]);
}
ajStrDel(&queryaln);
}
for(k=0;k<nkmers;k++)
{
AJFREE(wordsw[k]->seqindxs);
AJFREE(wordsw[k]->nSeqMatches);
for(j=0;j<wordsw[k]->nseqs;j++)
AJFREE(wordsw[k]->locs[j]);
AJFREE(wordsw[k]->nnseqlocs);
AJFREE(wordsw[k]->locs);
AJFREE(wordsw[k]);
}
embWordFreeTable(&kmers);
if(!ajAlignFormatShowsSequences(align))
ajMatrixfDel(&matrix);
AJFREE(path);
AJFREE(compass);
AJFREE(kmers);
AJFREE(wordsw);
AJFREE(matchlist);
AJFREE(lastlocation);
ajAlignClose(align);
ajAlignDel(&align);
ajSeqallDel(&queryseqs);
ajSeqDel(&queryseq);
ajSeqsetDel(&targetseqs);
ajFileClose(&errorf);
embExit();
return 0;
}
int main(int argc, char **argv)
{
AjPSeqall seqall;
AjPSeq sequence;
AjPFile outfile;
ajint minLen;
ajint maxLen;
ajint maxGap;
ajint beginPos;
ajint endPos;
ajint maxmismatches;
AjPStr seqstr;
ajint current;
ajint rev;
ajint count;
ajint gap;
ajint begin;
ajint end;
ajint mismatches;
ajint mismatchAtEnd;
ajint istart;
ajint iend;
ajint ic;
ajint ir;
AjBool alln; /* TRUE if all of palindrome is N's */
Palindrome pfirstpal;
Palindrome plastpal = NULL;
Palindrome ppal = NULL;
Palindrome pnext = NULL;
AjBool found = AJFALSE;
embInit("palindrome", argc, argv);
seqall = ajAcdGetSeqall("sequence");
minLen = ajAcdGetInt("minpallen");
maxLen = ajAcdGetInt("maxpallen");
maxGap = ajAcdGetInt("gaplimit");
outfile = ajAcdGetOutfile("outfile");
maxmismatches = ajAcdGetInt("nummismatches");
overlap = ajAcdGetBoolean("overlap");
while(ajSeqallNext(seqall, &sequence))
{
beginPos = ajSeqallGetseqBegin(seqall);
endPos = ajSeqallGetseqEnd(seqall);
/* set to NULL to indicate that we have no first palindrome find yet */
pfirstpal = NULL;
/* write header to file */
ajFmtPrintF(outfile, "Palindromes of: %s \n", ajSeqGetNameC(sequence));
ajFmtPrintF(outfile, "Sequence length is: %d \n", ajSeqGetLen(sequence));
ajFmtPrintF(outfile, "Start at position: %d\nEnd at position: %d\n",
beginPos, endPos);
ajFmtPrintF(outfile,"Minimum length of Palindromes is: %d \n", minLen);
ajFmtPrintF(outfile,"Maximum length of Palindromes is: %d \n", maxLen);
ajFmtPrintF(outfile,"Maximum gap between elements is: %d \n", maxGap);
ajFmtPrintF(outfile,"Number of mismatches allowed in Palindrome: %d\n",
maxmismatches);
ajFmtPrintF(outfile, "\n\n\n");
ajFmtPrintF(outfile, "Palindromes:\n");
/* set vars in readiness to enter loop */
seqstr = ajStrNewC(ajSeqGetSeqC(sequence));
begin = beginPos - 1;
end = endPos - 1;
ajStrFmtLower(&seqstr); /* make comparisons case independent */
/* loop to look for inverted repeats */
for(current = begin; current < end; current++)
{
iend = current + 2*(maxLen) + maxGap;
if(iend > end)
iend = end;
istart = current + minLen;
for(rev = iend; rev > istart; rev--)
{
count = 0;
mismatches = 0;
mismatchAtEnd = 0;
alln = ajTrue;
ic = current;
ir = rev;
if(ajStrGetCharPos(seqstr, ic) ==
ajBaseAlphacharComp(ajStrGetCharPos(seqstr, ir)))
while(mismatches <= maxmismatches && ic < ir)
{
if(ajStrGetCharPos(seqstr, ic++) ==
ajBaseAlphacharComp(ajStrGetCharPos(seqstr, ir--)))
{
mismatchAtEnd = 0;
if(ajStrGetCharPos(seqstr, ic-1) != 'n')
alln = ajFalse;
}
else
{
mismatches++;
mismatchAtEnd++;
}
count++;
}
count -= mismatchAtEnd;
gap = rev - current - count - count + 1;
/* Find out if there's a reverse repeat long enough */
if(count >= minLen && gap <= maxGap && !alln)
{
/* create new struct to hold palindrome data */
ppal = palindrome_New(current,(current+count),rev,
(rev-count));
/*
** if it is the first palindrome find then save it as start
** of palindrome list
*/
if(pfirstpal == NULL)
{
pfirstpal = ppal;
plastpal = ppal;
}
else
{
/* Is it a subset of a palindrome already met */
pnext = pfirstpal;
found = AJFALSE;
while(pnext != NULL)
{
if(overlap && palindrome_AInB(ppal, pnext))
{
found = AJTRUE;
break;
}
if(!overlap && palindrome_AOverB(ppal, pnext))
{
if(palindrome_Longer(ppal, pnext))
{
palindrome_Swap(ppal, pnext);
}
found = AJTRUE;
break;
}
pnext = pnext->next;
}
/* if new palindrome add to end of list */
if(!found)
{
plastpal->next = ppal;
plastpal = ppal;
}
else
AJFREE(ppal);
}
}
}
}
/* Print out palindromes */
ppal = pfirstpal;
while(ppal != NULL)
{
palindrome_Print(outfile, seqstr, ppal, maxLen);
ppal = ppal->next;
}
/* make a gap beween outputs of different sequences */
ajFmtPrintF(outfile, "\n\n\n");
/* free memory used for palindrome list */
ppal = pfirstpal;
while(ppal != NULL)
{
pnext = ppal->next;
AJFREE(ppal);
ppal = pnext;
}
ajStrDel(&seqstr);
}
ajFileClose(&outfile);
ajSeqallDel(&seqall);
ajSeqDel(&sequence);
ajStrDel(&seqstr);
embExit();
return 0;
}
static void primersearch_store_hits(const Primer primdata,
AjPList fhits, AjPList rhits,
const AjPSeq seq, AjBool reverse)
{
ajint amplen = 0;
AjIList fi;
AjIList ri;
PHit primerhit = NULL;
fi = ajListIterNewread(fhits);
while(!ajListIterDone(fi))
{
EmbPMatMatch fm = NULL;
EmbPMatMatch rm = NULL;
amplen = 0;
fm = ajListIterGet(fi);
ri = ajListIterNewread(rhits);
while(!ajListIterDone(ri))
{
ajint seqlen = ajSeqGetLen(seq);
ajint s = (fm->start);
ajint e;
rm = ajListIterGet(ri);
e = (rm->start-1);
amplen = seqlen-(s-1)-e;
if (amplen > 0) /* no point making a hit if -ve length! */
{
primerhit = NULL;
AJNEW(primerhit);
primerhit->desc=NULL; /* must be NULL for ajStrAss */
primerhit->seqname=NULL; /* must be NULL for ajStrAss */
primerhit->acc=NULL;
primerhit->forward=NULL;
primerhit->reverse=NULL;
ajStrAssignC(&primerhit->seqname,ajSeqGetNameC(seq));
ajStrAssignS(&primerhit->desc, ajSeqGetDescS(seq));
ajStrAssignS(&primerhit->acc, ajSeqGetAccS(seq));
primerhit->forward_pos = fm->start;
primerhit->reverse_pos = rm->start;
primerhit->forward_mismatch = fm->mm;
primerhit->reverse_mismatch = rm->mm;
primerhit->amplen = amplen;
if(!reverse)
{
ajStrAssignS(&primerhit->forward,
primdata->forward->patstr);
ajStrAssignS(&primerhit->reverse,
primdata->reverse->patstr);
}
else
{
ajStrAssignS(&primerhit->forward,
primdata->reverse->patstr);
ajStrAssignS(&primerhit->reverse,
primdata->forward->patstr);
}
ajListPushAppend(primdata->hitlist, primerhit);
}
}
/*
** clean up rListIter here as it will be new'ed again next
** time through
*/
ajListIterDel(&ri);
}
ajListIterDel(&fi);
return;
}
static void infoalign_Compare(const AjPSeq ref, const AjPSeq seq,
ajint * const *sub,
const AjPSeqCvt cvt, ajint *seqlength,
ajint *alignlength, ajint *gaps,
ajint *gapcount, ajint *idcount,
ajint *simcount, ajint *difcount,
float *change)
{
ajint i;
ajint lenseq;
ajint lenref;
const char *s;
const char *r;
AjBool inGap = ajFalse; /* true if in a gap in 'seq' */
ajint begin;
ajint end;
lenseq = ajSeqGetLen(seq);
lenref = ajSeqGetLen(ref);
s = ajSeqGetSeqC(seq);
r = ajSeqGetSeqC(ref);
/* initialise counts */
*seqlength = 0;
*alignlength = 0;
*gaps = 0;
*gapcount = 0;
*idcount = 0;
*simcount = 0;
*difcount = 0;
*change = 0.0;
/* ignore gaps at the ends of the sequence */
for(begin = 0; s[begin] == '-'; begin++);
for(end = lenseq-1; s[end] == '-'; end--);
for(i=begin; i<=end; i++)
{
/* count gaps and their length */
if(s[i] == '-')
{
if(!inGap)
{
inGap = ajTrue;
(*gaps)++;
}
(*gapcount)++;
}
else
{
inGap = ajFalse;
/*
** count identity, similarity, differences
** Past the end of the reference sequence ?
*/
if(i >= lenref)
(*difcount)++;
else
{
/* identity */
if((toupper((int)r[i]) == toupper((int)s[i])))
(*idcount)++;
/* similarity */
else if(sub[ajSeqcvtGetCodeK(cvt, r[i])][ajSeqcvtGetCodeK(cvt, s[i])] > 0)
(*simcount)++;
/* difference */
else
(*difcount)++;
}
}
}
*seqlength = *idcount + *simcount + *difcount;
*alignlength = end-begin+1;
*change = (float)(*alignlength - *idcount)*(float)100.0/
(float)(*alignlength);
return;
}
int main(int argc, char **argv)
{
AjPSeqset seqset;
const AjPSeq seq1;
const AjPSeq seq2;
ajint wordlen;
AjPTable seq1MatchTable = NULL;
AjPList matchlist ;
AjPGraph graph = 0;
ajuint i;
ajuint j;
float total=0;
ajuint acceptableticks[]=
{
1,10,50,100,200,500,1000,1500,10000,50000,
100000,500000,1000000,5000000
};
ajint numbofticks = 10;
ajint gap,tickgap;
AjBool boxit = AJTRUE;
AjBool dumpfeat = AJFALSE;
float xmargin;
float ymargin;
float k;
char ptr[10];
float ticklen;
float onefifth;
AjPFeattable *tabptr = NULL;
AjPFeattabOut seq1out = NULL;
AjPStr sajb = NULL;
float flen1;
float flen2;
ajuint tui;
embInit("polydot", argc, argv);
wordlen = ajAcdGetInt("wordsize");
seqset = ajAcdGetSeqset("sequences");
graph = ajAcdGetGraph("graph");
gap = ajAcdGetInt("gap");
boxit = ajAcdGetBoolean("boxit");
seq1out = ajAcdGetFeatout("outfeat");
dumpfeat = ajAcdGetToggle("dumpfeat");
sajb = ajStrNew();
embWordLength(wordlen);
AJCNEW(lines,ajSeqsetGetSize(seqset));
AJCNEW(pts,ajSeqsetGetSize(seqset));
AJCNEW(tabptr,ajSeqsetGetSize(seqset));
for(i=0;i<ajSeqsetGetSize(seqset);i++)
{
seq1 = ajSeqsetGetseqSeq(seqset, i);
total += ajSeqGetLen(seq1);
}
total +=(float)(gap*(ajSeqsetGetSize(seqset)-1));
xmargin = total*(float)0.15;
ymargin = total*(float)0.15;
ticklen = xmargin*(float)0.1;
onefifth = xmargin*(float)0.2;
i = 0;
while(acceptableticks[i]*numbofticks < ajSeqsetGetLen(seqset))
i++;
if(i<=13)
tickgap = acceptableticks[i];
else
tickgap = acceptableticks[13];
ajGraphAppendTitleS(graph, ajSeqsetGetUsa(seqset));
ajGraphOpenWin(graph, (float)0.0-xmargin,(total+xmargin)*(float)1.35,
(float)0.0-ymargin,
total+ymargin);
ajGraphicsSetCharscale((float)0.3);
for(i=0;i<ajSeqsetGetSize(seqset);i++)
{
which = i;
seq1 = ajSeqsetGetseqSeq(seqset, i);
tui = ajSeqGetLen(seq1);
flen1 = (float) tui;
if(embWordGetTable(&seq1MatchTable, seq1)){ /* get table of words */
for(j=0;j<ajSeqsetGetSize(seqset);j++)
{
seq2 = ajSeqsetGetseqSeq(seqset, j);
tui = ajSeqGetLen(seq2);
flen2 = (float) tui;
if(boxit)
ajGraphicsDrawposRect(xstart,ystart,
xstart+flen1,
ystart+flen2);
matchlist = embWordBuildMatchTable(seq1MatchTable, seq2,
ajTrue);
if(matchlist)
polydot_plotMatches(matchlist);
if(i<j && dumpfeat)
embWordMatchListConvToFeat(matchlist,&tabptr[i],
&tabptr[j],seq1,
seq2);
if(matchlist) /* free the match structures */
embWordMatchListDelete(&matchlist);
if(j==0)
{
for(k=0.0;k<ajSeqGetLen(seq1);k+=tickgap)
{
ajGraphicsDrawposLine(xstart+k,ystart,xstart+k,
ystart-ticklen);
sprintf(ptr,"%d",(ajint)k);
ajGraphicsDrawposTextAtmid(xstart+k,
ystart-(onefifth),
ptr);
}
ajGraphicsDrawposTextAtmid(
xstart+(flen1/(float)2.0),
ystart-(3*onefifth),
ajStrGetPtr(ajSeqsetGetseqNameS(seqset, i)));
}
if(i==0)
{
for(k=0.0;k<ajSeqGetLen(seq2);k+=tickgap)
{
ajGraphicsDrawposLine(xstart,ystart+k,xstart-ticklen,
ystart+k);
sprintf(ptr,"%d",(ajint)k);
ajGraphicsDrawposTextAtend(xstart-(onefifth),
ystart+k,
ptr);
}
ajGraphicsDrawposTextAtlineJustify(
xstart-(3*onefifth),
ystart+(flen2/(float)2.0),
xstart-(3*onefifth),ystart+flen2,
ajStrGetPtr(ajSeqsetGetseqNameS(seqset, j)),0.5);
}
ystart += flen2+(float)gap;
}
}
embWordFreeTable(&seq1MatchTable);
seq1MatchTable = NULL;
xstart += flen1+(float)gap;
ystart = 0.0;
}
ajGraphicsDrawposTextAtstart(total+onefifth,total-(onefifth),
"No. Length Lines Points Sequence");
for(i=0;i<ajSeqsetGetSize(seqset);i++)
{
seq1 = ajSeqsetGetseqSeq(seqset, i);
ajFmtPrintS(&sajb,"%3u %6d %5d %6d %s",i+1,
ajSeqGetLen(seq1),lines[i],
pts[i],ajSeqGetNameC(seq1));
ajGraphicsDrawposTextAtstart(total+onefifth,total-(onefifth*(i+2)),
ajStrGetPtr(sajb));
}
if(dumpfeat && seq1out)
{
for(i=0;i<ajSeqsetGetSize(seqset);i++)
{
ajFeattableWrite(seq1out, tabptr[i]);
ajFeattableDel(&tabptr[i]);
}
}
ajGraphicsClose();
ajGraphxyDel(&graph);
ajStrDel(&sajb);
AJFREE(lines);
AJFREE(pts);
AJFREE(tabptr);
ajSeqsetDel(&seqset);
ajFeattabOutDel(&seq1out);;
embExit();
return 0;
}
int main(int argc, char *argv[])
{
char *string/*, *line*/;
char *structure=NULL, *cstruc=NULL;
/*char fname[13], ffname[20], gfname[20];*/
/*char *ParamFile=NULL;*/
char *ns_bases=NULL, *c;
int i, length, l, sym/*, r*/;
double energy, min_en;
double kT, sfact=1.07;
int pf=0, noPS=0, istty;
int noconv=0;
int circ=0;
AjPSeq seq = NULL;
AjPFile confile = NULL;
AjPFile paramfile = NULL;
AjPFile outf = NULL;
AjPFile essfile = NULL;
AjPFile dotfilea = NULL;
AjPFile dotfileb = NULL;
AjPStr seqstring = NULL;
AjPStr constring = NULL;
AjPStr seqname = NULL;
float eT = 0.;
AjBool eGU;
AjBool ecirc = ajFalse;
AjBool eclose;
AjBool lonely;
AjBool convert;
AjPStr ensbases = NULL;
AjBool etloop;
AjPStr eenergy = NULL;
char ewt = '\0';
float escale = 0.;
AjPStr edangles = NULL;
char edangle = '\0';
ajint len;
embInitPV("vrnafold",argc,argv,"VIENNA",VERSION);
seqstring = ajStrNew();
constring = ajStrNew();
seqname = ajStrNew();
seq = ajAcdGetSeq("sequence");
confile = ajAcdGetInfile("constraintfile");
paramfile = ajAcdGetInfile("paramfile");
eT = ajAcdGetFloat("temperature");
ecirc = ajAcdGetBoolean("circular");
eGU = ajAcdGetBoolean("gu");
eclose = ajAcdGetBoolean("closegu");
lonely = ajAcdGetBoolean("lp");
convert = ajAcdGetBoolean("convert");
ensbases = ajAcdGetString("nsbases");
etloop = ajAcdGetBoolean("tetraloop");
eenergy = ajAcdGetListSingle("energy");
escale = ajAcdGetFloat("scale");
edangles = ajAcdGetListSingle("dangles");
outf = ajAcdGetOutfile("outfile");
essfile = ajAcdGetOutfile("ssoutfile");
/*
dotfilea = ajAcdGetOutfile("adotoutfile");
dotfileb = ajAcdGetOutfile("bdotoutfile");
*/
do_backtrack = 2;
pf = 0;
string = NULL;
istty = 0;
temperature = (double) eT;
circ = !!ecirc;
noGU = (eGU) ? 0 : 1;
no_closingGU = (eclose) ? 0 : 1;
noLonelyPairs = (lonely) ? 0 : 1;
noconv = (convert) ? 0 : 1;
ns_bases = (ajStrGetLen(ensbases)) ? MAJSTRGETPTR(ensbases) : NULL;
tetra_loop = !!etloop;
ewt = *ajStrGetPtr(eenergy);
if(ewt == '0')
energy_set = 0;
else if(ewt == '1')
energy_set = 1;
else if(ewt == '2')
energy_set = 2;
sfact = (double) escale;
edangle = *ajStrGetPtr(edangles);
if(edangle == '0')
dangles = 0;
else if(edangle == '1')
dangles = 1;
else if(edangle == '2')
dangles = 2;
else if(edangle == '3')
dangles = 3;
if(circ && noLonelyPairs)
{
ajWarn("Depending on the origin of the circular sequence\n"
"some structures may be missed when using -noLP\nTry "
"rotating your sequence a few times\n");
}
if(paramfile)
read_parameter_file(paramfile);
if (ns_bases != NULL)
{
nonstandards = space(33);
c=ns_bases;
i=sym=0;
if (*c=='-')
{
sym=1; c++;
}
while (*c!='\0')
{
if (*c!=',')
{
nonstandards[i++]=*c++;
nonstandards[i++]=*c;
if ((sym)&&(*c!=*(c-1)))
{
nonstandards[i++]=*c;
nonstandards[i++]=*(c-1);
}
}
c++;
}
}
if(confile)
vienna_GetConstraints(confile,&constring);
string = NULL;
structure = NULL;
length = ajSeqGetLen(seq);
string = (char *) space(length+1);
strcpy(string,ajSeqGetSeqC(seq));
len = ajStrGetLen(constring);
structure = (char *) space(length+1);
if(len)
{
fold_constrained = 1;
strcpy(structure,ajStrGetPtr(constring));
}
for (l = 0; l < length; l++) {
string[l] = toupper(string[l]);
if (!noconv && string[l] == 'T') string[l] = 'U';
}
/* initialize_fold(length); */
if (circ)
min_en = circfold(string, structure);
else
min_en = fold(string, structure);
ajFmtPrintF(outf,"%s\n%s", string, structure);
if (istty)
printf("\n minimum free energy = %6.2f kcal/mol\n", min_en);
else
ajFmtPrintF(outf," (%6.2f)\n", min_en);
if (!noPS)
{
if (length<2000)
(void) PS_rna_plot(string, structure, essfile);
else
ajWarn("Structure too long, not doing xy_plot\n");
}
if (length>=2000) free_arrays();
if (pf)
{
char *pf_struc;
pf_struc = (char *) space((unsigned) length+1);
if (dangles==1)
{
dangles=2; /* recompute with dangles as in pf_fold() */
min_en = (circ) ? energy_of_circ_struct(string, structure) :
energy_of_struct(string, structure);
dangles=1;
}
kT = (temperature+273.15)*1.98717/1000.; /* in Kcal */
pf_scale = exp(-(sfact*min_en)/kT/length);
if (length>2000)
ajWarn("scaling factor %f\n", pf_scale);
(circ) ? init_pf_circ_fold(length) : init_pf_fold(length);
if (cstruc!=NULL)
strncpy(pf_struc, cstruc, length+1);
energy = (circ) ? pf_circ_fold(string, pf_struc) :
pf_fold(string, pf_struc);
if (do_backtrack)
{
ajFmtPrintF(outf,"%s", pf_struc);
ajFmtPrintF(outf," [%6.2f]\n", energy);
}
if ((istty)||(!do_backtrack))
ajFmtPrintF(outf," free energy of ensemble = %6.2f kcal/mol\n",
energy);
if (do_backtrack)
{
plist *pl1,*pl2;
char *cent;
double dist, cent_en;
cent = centroid(length, &dist);
cent_en = (circ) ? energy_of_circ_struct(string, cent) :
energy_of_struct(string, cent);
ajFmtPrintF(outf,"%s {%6.2f d=%.2f}\n", cent, cent_en, dist);
free(cent);
pl1 = make_plist(length, 1e-5);
pl2 = b2plist(structure);
(void) PS_dot_plot_list(string, dotfilea, pl1, pl2, "");
free(pl2);
if (do_backtrack==2)
{
pl2 = stackProb(1e-5);
PS_dot_plot_list(string, dotfileb, pl1, pl2,
"Probabilities for stacked pairs (i,j)(i+1,j-1)");
free(pl2);
}
free(pl1);
free(pf_struc);
}
ajFmtPrintF(outf," frequency of mfe structure in ensemble %g; ",
exp((energy-min_en)/kT));
if (do_backtrack)
ajFmtPrintF(outf,"ensemble diversity %-6.2f", mean_bp_dist(length));
ajFmtPrintF(outf,"\n");
free_pf_arrays();
}
if (cstruc!=NULL)
free(cstruc);
free(string);
free(structure);
ajStrDel(&seqstring);
ajStrDel(&constring);
ajStrDel(&seqname);
ajStrDel(&ensbases);
ajStrDel(&eenergy);
ajStrDel(&edangles);
ajSeqDel(&seq);
ajFileClose(&confile);
ajFileClose(¶mfile);
ajFileClose(&outf);
ajFileClose(&essfile);
/*
ajFileClose(&dotfilea);
ajFileClose(&dotfileb);
*/
if (length<2000) free_arrays();
embExit();
return 0;
}
static AjBool ssematch_NWScore(AjPScop temp_scop,
AjPSeq pseq,
ajint mode,
AjPMatrixf matrix,
float gapopen,
float gapextend)
{
ajint start1 =0; /* Start of seq 1, passed as arg but not used.*/
ajint start2 =0; /* Start of seq 2, passed as arg but not used.*/
ajint maxarr =300; /* Initial size for matrix. */
ajint len;
ajint *compass;
const char *p; /* Query sequence. */
const char *q; /* Subject sequence from scop object. */
float **sub;
float id =0.; /* Passed as arg but not used here. */
float sim =0.;
float idx =0.; /* Passed as arg but not used here. */
float simx =0.; /* Passed as arg but not used here. */
float *path;
AjPStr pstr = NULL; /* m walk alignment for first sequence
Passed as arg but not used here. */
AjPStr qstr = NULL; /* n walk alignment for second sequence
Passed as arg but not used here. */
AjPSeq qseq = NULL; /* Subject sequence. */
ajint lenp; /* Length of query sequence. */
ajint lenq; /* Length of subject sequence. */
AjPSeqCvt cvt = 0;
AjBool show = ajFalse; /*Passed as arg but not used here. */
AJCNEW(path, maxarr);
AJCNEW(compass, maxarr);
pstr = ajStrNew();
qstr = ajStrNew();
gapopen = ajRoundFloat(gapopen,8);
gapextend = ajRoundFloat(gapextend,8);
sub = ajMatrixfGetMatrix(matrix);
cvt = ajMatrixfGetCvt(matrix);
/* Extract subject sequence from scop object, convert to 3 letter code. */
if (mode == 0)
qseq = ssematch_convertbases(temp_scop->Sse);
else if (mode == 1)
qseq = ssematch_convertbases(temp_scop->Sss);
lenp = ajSeqGetLen(pseq); /* Length of query sequence. */
lenq = ajSeqGetLen(qseq); /* Length of subject sequence. */
/* Start of main application loop */
/* Intitialise variables for use by alignment functions*/
len = (lenp * lenq);
if(len>maxarr)
{
AJCRESIZE(path,len);
AJCRESIZE(compass,len);
maxarr=len;
}
p = ajSeqGetSeqC(pseq);
q = ajSeqGetSeqC(qseq);
ajStrAssignC(&pstr,"");
ajStrAssignC(&qstr,"");
/* Check that no sequence length is 0. */
if((lenp == 0)||(lenq == 0))
{
AJFREE(compass);
AJFREE(path);
ajStrDel(&pstr);
ajStrDel(&qstr);
}
/* Call alignment functions. */
embAlignPathCalc(p,q,lenp,lenq, gapopen,
gapextend,path,sub,cvt,compass,show);
/*embAlignScoreNWMatrix(path,compass,gapopen,gapextend,
pseq, qseq,
lenp,lenq,sub,cvt,
&start1,&start2);*/
embAlignWalkNWMatrix(path,pseq,qseq,&pstr,&qstr,
lenp,lenq,&start1,&start2,
gapopen,gapextend,compass);
embAlignCalcSimilarity(pstr,qstr,sub,cvt,lenp,
lenq,&id,&sim,&idx, &simx);
/* Assign score. */
temp_scop->Score = sim;
/* Tidy up */
AJFREE(compass);
AJFREE(path);
ajStrDel(&pstr);
ajStrDel(&qstr);
ajSeqDel(&qseq);
/* Bye Bye */
return ajTrue;
}
int main(int argc, char **argv)
{
AjPAlign align;
AjPSeq a;
AjPSeq b;
AjPSeqout seqout;
AjPStr m;
AjPStr n;
AjPStr merged = NULL;
ajuint lena;
ajuint lenb;
const char *p;
const char *q;
ajint start1 = 0;
ajint start2 = 0;
float *path;
ajint *compass;
AjPMatrixf matrix;
AjPSeqCvt cvt = 0;
float **sub;
float gapopen;
float gapextend;
ajulong maxarr = 1000;
ajulong len; /* arbitrary. realloc'd if needed */
size_t stlen;
float score;
ajint begina;
ajint beginb;
embInit("merger", argc, argv);
a = ajAcdGetSeq("asequence");
b = ajAcdGetSeq("bsequence");
seqout = ajAcdGetSeqout("outseq");
matrix = ajAcdGetMatrixf("datafile");
gapopen = ajAcdGetFloat("gapopen");
gapextend = ajAcdGetFloat("gapextend");
align = ajAcdGetAlign("outfile");
gapopen = ajRoundFloat(gapopen, 8);
gapextend = ajRoundFloat(gapextend, 8);
AJCNEW(path, maxarr);
AJCNEW(compass, maxarr);
/*
** make the two sequences lowercase so we can show which one we are
** using in the merge by uppercasing it
*/
ajSeqFmtLower(a);
ajSeqFmtLower(b);
m = ajStrNew();
n = ajStrNew();
sub = ajMatrixfGetMatrix(matrix);
cvt = ajMatrixfGetCvt(matrix);
begina = ajSeqGetBegin(a);
beginb = ajSeqGetBegin(b);
lena = ajSeqGetLen(a);
lenb = ajSeqGetLen(b);
if(lenb > (ULONG_MAX/(ajulong)(lena+1)))
ajFatal("Sequences too big. Try 'supermatcher'");
len = lena*lenb;
if(len>maxarr)
{
ajDebug("merger: resize path, len to %d (%d * $d)\n",
len, lena, lenb);
stlen = (size_t) len;
AJCRESIZE(path,stlen);
AJCRESIZE(compass,stlen);
maxarr=len;
}
p = ajSeqGetSeqC(a);
q = ajSeqGetSeqC(b);
ajStrAssignC(&m,"");
ajStrAssignC(&n,"");
score = embAlignPathCalc(p,q,lena,lenb,gapopen,gapextend,path,sub,cvt,
compass, ajFalse);
/*score = embAlignScoreNWMatrix(path,compass,gapopen,gapextend,
a,b,lena,lenb,sub,cvt,
&start1,&start2);*/
embAlignWalkNWMatrix(path,a,b,&m,&n,lena,lenb, &start1,&start2,gapopen,
gapextend,compass);
/*
** now construct the merged sequence, uppercase the bits of the two
** input sequences which are used in the merger
*/
merger_Merge(align, &merged,p,q,m,n,start1,start2,
ajSeqGetNameC(a),ajSeqGetNameC(b));
embAlignReportGlobal(align, a, b, m, n,
start1, start2, gapopen, gapextend,
score, matrix, begina, beginb);
ajAlignWrite(align);
ajAlignReset(align);
/* write the merged sequence */
ajSeqAssignSeqS(a, merged);
ajSeqoutWriteSeq(seqout, a);
ajSeqoutClose(seqout);
ajSeqoutDel(&seqout);
ajSeqDel(&a);
ajSeqDel(&b);
ajAlignClose(align);
ajAlignDel(&align);
ajStrDel(&merged);
AJFREE(compass);
AJFREE(path);
ajStrDel(&n);
ajStrDel(&m);
embExit();
return 0;
}
static ajint trimest_get_tail(const AjPSeq seq, ajint direction,
ajint minlength, ajint mismatches)
{
char t;
const char *s;
char c;
ajint inc;
ajint start;
ajint end;
ajint i;
ajint mismatchcount; /* number of contiguous mismatches */
ajint polycount; /* length of poly-A/T since end/last mismatch */
ajint length; /* length of tail looked at so far */
ajint result; /* resulting length of tail */
if(direction == 5)
{
t = 'T';
inc = 1;
start = 0;
end = ajSeqGetLen(seq);
}
else
{
t = 'A';
inc = -1;
start = ajSeqGetLen(seq)-1;
end = -1;
}
s = ajSeqGetSeqC(seq);
mismatchcount = 0;
polycount = 0;
length = 0;
result = 0;
for(i = start; i != end; i += inc, length++)
{
c = toupper((int)s[i]);
ajDebug("end = %d, c=%c\n", direction, c);
if(c == t)
{
polycount++;
mismatchcount = 0;
}
else if(c != 'N')
{
/*
** There is a mismatch
** N is ignored - it is not a poly-tail or a mismatch
*/
polycount = 0;
mismatchcount++;
}
if(polycount >= minlength)
result = length+1;
ajDebug("end = %d, polycount = %d, so far tail=%d\n",
direction, polycount, result);
if(mismatchcount > mismatches)
break;
}
return result;
}
static void extractfeat_GetRegionPad(const AjPSeq seq, AjPStr *featstr,
ajint start, ajint end, AjBool sense,
AjBool beginning)
{
ajint tmp;
ajint pad;
AjPStr result;
ajDebug("In extractfeat_GetRegionPad start=%d, end=%d\n", start, end);
result = ajStrNew();
if(start > end)
return;
if(start < 0)
{
pad = -start;
if(ajSeqIsNuc(seq))
ajStrAppendCountK(&result, 'N', pad);
else
ajStrAppendCountK(&result, 'X', pad);
start = 0;
}
if(end > (ajint) ajSeqGetLen(seq)-1)
tmp = ajSeqGetLen(seq)-1;
else
tmp = end;
if(start <= (ajint) ajSeqGetLen(seq) && tmp >= 0)
{
ajDebug("Get subsequence %d-%d\n", start, tmp);
ajStrAppendSubS(&result, ajSeqGetSeqS(seq), start, tmp);
ajDebug("result=%S\n", result);
}
if(end > (ajint) ajSeqGetLen(seq)-1)
{
pad = end - ajSeqGetLen(seq)+1;
if(ajSeqIsNuc(seq))
ajStrAppendCountK(&result, 'N', pad);
else
ajStrAppendCountK(&result, 'X', pad);
ajDebug("result=%S\n", result);
}
/* if feature was in reverse sense, then get reverse complement */
if(!sense)
{
ajDebug("get reverse sense of subsequence\n");
ajSeqstrReverse(&result);
ajDebug("result=%S\n", result);
}
if(beginning)
{
ajDebug("Prepend to featstr: %S\n", result);
ajStrInsertS(featstr, 0, result);
}
else
{
ajDebug("Append to featstr: %S\n", result);
ajStrAppendS(featstr, result);
}
ajDebug("featstr=%S\n", *featstr);
ajStrDel(&result);
return;
}
int main(int argc, char **argv)
{
AjPSeqall seqall;
AjPSeqout seqout;
AjPSeq seq = NULL;
AjPStr str = NULL;
AjPStr desc = NULL;
ajint tail3;
ajint tail5 = 0;
ajint minlength;
ajint mismatches;
AjBool reverse;
AjBool fiveprime;
AjBool cvttolower;
embInit("trimest", argc, argv);
seqall = ajAcdGetSeqall("sequence");
seqout = ajAcdGetSeqoutall("outseq");
minlength = ajAcdGetInt("minlength");
mismatches = ajAcdGetInt("mismatches");
reverse = ajAcdGetBoolean("reverse");
fiveprime = ajAcdGetBoolean("fiveprime");
cvttolower = ajAcdGetToggle("tolower");
str = ajStrNew();
while(ajSeqallNext(seqall, &seq))
{
/* get sequence description */
ajStrAssignS(&desc, ajSeqGetDescS(seq));
/* get positions to cut in 5' poly-T and 3' poly-A tails */
if(fiveprime)
tail5 = trimest_get_tail(seq, 5, minlength, mismatches);
tail3 = trimest_get_tail(seq, 3, minlength, mismatches);
/* get a COPY of the sequence string */
ajStrAssignS(&str, ajSeqGetSeqS(seq));
/* cut off longest of 3' or 5' tail */
if(tail5 > tail3)
{
/* if 5' poly-T tail, then reverse the sequence */
ajDebug("Tail=%d\n", tail5);
if(cvttolower)
trimest_tolower(&str, 0, tail5-1);
else
ajStrKeepRange(&str, tail5, ajSeqGetLen(seq)-1);
ajStrAppendC(&desc, " [poly-T tail removed]");
}
else if(tail3 > tail5)
{
/* remove 3' poly-A tail */
ajDebug("Tail=%d\n", tail3);
if(cvttolower)
trimest_tolower(&str, ajSeqGetLen(seq)-tail3,
ajSeqGetLen(seq));
else
ajStrKeepRange(&str, 0, ajSeqGetLen(seq)-tail3-1);
ajStrAppendC(&desc, " [poly-A tail removed]");
}
/* write sequence out */
ajSeqAssignSeqS(seq, str);
/* reverse complement if poly-T found */
if(tail5 > tail3 && reverse)
{
ajSeqReverseForce(seq);
ajStrAppendC(&desc, " [reverse complement]");
}
/* set description */
ajSeqAssignDescS(seq, desc);
ajSeqoutWriteSeq(seqout, seq);
}
ajSeqoutClose(seqout);
ajStrDel(&str);
ajStrDel(&desc);
ajSeqallDel(&seqall);
ajSeqDel(&seq);
ajSeqoutDel(&seqout);
embExit();
return 0;
}
static ajint supermatcher_findstartpoints(AjPTable seq1MatchTable,
const AjPSeq b,
const AjPSeq a, ajint *start1,
ajint *start2, ajint *end1,
ajint *end2)
{
ajint max = -10;
ajint offset = 0;
AjPList matchlist = NULL;
AjPList ordered = NULL;
ajint amax;
ajint bmax;
ajint bega;
ajint begb;
amax = ajSeqGetLen(a)-1;
bmax = ajSeqGetLen(b)-1;
bega = ajSeqGetOffset(a);
begb = ajSeqGetOffset(b);
ajDebug("supermatcher_findstartpoints len %d %d off %d %d\n",
amax, bmax, bega, begb);
matchlist = embWordBuildMatchTable(seq1MatchTable, b, ajTrue);
if(!matchlist)
return 0;
else if(!matchlist->Count)
{
embWordMatchListDelete(&matchlist);
return 0;
}
/* order and add if the gap is gapmax or less */
/* create list header bit*/
ordered = ajListNew();
supermatcher_orderandconcat(matchlist, ordered);
/* this sets global structure conmax to point to a matchlist element */
ajListMap(ordered,supermatcher_findmax, &max);
ajDebug("findstart conmax off:%d count:%d total:%d\n",
conmax->offset, conmax->count, conmax->total,
ajListGetLength(conmax->list));
offset = conmax->offset;
/* the offset is all we needed! we can delete everything */
ajListMap(ordered,supermatcher_removelists, NULL);
ajListFree(&ordered);
embWordMatchListDelete(&matchlist); /* free the match structures */
if(offset > 0)
{
*start1 = offset;
*start2 = 0;
}
else
{
*start2 = 0-offset;
*start1 = 0;
}
*end1 = *start1;
*end2 = *start2;
ajDebug("++ end1 %d -> %d end2 %d -> %d\n", *end1, amax, *end2, bmax);
while(*end1<amax && *end2<bmax)
{
(*end1)++;
(*end2)++;
}
ajDebug("++ end1 %d end2 %d\n", *end1, *end2);
ajDebug("supermatcher_findstartpoints has %d..%d [%d] %d..%d [%d]\n",
*start1, *end1, ajSeqGetLen(a), *start2, *end2, ajSeqGetLen(b));
return 1;
}
void embPatternRegexSearch (AjPFeattable ftable, const AjPSeq seq,
const AjPPatternRegex pat, AjBool reverse)
{
ajint pos=0;
ajint off;
ajint len;
AjPFeature sf = NULL;
AjPStr substr = NULL;
AjPStr seqstr = NULL;
AjPStr tmpstr = NULL;
AjPStr tmp = ajStrNew();
AjPRegexp patexp = ajPatternRegexGetCompiled(pat);
ajint adj;
AjBool isreversed;
AjPSeq revseq;
ajint seqlen;
seqlen = ajSeqGetLen(seq);
if(!seqlen)
return;
isreversed = ajSeqIsReversedTrue(seq);
if(isreversed)
seqlen += ajSeqGetOffset(seq);
pos = ajSeqGetBeginTrue(seq);
adj = ajSeqGetEndTrue(seq);
if(!ajStrGetLen(featMotifProt))
ajStrAssignC(&featMotifProt, "SO:0001067");
if(!ajStrGetLen(featMotifNuc))
ajStrAssignC(&featMotifNuc, "SO:0000714");
/*ajDebug("embPatternRegexSearch pos: %d adj: %d reverse: %B\n",
pos, adj, reverse, isreversed);*/
/*ajDebug("seqlen:%d len: %d offset: %d offend: %d begin: %d end: %d\n",
seqlen , ajSeqGetLen(seq), ajSeqGetOffset(seq),
ajSeqGetOffend(seq), ajSeqGetBegin(seq), ajSeqGetEnd(seq));*/
if (reverse)
{
revseq = ajSeqNewSeq(seq);
ajStrAssignSubS(&seqstr, ajSeqGetSeqS(revseq), pos-1, adj-1);
ajSeqstrReverse(&seqstr);
}
ajStrAssignSubS(&seqstr, ajSeqGetSeqS(seq), pos-1, adj-1);
ajStrFmtUpper(&seqstr);
while(ajStrGetLen(seqstr) && ajRegExec(patexp, seqstr))
{
off = ajRegOffset(patexp);
len = ajRegLenI(patexp, 0);
if(off || len)
{
ajRegSubI(patexp, 0, &substr);
ajRegPost(patexp, &tmp);
ajStrAssignS(&seqstr, substr);
ajStrAppendS(&seqstr, tmp);
pos += off;
/*ajDebug("match pos: %d adj: %d len: %d off:%d\n",
pos, adj, len, off);*/
if (reverse)
sf = ajFeatNew(ftable, NULL, featMotifNuc,
adj - pos - len + 2,
adj - pos + 1,
0.0, '-', 0);
else
{
if(ajSeqIsProt(seq) || ajFeattableIsProt(ftable))
sf = ajFeatNewProt(ftable, NULL, featMotifProt,
pos, pos + len - 1,
0.0);
else
sf = ajFeatNew(ftable, NULL, featMotifNuc,
pos, pos + len - 1,
0.0, '.', 0);
}
if(isreversed)
ajFeatReverse(sf, seqlen);
ajFmtPrintS (&tmpstr,"*pat %S: %S",
ajPatternRegexGetName(pat),
ajPatternRegexGetPattern(pat));
ajFeatTagAdd (sf,NULL,tmpstr);
pos += 1;
ajStrCutStart(&seqstr, 1);
}
else
{
pos++;
ajStrCutStart(&seqstr, 1);
}
}
ajStrDel(&tmpstr);
ajStrDel(&tmp);
ajStrDel(&substr);
ajStrDel(&seqstr);
if(reverse)
ajSeqDel(&revseq);
return;
}
int main(int argc, char **argv)
{
AjPList list = NULL;
AjPSeq seq;
AjPSeq seq2;
AjPStr aa0str = 0;
AjPStr aa1str = 0;
const char *s1;
const char *s2;
char *strret = NULL;
ajuint i;
ajuint j;
ajuint k;
ajint l;
ajint abovethresh;
ajint total;
ajint starti = 0;
ajint startj = 0;
ajint windowsize;
float thresh;
AjPGraph graph = NULL;
AjPGraph xygraph = NULL;
float flen1;
float flen2;
ajuint len1;
ajuint len2;
AjPTime ajtime = NULL;
time_t tim;
AjBool boxit=AJTRUE;
/* Different ticks as they need to be different for x and y due to
length of string being important on x */
ajuint acceptableticksx[]=
{
1,10,50,100,500,1000,1500,10000,
500000,1000000,5000000
};
ajuint acceptableticks[]=
{
1,10,50,100,200,500,1000,2000,5000,10000,15000,
500000,1000000,5000000
};
ajint numbofticks = 10;
float xmargin;
float ymargin;
float ticklen;
float tickgap;
float onefifth;
float k2;
float max;
char ptr[10];
AjPMatrix matrix = NULL;
ajint** sub;
AjPSeqCvt cvt;
AjPStr subt = NULL;
ajint b1;
ajint b2;
ajint e1;
ajint e2;
AjPStr se1;
AjPStr se2;
ajint ithresh;
AjBool stretch;
PPoint ppt = NULL;
float xa[1];
float ya[1];
AjPGraphdata gdata=NULL;
AjPStr tit = NULL;
AjIList iter = NULL;
float x1 = 0.;
float x2 = 0.;
float y1 = 0.;
float y2 = 0.;
ajuint tui;
se1 = ajStrNew();
se2 = ajStrNew();
embInit("dotmatcher", argc, argv);
seq = ajAcdGetSeq("asequence");
seq2 = ajAcdGetSeq("bsequence");
stretch = ajAcdGetToggle("stretch");
graph = ajAcdGetGraph("graph");
xygraph = ajAcdGetGraphxy("xygraph");
windowsize = ajAcdGetInt("windowsize");
ithresh = ajAcdGetInt("threshold");
matrix = ajAcdGetMatrix("matrixfile");
sub = ajMatrixGetMatrix(matrix);
cvt = ajMatrixGetCvt(matrix);
thresh = (float)ithresh;
ajtime = ajTimeNew();
tim = time(0);
ajTimeSetLocal(ajtime, tim);
b1 = ajSeqGetBegin(seq);
b2 = ajSeqGetBegin(seq2);
e1 = ajSeqGetEnd(seq);
e2 = ajSeqGetEnd(seq2);
len1 = ajSeqGetLen(seq);
len2 = ajSeqGetLen(seq2);
tui = ajSeqGetLen(seq);
flen1 = (float) tui;
tui = ajSeqGetLen(seq2);
flen2 = (float) tui;
ajStrAssignSubC(&se1,ajSeqGetSeqC(seq),b1-1,e1-1);
ajStrAssignSubC(&se2,ajSeqGetSeqC(seq2),b2-1,e2-1);
ajSeqAssignSeqS(seq,se1);
ajSeqAssignSeqS(seq2,se2);
s1 = ajStrGetPtr(ajSeqGetSeqS(seq));
s2 = ajStrGetPtr(ajSeqGetSeqS(seq2));
aa0str = ajStrNewRes(1+len1); /* length plus trailing blank */
aa1str = ajStrNewRes(1+len2);
list = ajListNew();
for(i=0;i<len1;i++)
ajStrAppendK(&aa0str,(char)ajSeqcvtGetCodeK(cvt, *s1++));
for(i=0;i<len2;i++)
ajStrAppendK(&aa1str,(char)ajSeqcvtGetCodeK(cvt, *s2++));
max = (float)len1;
if(len2 > max)
max = (float) len2;
xmargin = ymargin = max *(float)0.15;
ticklen = xmargin*(float)0.1;
onefifth = xmargin*(float)0.2;
subt = ajStrNewC((strret=
ajFmtString("(windowsize = %d, threshold = %3.2f %D)",
windowsize,thresh,ajtime)));
if(!stretch)
{
if( ajStrGetLen(ajGraphGetSubtitleS(graph)) <=1)
ajGraphSetSubtitleS(graph,subt);
ajGraphOpenWin(graph, (float)0.0-ymargin,(max*(float)1.35)+ymargin,
(float)0.0-xmargin,(float)max+xmargin);
ajGraphicsDrawposTextAtmid(flen1*(float)0.5,
(float)0.0-(xmargin/(float)2.0),
ajGraphGetXlabelC(graph));
ajGraphicsDrawposTextAtlineJustify((float)0.0-(xmargin*(float)0.75),
flen2*(float)0.5,
(float)0.0-(xmargin*(float)0.75),flen1,
ajGraphGetYlabelC(graph),0.5);
ajGraphicsSetCharscale(0.5);
}
s1= ajStrGetPtr(aa0str);
s2 = ajStrGetPtr(aa1str);
for(j=0; (ajint)j < (ajint)len2-windowsize;j++)
{
i =0;
total = 0;
abovethresh =0;
k = j;
for(l=0;l<windowsize;l++)
total = total + sub[(ajint)s1[i++]][(ajint)s2[k++]];
if(total >= thresh)
{
abovethresh=1;
starti = i-windowsize;
startj = k-windowsize;
}
while(i < len1 && k < len2)
{
total = total - sub[(ajint)s1[i-windowsize]]
[(ajint)s2[k-windowsize]];
total = total + sub[(ajint)s1[i]][(ajint)s2[k]];
if(abovethresh)
{
if(total < thresh)
{
abovethresh = 0;
/* draw the line */
dotmatcher_pushpoint(&list,(float)starti,(float)startj,
(float)i-1,(float)k-1,stretch);
}
}
else if(total >= thresh)
{
starti = i-windowsize;
startj = k-windowsize;
abovethresh= 1;
}
i++;
k++;
}
if(abovethresh)
/* draw the line */
dotmatcher_pushpoint(&list,(float)starti,(float)startj,
(float)i-1,(float)k-1,
stretch);
}
for(i=0; (ajint)i < (ajint)len1-windowsize;i++)
{
j = 0;
total = 0;
abovethresh =0;
k = i;
for(l=0;l<windowsize;l++)
total = total + sub[(ajint)s1[k++]][(ajint)s2[j++]];
if(total >= thresh)
{
abovethresh=1;
starti = k-windowsize;
startj = j-windowsize;
}
while(k < len1 && j < len2)
{
total = total - sub[(ajint)s1[k-windowsize]]
[(ajint)s2[j-windowsize]];
total = total + sub[(ajint)s1[k]][(ajint)s2[j]];
if(abovethresh)
{
if(total < thresh)
{
abovethresh = 0;
/* draw the line */
dotmatcher_pushpoint(&list,(float)starti,(float)startj,
(float)k-1,(float)j-1,stretch);
}
}
else if(total >= thresh)
{
starti = k-windowsize;
startj = j-windowsize;
abovethresh= 1;
}
j++;
k++;
}
if(abovethresh)
/* draw the line */
dotmatcher_pushpoint(&list,(float)starti,(float)startj,
(float)k-1,(float)j-1,
stretch);
}
if(boxit && !stretch)
{
ajGraphicsDrawposRect(0.0,0.0,flen1, flen2);
i=0;
while(acceptableticksx[i]*numbofticks < len1)
i++;
if(i<=13)
tickgap = (float)acceptableticksx[i];
else
tickgap = (float)acceptableticksx[10];
ticklen = xmargin*(float)0.1;
onefifth = xmargin*(float)0.2;
if(len2/len1 > 10 )
{
/* if a lot smaller then just label start and end */
ajGraphicsDrawposLine((float)0.0,(float)0.0,(float)0.0,(float)0.0-ticklen);
sprintf(ptr,"%d",b1-1);
ajGraphicsDrawposTextAtmid((float)0.0,(float)0.0-(onefifth),ptr);
ajGraphicsDrawposLine(flen1,(float)0.0,
flen1,(float)0.0-ticklen);
sprintf(ptr,"%d",len1+b1-1);
ajGraphicsDrawposTextAtmid(flen1,(float)0.0-(onefifth),ptr);
}
else
for(k2=0.0;k2<len1;k2+=tickgap)
{
ajGraphicsDrawposLine(k2,(float)0.0,k2,(float)0.0-ticklen);
sprintf(ptr,"%d",(ajint)k2+b1-1);
ajGraphicsDrawposTextAtmid(k2,(float)0.0-(onefifth),ptr);
}
i = 0;
while(acceptableticks[i]*numbofticks < len2)
i++;
tickgap = (float)acceptableticks[i];
ticklen = ymargin*(float)0.01;
onefifth = ymargin*(float)0.02;
if(len1/len2 > 10 )
{
/* if a lot smaller then just label start and end */
ajGraphicsDrawposLine((float)0.0,(float)0.0,(float)0.0-ticklen,(float)0.0);
sprintf(ptr,"%d",b2-1);
ajGraphicsDrawposTextAtend((float)0.0-(onefifth),(float)0.0,ptr);
ajGraphicsDrawposLine((float)0.0,flen2,(float)0.0-ticklen,
flen2);
sprintf(ptr,"%d",len2+b2-1);
ajGraphicsDrawposTextAtend((float)0.0-(onefifth),flen2,ptr);
}
else
for(k2=0.0;k2<len2;k2+=tickgap)
{
ajGraphicsDrawposLine((float)0.0,k2,(float)0.0-ticklen,k2);
sprintf(ptr,"%d",(ajint)k2+b2-1);
ajGraphicsDrawposTextAtend((float)0.0-(onefifth),k2,ptr);
}
}
if(!stretch)
ajGraphicsClose();
else /* the xy graph for -stretch */
{
tit = ajStrNew();
ajFmtPrintS(&tit,"%S",ajGraphGetTitleS(xygraph));
gdata = ajGraphdataNewI(1);
xa[0] = (float)b1;
ya[0] = (float)b2;
ajGraphSetTitleC(xygraph,ajStrGetPtr(tit));
ajGraphSetXlabelC(xygraph,ajSeqGetNameC(seq));
ajGraphSetYlabelC(xygraph,ajSeqGetNameC(seq2));
ajGraphdataSetTypeC(gdata,"2D Plot Float");
ajGraphdataSetTitleS(gdata,subt);
ajGraphdataSetMinmax(gdata,(float)b1,(float)e1,(float)b2,
(float)e2);
ajGraphdataSetTruescale(gdata,(float)b1,(float)e1,(float)b2,
(float)e2);
ajGraphxySetXstartF(xygraph,(float)b1);
ajGraphxySetXendF(xygraph,(float)e1);
ajGraphxySetYstartF(xygraph,(float)b2);
ajGraphxySetYendF(xygraph,(float)e2);
ajGraphxySetXrangeII(xygraph,b1,e1);
ajGraphxySetYrangeII(xygraph,b2,e2);
if(list)
{
iter = ajListIterNewread(list);
while((ppt = ajListIterGet(iter)))
{
x1 = ppt->x1+b1-1;
y1 = ppt->y1+b2-1;
x2 = ppt->x2+b1-1;
y2 = ppt->y2+b2-1;
ajGraphAddLine(xygraph,x1,y1,x2,y2,0);
AJFREE(ppt);
}
ajListIterDel(&iter);
}
ajGraphdataAddXY(gdata,xa,ya);
ajGraphDataReplace(xygraph,gdata);
ajGraphxyDisplay(xygraph,ajFalse);
ajGraphicsClose();
ajStrDel(&tit);
}
ajListFree(&list);
ajSeqDel(&seq);
ajSeqDel(&seq2);
ajGraphxyDel(&graph);
ajGraphxyDel(&xygraph);
ajMatrixDel(&matrix);
ajTimeDel(&ajtime);
/* deallocate memory */
ajStrDel(&aa0str);
ajStrDel(&aa1str);
ajStrDel(&se1);
ajStrDel(&se2);
ajStrDel(&subt);
AJFREE(strret); /* created withing ajFmtString */
embExit();
return 0;
}
void embPatternSeqSearch (AjPFeattable ftable, const AjPSeq seq,
const AjPPatternSeq pat, AjBool reverse)
{
const void *tidy;
ajuint hits;
ajuint i;
AjPPatComp pattern;
EmbPMatMatch m = NULL;
AjPFeature sf = NULL;
AjPSeq revseq = NULL;
AjPList list = ajListNew();
AjPStr seqstr = ajStrNew();
AjPStr seqname = ajStrNew();
AjPStr tmp = ajStrNew();
ajint adj;
ajint begin;
AjBool isreversed;
ajint seqlen;
seqlen = ajSeqGetLen(seq);
if(!seqlen)
return;
isreversed = ajSeqIsReversedTrue(seq);
if(isreversed)
seqlen += ajSeqGetOffset(seq);
begin = ajSeqGetBeginTrue(seq);
adj = ajSeqGetEndTrue(seq);
if(!ajStrGetLen(featMotifProt))
ajStrAssignC(&featMotifProt, "SO:0001067");
if(!ajStrGetLen(featMotifNuc))
ajStrAssignC(&featMotifNuc, "SO:0000714");
ajStrAssignS(&seqname,ajSeqGetNameS(seq));
pattern = ajPatternSeqGetCompiled(pat);
if (reverse)
{
revseq = ajSeqNewSeq(seq);
ajStrAssignSubS(&seqstr, ajSeqGetSeqS(revseq),
begin-1,adj-1);
ajSeqstrReverse(&seqstr);
}
else
ajStrAssignSubS(&seqstr, ajSeqGetSeqS(seq),
begin-1,adj-1);
ajStrFmtUpper(&seqstr);
/*ajDebug("seqlen:%d len: %d offset: %d offend: %d begin: %d end: %d\n"
"'%S'\n",
seqlen , ajSeqGetLen(seq), ajSeqGetOffset(seq),
ajSeqGetOffend(seq), ajSeqGetBegin(seq), ajSeqGetEnd(seq),
seqstr);*/
ajDebug("embPatternSeqSearch '%S' protein: %B reverse: %B\n",
pattern->pattern, pat->Protein, reverse);
embPatFuzzSearchII(pattern,begin,seqname,seqstr,list,
ajPatternSeqGetMismatch(pat),&hits,&tidy);
ajDebug ("embPatternSeqSearch: found %d hits\n",hits);
if(!reverse)
ajListReverse(list);
for(i=0;i<hits;++i)
{
ajListPop(list,(void **)&m);
if (reverse)
sf = ajFeatNew(ftable, NULL, featMotifNuc,
adj - m->start - m->len + begin + 1,
adj - m->start + begin,
0.0, '-', 0);
else
{
if(ajSeqIsProt(seq) || ajFeattableIsProt(ftable))
sf = ajFeatNewProt(ftable, NULL, featMotifProt,
m->start,
m->start + m->len - 1,
0.0);
else
sf = ajFeatNew(ftable, NULL, featMotifNuc,
m->start,
m->start + m->len - 1,
0.0, '.', 0);
}
if(isreversed)
ajFeatReverse(sf, seqlen);
/*
ajUser("isrev: %B reverse: %B begin: %d adj: %d "
"start: %d len: %d seqlen: %d %d..%d '%c'\n",
isreversed, reverse, begin, adj, m->start, m->len, seqlen,
sf->Start, sf->End, sf->Strand);
*/
ajFeatSetScore(sf, (float) (m->len - m->mm));
ajFmtPrintS(&tmp, "*pat %S: %S",
ajPatternSeqGetName(pat),
ajPatternSeqGetPattern(pat));
ajFeatTagAdd(sf,NULL,tmp);
if(m->mm)
{
ajFmtPrintS(&tmp, "*mismatch %d", m->mm);
ajFeatTagAdd(sf, NULL, tmp);
}
embMatMatchDel(&m);
}
ajStrDel(&seqname);
ajStrDel(&seqstr);
ajStrDel(&tmp);
ajListFree(&list);
if (reverse)
ajSeqDel(&revseq);
return;
}
int main(int argc, char **argv)
{
AjPSeqall seqall;
AjPSeqout seqout;
AjPSeq seq;
ajint size;
ajint overlap;
ajint len;
ajint pos;
AjBool addover;
AjBool feature;
AjPStr outseq_name = ajStrNew();
ajint start;
ajint end;
embInit("splitter", argc, argv);
seqout = ajAcdGetSeqoutall("outseq");
seqall = ajAcdGetSeqall("sequence");
size = ajAcdGetInt("size");
overlap = ajAcdGetInt("overlap");
addover = ajAcdGetBoolean("addoverlap");
feature = ajAcdGetBoolean("feature");
while(ajSeqallNext(seqall, &seq))
{
ajSeqTrim(seq);
len = ajSeqGetLen(seq);
pos = 0;
ajStrAssignC(&outseq_name, "");
if (!addover)
{
while(pos+size <= len-1)
{
start = pos;
end = pos+size-1;
splitter_MakeSubSeqName (&outseq_name, seq, start, end);
splitter_ProcessChunk (seqout, seq, start, end,
outseq_name, feature);
pos += size-overlap;
}
}
else
{
while(pos+size+overlap < len-1)
{
start = pos;
end = pos+size+overlap-1;
splitter_MakeSubSeqName (&outseq_name, seq, start, end);
splitter_ProcessChunk (seqout, seq, start, end,
outseq_name, feature);
pos += size;
}
}
splitter_MakeSubSeqName(&outseq_name, seq, pos, len-1);
splitter_ProcessChunk (seqout, seq, pos, len-1,
outseq_name, feature);
}
ajSeqoutClose(seqout);
ajSeqallDel(&seqall);
ajSeqoutDel(&seqout);
ajSeqDel(&seq);
ajStrDel(&outseq_name);
embExit();
return 0;
}
int main(int argc, char *argv[])
{
char *sequence;
char *structure = NULL;
char *ns_bases = NULL, *c;
int i, length, l, sym;
int istty;
double deltap=0.;
int delta=100;
int n_back = 0;
int noconv=0;
int circ=0;
int dos=0;
AjPSeq seq = NULL;
AjPFile confile = NULL;
AjPFile paramfile = NULL;
AjPFile outf = NULL;
AjPStr constring = NULL;
float eT = 0.;
AjBool eGU;
AjBool eclose;
AjBool lonely;
AjBool convert;
AjPStr ensbases = NULL;
AjBool etloop;
AjPStr edangles = NULL;
char edangle = '\0';
ajint len;
float erange;
float prange;
embInitPV("vrnasubopt",argc,argv,"VIENNA",VERSION);
constring = ajStrNew();
seq = ajAcdGetSeq("sequence");
confile = ajAcdGetInfile("constraintfile");
paramfile = ajAcdGetInfile("paramfile");
eT = ajAcdGetFloat("temperature");
circ = !!ajAcdGetBoolean("circular");
dos = !!ajAcdGetBoolean("dos");
eGU = ajAcdGetBoolean("gu");
eclose = ajAcdGetBoolean("closegu");
lonely = ajAcdGetBoolean("lp");
convert = ajAcdGetBoolean("convert");
ensbases = ajAcdGetString("nsbases");
etloop = ajAcdGetBoolean("tetraloop");
erange = ajAcdGetFloat("erange");
prange = ajAcdGetFloat("prange");
subopt_sorted = !!ajAcdGetBoolean("sort");
logML = !!ajAcdGetBoolean("logml");
n_back = ajAcdGetInt("nrandom");
edangles = ajAcdGetListSingle("dangles");
outf = ajAcdGetOutfile("outfile");
if(dos)
print_energy = -999999;
do_backtrack = 1;
istty = 0;
temperature = (double) eT;
noGU = (eGU) ? 0 : 1;
no_closingGU = (eclose) ? 0 : 1;
noLonelyPairs = (lonely) ? 0 : 1;
noconv = (convert) ? 0 : 1;
ns_bases = (ajStrGetLen(ensbases)) ? MAJSTRGETPTR(ensbases) : NULL;
tetra_loop = !!etloop;
delta = (int) (0.1 + erange * 100);
deltap = prange;
edangle = *ajStrGetPtr(edangles);
if(edangle == '0')
dangles = 0;
else if(edangle == '1')
dangles = 1;
else if(edangle == '2')
dangles = 2;
else if(edangle == '3')
dangles = 3;
if(paramfile)
read_parameter_file(paramfile);
if (ns_bases != NULL)
{
nonstandards = space(33);
c=ns_bases;
i=sym=0;
if (*c=='-')
{
sym=1;
c++;
}
while (*c)
{
if (*c!=',')
{
nonstandards[i++]=*c++;
nonstandards[i++]=*c;
if ((sym)&&(*c!=*(c-1)))
{
nonstandards[i++]=*c;
nonstandards[i++]=*(c-1);
}
}
c++;
}
}
if(confile)
vienna_GetConstraints(confile,&constring);
if(n_back)
init_rand();
sequence = NULL;
structure = NULL;
length = ajSeqGetLen(seq);
sequence = (char *) space(length+1);
strcpy(sequence,ajSeqGetSeqC(seq));
len = ajStrGetLen(constring);
structure = (char *) space(length+1);
if(len)
{
fold_constrained = 1;
strcpy(structure,ajStrGetPtr(constring));
}
istty = 0;
if (fold_constrained)
{
for (i=0; i<length; i++)
if (structure[i]=='|')
ajFatal("Constraints of type '|' are not allowed\n");
}
for (l = 0; l < length; l++)
{
sequence[l] = toupper(sequence[l]);
if (!noconv && sequence[l] == 'T')
sequence[l] = 'U';
}
if ((logML!=0 || dangles==1 || dangles==3) && dos==0)
if (deltap<=0) deltap=delta/100. +0.001;
if (deltap>0)
print_energy = deltap;
/* first lines of output (suitable for sort +1n) */
ajFmtPrintF(outf,"> %s [%d]\n", ajSeqGetNameC(seq), delta);
if(n_back>0)
{
int i;
double mfe, kT;
char *ss;
st_back=1;
ss = (char *) space(strlen(sequence)+1);
strncpy(ss, structure, length);
mfe = (circ) ? circfold(sequence, ss) : fold(sequence, ss);
kT = (temperature+273.15)*1.98717/1000.; /* in Kcal */
pf_scale = exp(-(1.03*mfe)/kT/length);
strncpy(ss, structure, length);
/*
** we are not interested in the free energy but in the bppm, so we
** drop free energy into the void
*/
(circ) ? (void) pf_circ_fold(sequence, ss) :
(void) pf_fold(sequence, ss);
free(ss);
for (i=0; i<n_back; i++)
{
char *s;
s = (circ) ? pbacktrack_circ(sequence) : pbacktrack(sequence);
ajFmtPrintF(outf,"%s\n", s);
free(s);
}
free_pf_arrays();
}
else
{
(circ) ? subopt_circ(sequence, structure, delta, ajFileGetFileptr(outf)) :
subopt(sequence, structure, delta, ajFileGetFileptr(outf));
}
free(sequence);
free(structure);
ajSeqDel(&seq);
ajStrDel(&ensbases);
ajStrDel(&edangles);
ajFileClose(&confile);
ajFileClose(&outf);
ajFileClose(¶mfile);
embExit();
return 0;
}