static void silent_split_hits(AjPList *hits, AjPList *silents,
AjPList *nonsilents, AjBool allmut)
{
PSilent res;
while(ajListPop(*hits,(void **)&res))
{
if(res->issilent)
{
ajListPushAppend(*silents,(void *)res);
continue;
}
if(allmut)
ajListPushAppend(*nonsilents,(void *)res);
else
{
ajStrDel(&res->code);
ajStrDel(&res->site);
ajStrDel(&res->seqaa);
ajStrDel(&res->reaa);
AJFREE(res);
}
}
return;
}
static void prima_prune_nearby(AjPList pairlist, ajint *npair, ajint range)
{
PPair pair;
ajint count;
ajint fst;
ajint fst2;
ajint blim;
ajint blim2;
ajint elim;
ajint elim2;
ajint i;
ajint j;
ajint len;
for(i=0;i<*npair;++i)
{
for(j=0;j<i;++j) /* Ignore those already processed */
{
ajListPop(pairlist,(void **)&pair);
ajListPushAppend(pairlist,(void *)pair);
}
ajListPop(pairlist,(void **)&pair);/* Get next high scoring pair */
len = pair->f->primerlen;
fst = pair->f->start + len -1;
blim = fst-range;
elim = fst+range;
len = pair->r->primerlen;
blim2 = pair->r->start - range;
elim2 = pair->r->start + range;
ajListPushAppend(pairlist,(void *)pair);
count = *npair;
for(j=i+1;j<*npair;++j)
{
ajListPop(pairlist,(void **)&pair);
fst2=pair->f->start+pair->f->primerlen-1;
if((fst2<blim || fst2>elim) && (pair->r->start<blim2 ||
pair->r->start>elim2))
ajListPushAppend(pairlist,(void *)pair);
else
{
prima_PrimerDel(&pair->f);
prima_PrimerDel(&pair->r);
AJFREE(pair);
--count;
}
}
*npair = count;
}
return;
}
static void jaspscan_GetFileList(const AjPStr dir, const AjPStr jaspdir,
const char *wild, AjPList list)
{
AjPList tlist = NULL;
AjPStr jdir = NULL;
AjPStr wstr = NULL;
AjPStr str = NULL;
tlist = ajListNew();
jdir = ajStrNew();
wstr = ajStrNew();
ajFmtPrintS(&jdir,"%S%S",dir,jaspdir);
if(!ajDirnameFixExists(&jdir))
ajFatal("EMBOSS_DATA undefined or 'jaspextract' needs to be run");
ajFmtPrintS(&wstr,"%s%s",wild,J_EXT);
ajFilelistAddPathWild(tlist,jdir,wstr);
if(!ajListGetLength(tlist))
ajWarn("Matrix file(s) %S not found",wstr);
while(ajListPop(tlist,(void **)&str))
ajListPushAppend(list,(void *)str);
ajStrDel(&wstr);
ajStrDel(&jdir);
ajListFree(&tlist);
return;
}
static void jaspextract_getjaspdirs(AjPList jdirlist)
{
const AjPStr datadir = NULL;
AjPStr line = NULL;
ajuint n = 0;
ajuint i = 0;
datadir = ajDatafileValuePath();
if(!datadir)
ajFatal("jaspextract: Cannot determine the EMBOSS data directory");
n = ajFilelistAddPathDir(jdirlist, datadir);
for(i=0; i < n; ++i)
{
ajListPop(jdirlist,(void **)&line);
if(ajStrFindC(line,"JASPAR_") != -1)
ajListPushAppend(jdirlist,(void *)line);
else
ajStrDel(&line);
}
return;
}
static void assemblyexceptionadaptorFetchAll(const void *key,
void **value,
void *cl)
{
AjIList iter = NULL;
EnsPAssemblyexception ae = NULL;
if(!key)
return;
if(!value)
return;
if(!*value)
return;
if(!cl)
return;
iter = ajListIterNew(*((AjPList *) value));
while(!ajListIterDone(iter))
{
ae = (EnsPAssemblyexception) ajListIterGet(iter);
ajListPushAppend((AjPList) cl,
(void *) ensAssemblyexceptionNewRef(ae));
}
ajListIterDel(&iter);
return;
}
static void prima_TwoSortscorepos(AjPList *pairlist)
{
PPair tmp = NULL;
AjPList intlist = NULL;
AjPList filist = NULL;
PPair save = NULL;
float score = 0.0;
ajListSort(*pairlist,prima_Compare);
intlist = ajListNew();
filist = ajListNew();
score = (float) -1.0;
while(ajListPop(*pairlist,(void **)&tmp))
{
if(tmp->f->score == score)
{
ajListPush(intlist,(void *)tmp);
continue;
}
save = tmp;
ajListSort(intlist,prima_PosCompare);
score = tmp->f->score;
prima_RevSort(&intlist);
while(ajListPop(intlist,(void **)&tmp))
ajListPushAppend(filist,(void *)tmp);
ajListPush(intlist,(void *)save);
}
ajListSort(intlist,prima_PosCompare);
prima_RevSort(&intlist);
while(ajListPop(intlist,(void **)&tmp))
ajListPushAppend(filist,(void *)tmp);
ajListFree(&intlist);
ajListFree(pairlist);
*pairlist = filist;
return;
}
void ajPatlistAddSeq (AjPPatlistSeq thys, AjPPatternSeq pat)
{
ajDebug ("ajPatlistAddSeq list size %d '%S' '%S' '%B' '%d'\n",
ajListGetLength (thys->Patlist), pat->Name,
pat->Pattern, pat->Protein, pat->Mismatch);
ajListPushAppend(thys->Patlist, pat);
return;
}
void ajPatlistAddRegex (AjPPatlistRegex thys, AjPPatternRegex pat)
{
ajDebug ("ajPatlistAddRegex list size %d '%S' '%S' '%d'\n",
ajListGetLength (thys->Patlist), pat->Name,
pat->Pattern, pat->Type);
ajListPushAppend(thys->Patlist, pat);
return;
}
static void supermatcher_matchListOrder(void **x,void *cl)
{
EmbPWordMatch p;
AjPList ordered;
ajint offset;
AjIList listIter;
concat *con;
concat *c=NULL;
p = (EmbPWordMatch)*x;
ordered = (AjPList) cl;
offset = (*p).seq1start-(*p).seq2start;
/* iterate through ordered list to find if it exists already*/
listIter = ajListIterNewread(ordered);
while(!ajListIterDone( listIter))
{
con = ajListIterGet(listIter);
if(con->offset == offset)
{
/* found so add count and set offset to the new value */
con->offset = offset;
con->total+= (*p).length;
con->count++;
ajListPushAppend(con->list,p);
ajListIterDel(&listIter);
return;
}
}
ajListIterDel(&listIter);
/* not found so add it */
AJNEW(c);
c->offset = offset;
c->total = (*p).length;
c->count = 1;
c->list = ajListNew();
ajListPushAppend(c->list,p);
ajListPushAppend(ordered, c);
return;
}
static void prima_check_overlap(AjPList pairlist, ajint *npair, ajint overlap)
{
PPair pair;
ajint i;
ajint j;
ajint end;
ajint limit;
ajint count;
for(i=0;i<*npair;++i)
{
for(j=0;j<i;++j)
{
ajListPop(pairlist,(void **)&pair);
ajListPushAppend(pairlist,(void *)pair);
}
ajListPop(pairlist,(void **)&pair);
end = pair->r->start;
limit = end-overlap;
ajListPushAppend(pairlist,(void *)pair);
count = *npair;
for(j=i+1;j<*npair;++j)
{
ajListPop(pairlist,(void **)&pair);
if(pair->f->start+pair->f->primerlen-1 < limit)
{
prima_PrimerDel(&pair->f);
prima_PrimerDel(&pair->r);
AJFREE(pair);
--count;
}
else
ajListPushAppend(pairlist,(void *)pair);
}
*npair = count;
}
return;
}
static void acdrelations_readtypefile
(AjPFile inf,
PKtype *T)
{
AjPStr line = NULL;
PKtypedat dattmp = NULL;
AjPList datlist = NULL;
if(!T)
ajFatal("Null arg error 1 in acdrelations_readtypefile");
if(!inf)
ajFatal("Null arg error 3 in acdrelations_readtypefile");
/* Allocate memory */
line = ajStrNew();
datlist = ajListNew();
/* Read data from file */
while(ajReadline(inf,&line))
{
/* Discard comment lines */
if(ajStrPrefixC(line,"#"))
continue;
/* Create object for holding line */
dattmp = ajKtypedatNew();
/* Tokenise line delimited by '|'
Parse first token (value of knowntype: attribute) */
ajStrAssignS(&dattmp->ktype, ajStrParseC(line, "|"));
ajStrRemoveSetC(&dattmp->ktype, "_");
ajStrRemoveWhite(&dattmp->ktype);
/* Parse second token (ACD datatype) */
ajStrAssignS(&dattmp->acdtype, ajStrParseC(NULL, "|"));
/* Parse third token (EDAM relations: value ) */
ajStrAssignS(&dattmp->edam, ajStrParseC(NULL, "|"));
/* Push line onto list */
ajListPushAppend(datlist, dattmp);
}
/* Write PKtype structure */
((*T)->n) = ajListToarray(datlist, (void***) &((*T)->dat));
/* Free memory */
ajStrDel(&line);
ajListFree(&datlist);
return;
}
static void prima_RevSort(AjPList *alist)
{
PPair tmp = NULL;
AjPList intlist = NULL;
AjPList filist = NULL;
PPair save = NULL;
ajint pos = -1;
intlist = ajListNew();
filist = ajListNew();
pos = -1;
while(ajListPop(*alist,(void **)&tmp))
{
if(tmp->f->start+tmp->f->primerlen == pos)
{
ajListPush(intlist,(void *)tmp);
continue;
}
save = tmp;
ajListSort(intlist,prima_PosEndCompare);
pos = tmp->f->start+tmp->f->primerlen;
while(ajListPop(intlist,(void **)&tmp))
ajListPushAppend(filist,(void *)tmp);
ajListPush(intlist,(void *)save);
}
ajListSort(intlist,prima_PosEndCompare);
while(ajListPop(intlist,(void **)&tmp))
ajListPushAppend(filist,(void *)tmp);
ajListFree(&intlist);
ajListFree(alist);
*alist = filist;
return;
}
/* @funcstatic skipredundant_SeqsetToList **************************************
**
** Builds a list of sequences from a sequence set.
** The sequences are NOT copied (only a reference is pushed onto the list)
**
** @param [u] list [AjPList] List
** @param [w] seqset [AjPSeqset] Sequence set
** @return [AjBool] True on success
******************************************************************************/
static AjBool skipredundant_SeqsetToList (AjPList list, AjPSeqset seqset)
{
ajint n = 0;
ajint x = 0;
EmbPDmxNrseq seq_tmp = NULL; /* Temp. pointer for making seq_list. */
if(!list || !seqset)
return ajFalse;
n = ajSeqsetGetSize(seqset);
for(x=0; x<n; x++)
{
seq_tmp = embDmxNrseqNew(ajSeqsetGetseqSeq(seqset, x));
ajListPushAppend(list, seq_tmp);
seq_tmp = NULL;
}
return ajTrue;
}
AjBool ensAssemblyexceptionadaptorFetchAllBySeqregionIdentifier(
const EnsPAssemblyexceptionadaptor aea,
ajuint srid,
AjPList aes)
{
AjIList iter = NULL;
AjPList list = NULL;
EnsPAssemblyexception ae = NULL;
if(ajDebugTest("ensAssemblyexceptionadaptorFetchAllBySeqregionIdentifier"))
ajDebug("ensAssemblyexceptionadaptorFetchAllBySeqregionIdentifier\n"
" aea %p\n"
" srid %u\n"
" aes %p\n",
aea,
srid,
aes);
if(!aea)
return ajFalse;
if(!aes)
return ajFalse;
list = (AjPList) ajTableFetch(aea->CacheBySeqregionIdentifier,
(const void *) &srid);
iter = ajListIterNew(list);
while(!ajListIterDone(iter))
{
ae = (EnsPAssemblyexception) ajListIterGet(iter);
ajListPushAppend(aes, (void *) ensAssemblyexceptionNewRef(ae));
}
ajListIterDel(&iter);
return ajTrue;
}
AjBool ajResquerylistClone(const AjPList src, AjPList dest)
{
AjIList iter;
AjPResquery qryout = NULL;
AjPResquery qryin = NULL;
if(ajListGetLength(dest))
return ajFalse;
iter = ajListIterNewread(src);
while ((qryin = (AjPResquery) ajListIterGet(iter)))
{
qryout = ajResqueryNewResquery(qryin);
ajListPushAppend(dest, qryout);
}
ajListIterDel(&iter);
return ajTrue;
}
AjBool ajRestermlistClone(const AjPList src, AjPList dest)
{
AjIList iter;
AjPResterm termout = NULL;
AjPResterm termin = NULL;
if(ajListGetLength(dest))
return ajFalse;
iter = ajListIterNewread(src);
while ((termin = (AjPResterm) ajListIterGet(iter)))
{
termout = ajRestermNewResterm(termin);
ajListPushAppend(dest, termout);
}
ajListIterDel(&iter);
return ajTrue;
}
static void jaspscan_ParseInput(const AjPStr dir, const AjPStr jaspdir,
const AjPStr mats, const AjPStr excl,
ajuint *recurs, AjPList ret)
{
ajuint nm = 0;
ajuint ne = 0;
AjPStr *carr = NULL;
AjPStr *earr = NULL;
AjPFile inf = NULL;
AjPStr line = NULL;
AjPStr comm = NULL;
AjPStr val = NULL;
ajuint i;
ajuint j;
char c;
ajuint rlen = 0;
if(*recurs > JASPSCAN_RECURS)
ajFatal("Too many recursion levels in matrix list files");
line = ajStrNew();
comm = ajStrNew();
if(mats)
{
nm = ajArrCommaList(mats,&carr);
for(i=0; i < nm; ++i)
{
if(ajStrGetCharFirst(carr[i]) != '@')
ajStrFmtUpper(&carr[i]);
if(ajStrMatchC(carr[i],"ALL"))
{
jaspscan_GetFileList(dir, jaspdir, "*", ret);
ajListSortUnique(ret, ajStrVcmp, jaspscan_strdel);
}
else if(ajStrGetCharFirst(carr[i]) == '@')
{
ajStrTrimStartC(&carr[i],"@");
inf = ajFileNewInNameS(carr[i]);
if(!inf)
ajFatal("Cannot open list file %S",carr[i]);
while(ajReadlineTrim(inf,&line))
{
ajStrRemoveWhite(&line);
c = ajStrGetCharFirst(line);
if(c == '#' || c== '!')
continue;
if(ajStrGetLen(comm))
ajStrAppendC(&comm,",");
ajStrFmtUpper(&line);
ajStrAppendS(&comm,line);
}
*recurs += 1;
jaspscan_ParseInput(dir,jaspdir,comm,NULL,recurs,ret);
*recurs -= 1;
ajListSortUnique(ret, ajStrVcmp, jaspscan_strdel);
ajFileClose(&inf);
}
else
{
jaspscan_GetFileList(dir,jaspdir,ajStrGetPtr(carr[i]),ret);
ajListSortUnique(ret, ajStrVcmp, jaspscan_strdel);
}
}
for(i=0; i < nm; ++i)
ajStrDel(&carr[i]);
AJFREE(carr);
}
if(excl)
{
ne = ajArrCommaList(excl,&earr);
for(i=0; i < ne; ++i)
{
if(ajStrGetCharFirst(earr[i]) != '@')
ajStrFmtUpper(&earr[i]);
if(ajStrGetCharFirst(earr[i]) == '@')
{
ajStrTrimStartC(&earr[i],"@");
inf = ajFileNewInNameS(earr[i]);
if(!inf)
ajFatal("Cannot open list file %S",earr[i]);
while(ajReadlineTrim(inf,&line))
{
ajStrRemoveWhite(&line);
c = ajStrGetCharFirst(line);
if(c == '#' || c== '!')
continue;
if(ajStrGetLen(comm))
ajStrAppendC(&comm,",");
ajStrFmtUpper(&line);
ajStrAppendS(&comm,line);
}
*recurs += 1;
jaspscan_ParseInput(dir,jaspdir,NULL,comm,recurs,ret);
*recurs -= 1;
ajListSortUnique(ret, ajStrVcmp, jaspscan_strdel);
ajFileClose(&inf);
}
else
{
ajStrAssignS(&line,earr[i]);
ajStrAppendC(&line,J_EXT);
rlen = ajListGetLength(ret);
for(j=0; j < rlen; ++j)
{
ajListPop(ret,(void **)&val);
if(ajStrSuffixS(val,line))
ajStrDel(&val);
else
ajListPushAppend(ret,(void *)val);
}
}
}
for(i=0; i < ne; ++i)
ajStrDel(&earr[i]);
AJFREE(earr);
}
ajStrDel(&line);
ajStrDel(&comm);
return;
}
AjPMatrix ajMatrixNewFile(const AjPStr filename)
{
AjPMatrix ret = NULL;
AjPStr buffer = NULL;
const AjPStr tok = NULL;
AjPStr firststring = NULL;
AjPStr *orderstring = NULL;
AjPFile file = NULL;
AjBool first = ajTrue;
const char *ptr = NULL;
ajint **matrix = NULL;
ajint minval = -1;
ajint i = 0;
ajint l = 0;
ajint k = 0;
ajint cols = 0;
ajint rows = 0;
ajint *templine = NULL;
AjPList rlabel_list = NULL;
AjPStr *rlabel_arr = NULL;
#ifndef WIN32
static const char *delimstr = " :\t\n";
#else
static const char *delimstr = " :\t\n\r";
#endif
rlabel_list = ajListNew();
firststring = ajStrNew();
file = ajDatafileNewInNameS(filename);
if(!file)
{
ajStrDel(&firststring);
ajListFree(&rlabel_list);
return NULL;
}
/* Read row labels */
while(ajReadline(file,&buffer))
{
ptr = ajStrGetPtr(buffer);
#ifndef WIN32
if(*ptr != '#' && *ptr != '\n')
#else
if(*ptr != '#' && *ptr != '\n' && *ptr != '\r')
#endif
{
if(first)
first = ajFalse;
else
{
ajFmtScanC(ptr, "%S", &firststring);
ajListPushAppend(rlabel_list, firststring);
firststring = ajStrNew();
}
}
}
first = ajTrue;
ajStrDel(&firststring);
rows = ajListToarray(rlabel_list, (void ***) &rlabel_arr);
ajFileSeek(file, 0, 0);
while(ajReadline(file,&buffer))
{
ajStrRemoveWhiteExcess(&buffer);
ptr = ajStrGetPtr(buffer);
if(*ptr && *ptr != '#')
{
if(first)
{
cols = ajStrParseCountC(buffer,delimstr);
AJCNEW0(orderstring, cols);
for(i=0; i<cols; i++)
orderstring[i] = ajStrNew();
tok = ajStrParseC(buffer, " :\t\n");
ajStrAssignS(&orderstring[l++], tok);
while((tok = ajStrParseC(NULL, " :\t\n")))
ajStrAssignS(&orderstring[l++], tok);
first = ajFalse;
ret = ajMatrixNewAsym(orderstring, cols,
rlabel_arr, rows,
filename);
matrix = ret->Matrix;
}
else
{
ajFmtScanC(ptr, "%S", &firststring);
/* JISON 19/7/4
k = ajSeqcvtGetCodeK(ret->Cvt, ajStrGetCharFirst(firststring)); */
k = ajSeqcvtGetCodeS(ret->Cvt, firststring);
/*
** cols+1 is used below because 2nd and subsequent lines have
** one more string in them (the residue label)
*/
templine = ajArrIntLine(buffer,delimstr,2,cols+1);
for(i=0; i<cols; i++)
{
if(templine[i] < minval)
minval = templine[i];
/* JISON 19/7/4
matrix[k][ajSeqcvtGetCodeK(ret->Cvt,
ajStrGetCharFirst(orderstring[i]))]
= templine[i]; */
matrix[k][ajSeqcvtGetCodeAsymS(ret->Cvt,
orderstring[i])]
= templine[i];
}
AJFREE(templine);
}
}
}
ajDebug("fill rest with minimum value %d\n", minval);
ajFileClose(&file);
ajStrDel(&buffer);
for(i=0; i<cols; i++)
ajStrDel(&orderstring[i]);
AJFREE(orderstring);
ajDebug("read matrix file %S\n", filename);
ajStrDel(&firststring);
for(i=0; i<rows; i++)
ajStrDel(&rlabel_arr[i]);
AJFREE(rlabel_arr);
ajListFree(&rlabel_list);
return ret;
}
int main(int argc, char **argv)
{
AjPSeqall seqall;
AjPSeq seq = NULL;
AjPReport report = NULL;
AjPStr jaspdir = NULL;
AjPStr menu = NULL;
AjPStr substr = NULL;
AjPStr mats = NULL;
AjPStr excl = NULL;
float thresh = 0.;
ajuint recurs = 0;
AjPStr dir = NULL;
AjPStr mfname = NULL;
AjPList flist = NULL;
AjPList hits = NULL;
AjPStr head = NULL;
ajint begin;
ajint end;
ajuint mno;
char cp;
ajuint i;
AjPTable mattab = NULL;
AjPFeattable TabRpt = NULL;
AjBool both = ajFalse;
embInit("jaspscan", argc, argv);
seqall = ajAcdGetSeqall("sequence");
menu = ajAcdGetListSingle("menu");
mats = ajAcdGetString("matrices");
excl = ajAcdGetString("exclude");
thresh = ajAcdGetFloat("threshold");
report = ajAcdGetReport("outfile");
both = ajAcdGetBoolean("both");
jaspdir = ajStrNew();
substr = ajStrNew();
flist = ajListNew();
hits = ajListNew();
dir = ajStrNew();
head = ajStrNew();
cp = ajStrGetCharFirst(menu);
if(cp=='C')
ajStrAssignC(&jaspdir,J_COR);
else if(cp=='F')
ajStrAssignC(&jaspdir,J_FAM);
else if(cp=='P')
ajStrAssignC(&jaspdir,J_PHY);
else if(cp=='N')
ajStrAssignC(&jaspdir,J_CNE);
else if(cp=='O')
ajStrAssignC(&jaspdir,J_POL);
else if(cp=='S')
ajStrAssignC(&jaspdir,J_SPL);
else
ajFatal("Invalid JASPAR database selection");
ajStrAssignS(&dir, ajDatafileValuePath());
if(!ajStrGetLen(dir))
ajFatal("EMBOSS DATA directory couldn't be determined");
jaspscan_ParseInput(dir, jaspdir, mats, excl, &recurs, flist);
mno = ajListGetLength(flist);
if(cp == 'C')
mattab = jaspscan_ReadCoreList(jaspdir);
if(cp == 'F')
mattab = jaspscan_ReadFamList(jaspdir);
if(cp == 'P')
mattab = jaspscan_ReadCoreList(jaspdir);
if(cp == 'N')
mattab = jaspscan_ReadCoreList(jaspdir);
if(cp == 'O')
mattab = jaspscan_ReadCoreList(jaspdir);
if(cp == 'S')
mattab = jaspscan_ReadCoreList(jaspdir);
ajFmtPrintS(&head,"Database scanned: %S Threshold: %.3f",jaspdir,thresh);
ajReportSetHeaderS(report,head);
while(ajSeqallNext(seqall, &seq))
{
begin = ajSeqallGetseqBegin(seqall);
end = ajSeqallGetseqEnd(seqall);
ajStrAssignSubC(&substr,ajSeqGetSeqC(seq),begin-1,end-1);
ajStrFmtUpper(&substr);
TabRpt = ajFeattableNewSeq(seq);
for(i=0; i < mno; ++i)
{
ajListPop(flist,(void **)&mfname);
jaspscan_scan(substr,begin,mfname, cp, thresh, both, hits);
ajListPushAppend(flist, (void **)mfname);
}
jaspscan_ReportHits(TabRpt,mattab,hits);
ajReportWrite(report, TabRpt, seq);
ajFeattableDel(&TabRpt);
}
while(ajListPop(flist,(void **)&mfname))
ajStrDel(&mfname);
ajStrDel(&dir);
ajStrDel(&menu);
ajStrDel(&excl);
ajStrDel(&substr);
ajStrDel(&mats);
ajStrDel(&head);
ajStrDel(&jaspdir);
ajSeqDel(&seq);
ajTableMapDel(mattab,jaspscan_ClearTable,NULL);
ajTableFree(&mattab);
ajListFree(&flist);
ajListFree(&hits);
ajSeqallDel(&seqall);
ajReportDel(&report);
embExit();
return 0;
}
/* @funcstatic sigscanlig_score_ligands_site *********************************
**
** Writes score data for ligands (Lighit objects) from individual hits to
** signatures (Hit objects). Site score method is used. List of hit objects
** must be sorted by ligand type and site number.
**
** @param [r] hits [const AjPList] List of hit objects.
** @return [AjBool] True on success
** @@
******************************************************************************/
AjPList sigscanlig_score_ligands_site(AjPList hits)
{
AjPList ret = NULL;
AjIList iter = NULL; /* Iterator. */
EmbPHit hit = NULL;
AjPStr prev_ligand = NULL;
SigPLighit lighit = NULL;
ajint prevsn = -1; /* Previous site number */
float score = 0.0; /* Score for current ligand */
ajint nhits = 0; /* No. of hits (patches) for current
ligand. */
ajint nsites = 0; /* No. of sites for current ligand */
float score_site = 0.0; /* Score for this site. */
ajint patch_cnt = 0; /* No. of patches in current site. */
ret = ajListNew();
prev_ligand = ajStrNew();
iter = ajListIterNew(hits);
/* Hits are already sorted by ligid & site number */
while((hit = (EmbPHit) ajListIterGet(iter)))
{
/* ajFmtPrint("Current hit: %S (ligid: %S, sn: %d) score: %f\n",
hit->Acc, hit->Sig->Ligid, hit->Sig->sn, hit->Score); */
/* New site */
if(hit->Sig->sn != prevsn)
{
if(patch_cnt)
score_site /= patch_cnt;
score += score_site;
patch_cnt = 0;
score_site = 0;
}
/* New ligand */
if((!ajStrMatchS(hit->Sig->Ligid, prev_ligand)))
{
if(nsites)
score /= nsites;
if(lighit)
{
/* lighit->ns = snarr_siz; */
lighit->ns = nsites;
lighit->np = nhits;
lighit->score = score;
ajStrAssignS(&lighit->ligid, prev_ligand);
ajListPushAppend(ret, lighit);
}
lighit = sigscanlig_LighitNew();
nsites = 0;
nhits=0;
prevsn = -1;
score = 0;
}
/* Increment count of sites and hits/patches (for current
ligand) and patches (for current site) */
if(hit->Sig->sn != prevsn)
nsites++;
score_site += hit->Score;
nhits++;
patch_cnt++;
ajStrAssignS(&prev_ligand, hit->Sig->Ligid);
prevsn = hit->Sig->sn;
}
/* Process last hit */
if(patch_cnt)
score_site /= patch_cnt;
score += score_site;
if(nsites)
score /= nsites;
if(lighit)
{
/* lighit->ns = snarr_siz; */
lighit->ns = nsites;
lighit->np = nhits;
lighit->score = score;
ajStrAssignS(&lighit->ligid, prev_ligand);
ajListPushAppend(ret, lighit);
}
ajListSort(ret, sigscanlig_MatchinvScore);
ajListIterDel(&iter);
ajStrDel(&prev_ligand);
return ret;
}
AjPList sigscanlig_score_ligands_patch(AjPList hits)
{
AjPList ret = NULL;
AjIList iter = NULL; /* Iterator. */
EmbPHit hit = NULL;
AjPStr prev_ligand = NULL;
SigPLighit lighit = NULL;
float score = 0.0;
ajint nhits = 0; /* No. of hits (patches) for current ligand. */
ajint nsites = 0; /* No. of sites for current ligand. */
ajint prevsn = -1; /* Previous site number */
ret = ajListNew();
prev_ligand = ajStrNew();
iter = ajListIterNew(hits);
while((hit = (EmbPHit) ajListIterGet(iter)))
{
/* New ligand */
if((!ajStrMatchS(hit->Sig->Ligid, prev_ligand)))
{
if(nhits)
score /= nhits;
if(lighit)
{
lighit->ns = nsites;
lighit->np = nhits;
lighit->score = score;
ajStrAssignS(&lighit->ligid, prev_ligand);
ajListPushAppend(ret, lighit);
}
lighit = sigscanlig_LighitNew();
nsites = 0;
nhits=0;
prevsn = -1;
score = 0;
}
/* Increment count of sites and hits/patches (for current ligand)
and patches (for current site) */
if(hit->Sig->sn != prevsn)
nsites++;
score+= hit->Score;
nhits++;
ajStrAssignS(&prev_ligand, hit->Sig->Ligid);
prevsn = hit->Sig->sn;
}
/* Process last hit */
if(nhits)
score /= nhits;
if(lighit)
{
lighit->ns = nsites;
lighit->np = nhits;
lighit->score = score;
ajStrAssignS(&lighit->ligid, prev_ligand);
ajListPushAppend(ret, lighit);
}
ajListSort(ret, sigscanlig_MatchinvScore);
ajListIterDel(&iter);
ajStrDel(&prev_ligand);
return ret;
}
int main(int argc, char **argv)
{
AjPList sigin = NULL; /* Signature input file names. */
AjPStr signame = NULL; /* Name of signature file. */
AjPFile sigf = NULL; /* Signature input file. */
EmbPSignature sig = NULL; /* Signature. */
AjPList siglist = NULL; /* List of signatures. */
AjIList sigiter = NULL; /* Iterator for siglist. */
AjBool sigok = ajFalse; /* True if signature processed ok. */
EmbPHit hit = NULL; /* Hit to store signature-sequence match. */
AjPList hits = NULL; /* List of hits */
AjPList ligands = NULL; /* List of top-scoring ligands. */
AjPSeqall database=NULL; /* Protein sequences to match signature
against. */
AjPSeq seq = NULL; /* Current sequence. */
AjPMatrixf sub =NULL; /* Residue substitution matrix. */
float gapo =0.0; /* Gap insertion penalty. */
float gape =0.0; /* Gap extension penalty. */
AjPStr nterm=NULL; /* Holds N-terminal matching options from
acd. */
ajint ntermi=0; /* N-terminal option as int. */
AjPFile hitsf =NULL; /* Hits output file.
sequence matches. */
AjPDirout hitsdir=NULL; /* Directory of hits files (output). */
AjPFile alignf =NULL; /* Alignment output file. */
AjPDirout aligndir=NULL; /* Directory of alignment files (output). */
AjPFile resultsf =NULL; /* Results file (output). */
AjPDirout resultsdir=NULL; /* Directory of results files (output). */
AjPStr mode = NULL; /* Mode, 1: Patch score mode, 2:
Site score mode. */
ajint modei = 0; /* Selected mode as integer. */
SigPLighit lighit = NULL;
embInitPV("sigscanlig", argc, argv, "SIGNATURE",VERSION);
/* GET VALUES FROM ACD */
sigin = ajAcdGetDirlist("siginfilesdir");
database = ajAcdGetSeqall("dbseqall");
sub = ajAcdGetMatrixf("sub");
gapo = ajAcdGetFloat("gapo");
gape = ajAcdGetFloat("gape");
nterm = ajAcdGetListSingle("nterm");
hitsdir = ajAcdGetOutdir("hitsoutdir");
aligndir = ajAcdGetOutdir("alignoutdir");
resultsdir = ajAcdGetOutdir("resultsoutdir");
mode = ajAcdGetListSingle("mode");
/*Assign N-terminal matching option etc. */
ajFmtScanS(nterm, "%d", &ntermi);
modei = (ajint) ajStrGetCharFirst(mode)-48;
/* READ & PROCESS SIGNATURES */
siglist = ajListNew();
while(ajListPop(sigin, (void **) &signame))
{
/* Read signature files, compile signatures and populate list. */
sigok = ajFalse;
if((sigf = ajFileNewInNameS(signame)))
if((sig = embSignatureReadNew(sigf)))
if(embSignatureCompile(&sig, gapo, gape, sub))
{
sigok=ajTrue;
ajListPushAppend(siglist, sig);
/*
ajFmtPrint("Id: %S\nDomid: %S\nLigid: %S\nns: %d\n"
"sn: %d\nnp: %d\npn: %d\nminpatch: %d\n"
"maxgap: %d\n",
sig->Id, sig->Domid, sig->Ligid, sig->ns,
sig->sn, sig->np, sig->pn, sig->minpatch,
sig->maxgap); */
}
if(!sigok)
{
ajWarn("Could not process %S", signame);
embSignatureDel(&sig);
ajFileClose(&sigf);
ajStrDel(&signame);
continue;
}
ajFileClose(&sigf);
ajStrDel(&signame);
}
ajListFree(&sigin);
/* ALIGN EACH QUERY SEQUENCE TO LIST OF SIGNATURE */
while(ajSeqallNext(database, &seq))
{
/* Do sequence-signature alignment and save results */
hits = ajListNew();
sigiter = ajListIterNew(siglist);
while((sig = (EmbPSignature) ajListIterGet(sigiter)))
{
if(embSignatureAlignSeq(sig, seq, &hit, ntermi))
{
hit->Sig = sig;
ajListPushAppend(hits, hit);
hit=NULL; /* To force reallocation by embSignatureAlignSeq */
}
/* There has to be a hit for each signature for correct
generation of the LHF by sigscanlig_WriteFasta. So push
an empty hit if necessary. 'hit'=NULL forces
reallocation by embSignatureAlignSeq. */
/*
else
{
hit = embHitNew();
ajListPushAppend(hits, hit);
hit=NULL;
}
*/
}
ajListIterDel(&sigiter);
/* Rank-order the list of hits by score */
ajListSort(hits, embMatchinvScore);
/* Write ligand hits & alignment files (output) */
hitsf = ajFileNewOutNameDirS(ajSeqGetNameS(seq), hitsdir);
alignf = ajFileNewOutNameDirS(ajSeqGetNameS(seq), aligndir);
resultsf = ajFileNewOutNameDirS(ajSeqGetNameS(seq), resultsdir);
/* if((!sigscanlig_WriteFasta(hitsf, siglist, hits)))
ajFatal("Bad args to sigscanlig_WriteFasta"); */
if((!sigscanlig_WriteFasta(hitsf, hits)))
ajFatal("Bad args to sigscanlig_WriteFasta");
if((!sigscanlig_SignatureAlignWriteBlock(alignf, hits)))
ajFatal("Bad args to sigscanlig_SignatureAlignWriteBlock");
/* if((!sigscanlig_SignatureAlignWriteBlock(alignf, siglist, hits)))
ajFatal("Bad args to sigscanlig_SignatureAlignWriteBlock"); */
/* Sort list of hits by ligand type and site number.
Process list of ligands and print out. */
ajListSortTwo(hits, embMatchLigid, embMatchSN);
if(modei==1)
ligands = sigscanlig_score_ligands_patch(hits);
else if(modei==2)
ligands = sigscanlig_score_ligands_site(hits);
else
ajFatal("Unrecognised mode");
sigscanlig_WriteResults(ligands, resultsf);
ajFileClose(&hitsf);
ajFileClose(&alignf);
ajFileClose(&resultsf);
/* Memory management */
while(ajListPop(hits, (void **) &hit))
embHitDel(&hit);
ajListFree(&hits);
while(ajListPop(ligands, (void **) &lighit))
sigscanlig_LigHitDel(&lighit);
ajListFree(&ligands);
}
/* MEMORY MANAGEMENT */
while(ajListPop(siglist, (void **) &sig))
embSignatureDel(&sig);
ajListFree(&siglist);
ajSeqallDel(&database);
ajMatrixfDel(&sub);
ajStrDel(&nterm);
ajDiroutDel(&hitsdir);
ajDiroutDel(&aligndir);
ajDiroutDel(&resultsdir);
ajStrDel(&mode);
embExit();
return 0;
}
int main(int argc, char **argv)
{
/* Variable declarations */
AjPOutfile outfile = NULL;
AjPOboall oboall = NULL;
AjBool subclasses = ajFalse;
AjBool obsolete = ajFalse;
AjPObo obo = NULL;
AjPObo obotest = NULL;
AjPList obolist = ajListNew();
AjPTable foundtable = NULL;
ajuint ifound = 0;
ajuint ikeep = 0;
/* ACD processing */
embInit("ontoget", argc, argv);
oboall = ajAcdGetOboall("oboterms");
outfile = ajAcdGetOutobo("outfile");
subclasses = ajAcdGetBoolean("subclasses");
obsolete = ajAcdGetBoolean("obsolete");
foundtable = ajTablestrNew(600);
while(ajOboallNext(oboall, &obo))
{
ajListPushAppend(obolist, ajOboNewObo(obo));
if(subclasses)
ajOboGetTree(obo, obolist);
while(ajListGetLength(obolist))
{
ajListPop(obolist, (void**) &obotest);
if(!obsolete && ajOboIsObsolete(obotest))
{
ajOboDel(&obotest);
continue;
}
ifound++;
if(!ajTableMatchS(foundtable, obotest->Id))
{
ajObooutWrite(outfile, obotest);
ajTablePut(foundtable, ajStrNewS(obotest->Id),
(void *) 1);
ikeep++;
}
ajOboDel(&obotest);
}
}
if(!ifound)
ajErr("No matching terms");
/* Memory clean-up and exit */
ajListFree(&obolist);
ajOboallDel(&oboall);
ajOboDel(&obo);
ajOutfileClose(&outfile);
ajTablestrFreeKey(&foundtable);
embExit();
return 0;
}
AjPMatrixf ajMatrixfNewFile(const AjPStr filename)
{
AjPMatrixf ret = NULL;
AjPStr *orderstring = NULL;
AjPStr buffer = NULL;
AjPStr firststring = NULL;
AjPStr reststring = NULL;
const AjPStr tok = NULL;
ajint len = 0;
ajint i = 0;
ajint l = 0;
ajint k = 0;
ajint cols = 0;
ajint rows = 0;
const char *ptr = NULL;
AjPFile file = NULL;
AjBool first = ajTrue;
float **matrix = NULL;
float *templine = NULL;
float minval = -1.0;
AjPList rlabel_list = NULL;
AjPStr *rlabel_arr = NULL;
#ifndef WIN32
static const char *delimstr = " :\t\n";
#else
static const char *delimstr = " :\t\n\r";
#endif
rlabel_list = ajListNew();
firststring = ajStrNew();
reststring = ajStrNew();
file = ajDatafileNewInNameS(filename);
if(!file)
{
ajStrDel(&firststring);
ajStrDel(&reststring);
return NULL;
}
/* Read row labels */
while(ajReadline(file,&buffer))
{
ptr = ajStrGetPtr(buffer);
#ifndef WIN32
if(*ptr != '#' && *ptr != '\n')
#else
if(*ptr != '#' && *ptr != '\n' && *ptr != '\r')
#endif
{
if(first)
first = ajFalse;
else
{
ajFmtScanC(ptr, "%S", &firststring);
ajListPushAppend(rlabel_list, firststring);
firststring = ajStrNew();
}
}
}
first = ajTrue;
ajStrDel(&firststring);
rows = ajListToarray(rlabel_list, (void ***) &rlabel_arr);
ajFileSeek(file, 0, 0);
while(ajReadline(file,&buffer))
{
ajStrRemoveWhiteExcess(&buffer);
ptr = ajStrGetPtr(buffer);
if(*ptr && *ptr != '#')
{
if(first)
{
cols = ajStrParseCountC(buffer,delimstr);
AJCNEW0(orderstring, cols);
for(i=0; i<cols; i++)
orderstring[i] = ajStrNew();
tok = ajStrParseC(buffer, " :\t\n");
ajStrAssignS(&orderstring[l++], tok);
while((tok = ajStrParseC(NULL, " :\t\n")))
ajStrAssignS(&orderstring[l++], tok);
first = ajFalse;
ret = ajMatrixfNewAsym(orderstring, cols,
rlabel_arr, rows,
filename);
matrix = ret->Matrixf;
}
else
{
ajFmtScanC(ptr, "%S", &firststring);
/* JISON 19/7/4
k = ajSeqcvtGetCodeK(ret->Cvt,
ajStrGetCharFirst(firststring)); */
k = ajSeqcvtGetCodeS(ret->Cvt, firststring);
len = MAJSTRGETLEN(firststring);
ajStrAssignSubC(&reststring, ptr, len, -1);
/*
** Must discard the first string (label) and use
** reststring otherwise ajArrFloatLine would fail (it
** cannot convert a string to a float)
**
** Use cols,1,cols in below because although 2nd and
** subsequent lines have one more string in them (the
** residue label in the 1st column) we've discarded that
** from the string that's passed
*/
templine = ajArrFloatLine(reststring,delimstr,1,cols);
for(i=0; i<cols; i++)
{
if(templine[i] < minval)
minval = templine[i];
/* JISON 19/7/4
matrix[k][ajSeqcvtGetCodeK(ret->Cvt,
ajStrGetCharFirst(orderstring[i]))]
= templine[i]; */
matrix[k][ajSeqcvtGetCodeAsymS(ret->Cvt,
orderstring[i])]
= templine[i];
}
AJFREE(templine);
}
}
}
ajDebug("fill rest with minimum value %d\n", minval);
ajFileClose(&file);
ajStrDel(&buffer);
for(i=0; i<cols; i++)
ajStrDel(&orderstring[i]);
AJFREE(orderstring);
ajDebug("read matrix file %S\n", filename);
ajStrDel(&firststring);
ajStrDel(&reststring);
for(i=0; i<rows; i++)
ajStrDel(&rlabel_arr[i]);
AJFREE(rlabel_arr);
ajListFree(&rlabel_list);
return ret;
}
static AjBool cacheNodeInsert(EnsPCache cache, CachePNode node)
{
CachePNode old = NULL;
if(!cache)
return ajFalse;
if(!node)
return ajFalse;
if(cache->MaxSize && (node->Bytes > cache->MaxSize))
return ajFalse;
/* Insert the node into the AJAX List. */
ajListPushAppend(cache->List, (void *) node);
/* Insert the node into the AJAX Table. */
ajTablePut(cache->Table, node->Key, (void *) node);
/* Update the cache statistics. */
cache->Bytes += node->Bytes;
cache->Count++;
cache->Stored++;
/* If the cache is too big, remove the top node(s). */
while((cache->MaxBytes && (cache->Bytes > cache->MaxBytes)) ||
(cache->MaxCount && (cache->Count > cache->MaxCount)))
{
/* Remove the top node from the AJAX List. */
ajListPop(cache->List, (void **) &old);
/* Remove the node also from the AJAX Table. */
ajTableRemove(cache->Table, old->Key);
/* Update the cache statistics. */
cache->Bytes -= old->Bytes;
cache->Count--;
cache->Dropped++;
/* Write changes of value data to disk if any. */
if(cache->Write && old->Value && old->Dirty)
(*cache->Write)(old->Value);
/* Both, key and value data are deleted via cacheNodeDel. */
cacheNodeDel(cache, &old);
}
return ajTrue;
}
void* ensCacheFetch(EnsPCache cache, void *key)
{
void *value = NULL;
AjIList iter = NULL;
CachePNode lnode = NULL;
CachePNode tnode = NULL;
if(!cache)
return NULL;
if(!key)
return NULL;
tnode = (CachePNode) ajTableFetch(cache->Table, key);
if(tnode)
{
cache->Hit++;
/* Move the node to the end of the list. */
iter = ajListIterNew(cache->List);
while(!ajListIterDone(iter))
{
lnode = (CachePNode) ajListIterGet(iter);
if(lnode == tnode)
{
ajListIterRemove(iter);
ajListPushAppend(cache->List, (void *) lnode);
break;
}
}
ajListIterDel(&iter);
/*
** Reference the object when returned by the cache so that external
** code has to delete it irrespectively whether it was read from the
** cache or instantiated by the cache->Read function.
*/
if(cache->Reference && tnode->Value)
value = (*cache->Reference)(tnode->Value);
}
else
{
cache->Miss++;
if(cache->Read)
{
value = (*cache->Read)(key);
if(value)
{
tnode = cacheNodeNew(cache, key, value);
if(!cacheNodeInsert(cache, tnode))
cacheNodeDel(cache, &tnode);
}
}
}
return value;
}
static AjBool dbiblast_parseNcbi(const AjPStr line, AjPFile * alistfile,
AjBool systemsort, AjPStr const * fields,
ajint* maxFieldLen,
ajuint* countfield,
AjPStr* myid,
AjPList* fdlist)
{
char* fd;
static ajint numFields;
static ajint accfield = -1;
static ajint desfield = -1;
static ajint svnfield = -1;
static AjBool reset = AJTRUE;
if(!fields)
{
reset = ajTrue;
accfield = svnfield = desfield = -1;
return ajFalse;
}
if(reset)
{
numFields = 0;
while(fields[numFields])
{
if(ajStrMatchCaseC(fields[numFields], "acc"))
accfield=numFields;
else if(ajStrMatchCaseC(fields[numFields], "sv"))
svnfield=numFields;
else if(ajStrMatchCaseC(fields[numFields], "des"))
desfield=numFields;
else
ajWarn("EMBL parsing unknown field '%S' ignored",
fields[numFields]);
numFields++;
}
reset = ajFalse;
}
if(!wrdexp)
wrdexp = ajRegCompC("([A-Za-z0-9]+)");
ajStrAssignC(&tmpdes,"");
ajStrAssignC(&t,"");
ajStrAssignC(&tmpac,"");
ajStrAssignC(&tmpsv,"");
ajStrAssignC(&tmpgi,"");
ajStrAssignC(&tmpdb,"");
ajFmtPrintS(&t,">%S",line);
if(!ajSeqParseNcbi(t,myid,&tmpac,&tmpsv,&tmpgi,&tmpdb,&tmpdes))
return ajFalse;
if(ajStrGetLen(tmpac))
ajStrFmtUpper(&tmpac);
if(accfield >= 0)
embDbiMaxlen(&tmpac, &maxFieldLen[accfield]);
if(svnfield >= 0)
{
embDbiMaxlen(&tmpsv, &maxFieldLen[svnfield]);
embDbiMaxlen(&tmpgi, &maxFieldLen[svnfield]);
}
ajStrFmtUpper(myid);
/* ajDebug("parseNCBI success\n"); */
if(systemsort)
{
if(accfield >= 0 && ajStrGetLen(tmpac))
{
countfield[accfield]++;
ajFmtPrintF(alistfile[accfield], "%S %S\n", *myid, tmpac);
}
if(svnfield >= 0 && ajStrGetLen(tmpsv))
{
countfield[svnfield]++;
ajFmtPrintF(alistfile[svnfield], "%S %S\n", *myid, tmpsv);
}
if(svnfield >= 0 && ajStrGetLen(tmpgi))
{
countfield[svnfield]++;
ajFmtPrintF(alistfile[svnfield], "%S %S\n", *myid, tmpgi);
}
if(desfield >= 0 && ajStrGetLen(tmpdes))
while(ajRegExec(wrdexp, tmpdes))
{
ajRegSubI(wrdexp, 1, &tmpfd);
embDbiMaxlen(&tmpfd, &maxFieldLen[desfield]);
ajStrFmtUpper(&tmpfd);
ajDebug("++des '%S'\n", tmpfd);
countfield[desfield]++;
ajFmtPrintF(alistfile[desfield], "%S %S\n", *myid, tmpfd);
ajRegPost(wrdexp, &tmpdes);
}
}
else
{
if(accfield >= 0 && ajStrGetLen(tmpac))
{
fd = ajCharNewS(tmpac);
countfield[accfield]++;
ajListPushAppend(fdlist[accfield], fd);
}
if(svnfield >= 0 && ajStrGetLen(tmpsv))
{
fd = ajCharNewS(tmpsv);
countfield[svnfield]++;
ajListPushAppend(fdlist[svnfield], fd);
}
if(svnfield >= 0 && ajStrGetLen(tmpgi))
{
fd = ajCharNewS(tmpgi);
ajListPushAppend(fdlist[svnfield], fd);
}
if(desfield >= 0 && ajStrGetLen(tmpdes))
{
while(ajRegExec(wrdexp, tmpdes))
{
ajRegSubI(wrdexp, 1, &tmpfd);
embDbiMaxlen(&tmpfd, &maxFieldLen[desfield]);
ajStrFmtUpper(&tmpfd);
ajDebug("++des '%S'\n", tmpfd);
fd = ajCharNewS(tmpfd);
countfield[desfield]++;
ajListPushAppend(fdlist[desfield], fd);
ajRegPost(wrdexp, &tmpdes);
}
}
}
/* ajDebug("parseNCBI '%S' '%S'\n", *myid, tmpac); */
return ajTrue;
}
static void jaspscan_scan(const AjPStr seq, const ajuint begin,
const AjPStr mfname, const char type,
const float threshold,
const AjBool both, AjPList hits)
{
PJsphits val = NULL;
AjPStr mname = NULL;
float **matrix = NULL;
ajuint cols;
ajuint i;
ajuint rc;
ajuint cc;
ajuint limit;
ajuint slen;
const char *p;
char schar;
float sum = 0.;
float rmax = 0.;
float scorepc = 0.;
float maxscore = 0.;
cols = jaspscan_readmatrix(mfname, &matrix);
maxscore = 0.;
for(cc = 0; cc < cols; ++cc)
{
rmax = 0.;
for(rc = 0; rc < 4; ++rc)
rmax = (rmax > matrix[rc][cc]) ? rmax : matrix[rc][cc];
maxscore += rmax;
}
slen = ajStrGetLen(seq);
limit = (slen - cols) + 1;
if(limit <= 0)
return;
mname = ajStrNew();
ajStrAssignS(&mname,mfname);
ajFilenameTrimPath(&mname);
ajFilenameTrimExt(&mname);
p = ajStrGetPtr(seq);
for(i=0; i < limit; ++i)
{
sum = 0.;
schar = p[i];
for(cc = 0; cc < cols; ++cc)
{
schar = p[i+cc];
if(schar == 'A')
sum += matrix[0][cc];
else if(schar == 'C')
sum += matrix[1][cc];
else if(schar == 'G')
sum += matrix[2][cc];
else if(schar == 'T')
sum += matrix[3][cc];
}
scorepc = (sum * (float)100.) / maxscore;
if(scorepc >= threshold)
{
val = jaspscan_hitsnew();
val->type = type;
ajStrAssignS(&val->matname,mname);
val->start = i + begin;
val->end = val->start + cols - 1;
val->score = sum;
val->threshold = threshold;
val->scorepc = scorepc;
val->maxscore = maxscore;
ajListPushAppend(hits,(void *)val);
}
}
if(both)
{
jaspscan_CompMat(matrix, cols);
p = ajStrGetPtr(seq);
for(i=0; i < limit; ++i)
{
sum = 0.;
schar = p[i];
for(cc = 0; cc < cols; ++cc)
{
schar = p[i+cc];
if(schar == 'A')
sum += matrix[0][cc];
else if(schar == 'C')
sum += matrix[1][cc];
else if(schar == 'G')
sum += matrix[2][cc];
else if(schar == 'T')
sum += matrix[3][cc];
}
scorepc = (sum * (float)100.) / maxscore;
if(scorepc >= threshold)
{
val = jaspscan_hitsnew();
val->type = type;
ajStrAssignS(&val->matname,mname);
val->end = i + begin;
val->start = val->end + cols - 1;
val->score = sum;
val->threshold = threshold;
val->scorepc = scorepc;
val->maxscore = maxscore;
ajListPushAppend(hits,(void *)val);
}
}
}
for(i = 0; i < 4; ++i)
AJFREE(matrix[i]);
AJFREE(matrix);
ajStrDel(&mname);
return;
}
int main(int argc, char **argv)
{
/* Variable declarations */
AjPOutfile outfile = NULL;
AjPTaxall taxall = NULL;
AjPTax tax = NULL;
AjPTax taxtest = NULL;
AjPList taxlist = ajListNew();
AjPTable foundtable = NULL;
ajuint ifound = 0;
ajuint ikeep = 0;
/* ACD processing */
embInit("taxgetspecies", argc, argv);
taxall = ajAcdGetTaxonall("taxons");
outfile = ajAcdGetOuttaxon("outfile");
foundtable = ajTablestrNew(600);
while(ajTaxallNext(taxall, &tax))
{
ajListPushAppend(taxlist, ajTaxNewTax(tax));
ajTaxGetTree(tax, taxlist);
while(ajListGetLength(taxlist))
{
ajListPop(taxlist, (void**) &taxtest);
if(!ajStrMatchC(ajTaxGetRank(taxtest), "species"))
{
ajTaxDel(&taxtest);
continue;
}
ifound++;
if(!ajTableMatchS(foundtable, taxtest->Id))
{
ajTaxoutWrite(outfile, taxtest);
ajTablePut(foundtable, ajStrNewS(taxtest->Id),
(void *) 1);
ikeep++;
}
ajTaxDel(&taxtest);
}
}
if(!ifound)
ajErr("No matching terms");
/* Memory clean-up and exit */
ajListFree(&taxlist);
ajTaxallDel(&taxall);
ajTaxDel(&tax);
ajOutfileClose(&outfile);
ajTablestrFreeKey(&foundtable);
embExit();
return 0;
}
static AjBool dbiblast_parseSimple(const AjPStr line,
AjPFile * alistfile,
AjBool systemsort, AjPStr const * fields,
ajint* maxFieldLen,
ajuint* countfield,
AjPStr* myid,
AjPList* myfdl)
{
static AjPRegexp idexp = NULL;
static AjPStr mytmpac = NULL;
char* ac;
static ajint numFields;
static ajint accfield = -1;
static AjBool reset = AJTRUE;
if(!fields)
{
reset = ajTrue;
accfield = -1;
return ajFalse;
}
if(reset)
{
numFields = 0;
while(fields[numFields])
{
if(ajStrMatchCaseC(fields[numFields], "acc"))
accfield=numFields;
else if(!ajStrMatchCaseC(fields[numFields], "sv") &&
!ajStrMatchCaseC(fields[numFields], "des"))
ajWarn("Simple ID parsing unknown field '%S' ignored",
fields[numFields]);
numFields++;
}
reset = ajFalse;
}
if(!idexp)
idexp = ajRegCompC("^([^ ]+)( +([A-Za-z][A-Za-z0-9]+[0-9]))");
if(!ajRegExec(idexp, line))
return ajFalse;
ajRegSubI(idexp, 1, myid);
ajRegSubI(idexp, 3, &mytmpac);
ajStrFmtUpper(myid);
ajStrFmtUpper(&mytmpac); /* GCG mixes case on new SwissProt acnums */
if(accfield >= 0)
{
embDbiMaxlen(&mytmpac, &maxFieldLen[accfield]);
countfield[accfield]++;
if(systemsort)
ajFmtPrintF(alistfile[accfield], "%S %S\n", *myid, mytmpac);
else
{
ac = ajCharNewS(mytmpac);
ajListPushAppend(myfdl[accfield], ac);
}
}
ajDebug("parseSimple '%S' '%S'\n", *myid, mytmpac);
return ajTrue;
}