static void dbiblast_newname(AjPStr* nname, const AjPStr oname,
const char *suff)
{
ajStrAssignS(nname, oname);
if(ajStrGetCharFirst(oname)=='@')
ajStrCutStart(nname, 1);
ajStrAppendK(nname, '.');
ajStrAppendC(nname, suff);
return;
}
static void dbiblast_dbname(AjPStr* dbname,
const AjPStr oname, const char *suff)
{
AjPStr suffix = NULL;
ajFmtPrintS(&suffix, ".%s", suff);
ajStrAssignS(dbname, oname);
if(ajStrGetCharFirst(oname)=='@')
ajStrCutStart(dbname, 1);
if(!ajStrSuffixS(*dbname, suffix))
{
ajStrDel(&suffix);
return;
}
ajStrCutEnd(dbname, ajStrGetLen(suffix));
ajStrDel(&suffix);
return;
}
int main(int argc, char **argv)
{
AjPSeqall seqall = NULL;
AjPFile dend_outfile = NULL;
AjPStr tmp_dendfilename = NULL;
AjPFile tmp_dendfile = NULL;
AjPStr tmp_aln_outfile = NULL;
AjPSeqset seqset = NULL;
AjPSeqout seqout = NULL;
AjPSeqin seqin = NULL;
AjBool only_dend;
AjBool are_prot = ajFalse;
AjBool do_slow;
AjBool use_dend;
AjPFile dend_file = NULL;
AjPStr dend_filename = NULL;
ajint ktup;
ajint gapw;
ajint topdiags;
ajint window;
AjBool nopercent;
AjPStr pw_matrix = NULL;
AjPStr pw_dna_matrix = NULL;
AjPFile pairwise_matrix = NULL;
float pw_gapc;
float pw_gapv;
AjPStr pwmstr = NULL;
char pwmc = '\0';
AjPStr pwdstr = NULL;
char pwdc = '\0';
AjPStr m1str = NULL;
AjPStr m2str = NULL;
char m1c = '\0';
char m2c = '\0';
AjPStr matrix = NULL;
AjPStr dna_matrix = NULL;
AjPFile ma_matrix = NULL;
float gapc;
float gapv;
AjBool endgaps;
AjBool norgap;
AjBool nohgap;
ajint gap_dist;
ajint maxdiv;
AjPStr hgapres = NULL;
AjPSeqout fil_file = NULL;
AjPSeq seq = NULL;
AjPStr cmd = NULL;
AjPStr tmp = NULL;
AjPStr tmpFilename;
AjPStr line = NULL;
ajint nb = 0;
/* get all the parameters */
embInit("emma", argc, argv);
pwmstr = ajStrNew();
pwdstr = ajStrNew();
m1str = ajStrNew();
m2str = ajStrNew();
seqall = ajAcdGetSeqall("sequence");
seqout = ajAcdGetSeqoutset("outseq");
dend_outfile = ajAcdGetOutfile("dendoutfile");
only_dend = ajAcdGetToggle("onlydend");
use_dend = ajAcdGetToggle("dendreuse");
dend_file = ajAcdGetInfile("dendfile");
if (dend_file)
ajStrAssignS(&dend_filename, ajFileGetPrintnameS(dend_file));
ajFileClose(&dend_file);
do_slow = ajAcdGetToggle("slowalign");
ktup = ajAcdGetInt("ktup");
gapw = ajAcdGetInt("gapw");
topdiags = ajAcdGetInt("topdiags");
window = ajAcdGetInt("window");
nopercent = ajAcdGetBoolean("nopercent");
pw_matrix = ajAcdGetListSingle("pwmatrix");
pwmc = ajStrGetCharFirst(pw_matrix);
if(pwmc=='b')
ajStrAssignC(&pwmstr,"blosum");
else if(pwmc=='p')
ajStrAssignC(&pwmstr,"pam");
else if(pwmc=='g')
ajStrAssignC(&pwmstr,"gonnet");
else if(pwmc=='i')
ajStrAssignC(&pwmstr,"id");
else if(pwmc=='o')
ajStrAssignC(&pwmstr,"own");
pw_dna_matrix = ajAcdGetListSingle("pwdnamatrix");
pwdc = ajStrGetCharFirst(pw_dna_matrix);
if(pwdc=='i')
ajStrAssignC(&pwdstr,"iub");
else if(pwdc=='c')
ajStrAssignC(&pwdstr,"clustalw");
else if(pwdc=='o')
ajStrAssignC(&pwdstr,"own");
pairwise_matrix = ajAcdGetInfile("pairwisedatafile");
pw_gapc = ajAcdGetFloat( "pwgapopen");
pw_gapv = ajAcdGetFloat( "pwgapextend");
matrix = ajAcdGetListSingle( "matrix");
m1c = ajStrGetCharFirst(matrix);
if(m1c=='b')
ajStrAssignC(&m1str,"blosum");
else if(m1c=='p')
ajStrAssignC(&m1str,"pam");
else if(m1c=='g')
ajStrAssignC(&m1str,"gonnet");
else if(m1c=='i')
ajStrAssignC(&m1str,"id");
else if(m1c=='o')
ajStrAssignC(&m1str,"own");
dna_matrix = ajAcdGetListSingle( "dnamatrix");
m2c = ajStrGetCharFirst(dna_matrix);
if(m2c=='i')
ajStrAssignC(&m2str,"iub");
else if(m2c=='c')
ajStrAssignC(&m2str,"clustalw");
else if(m2c=='o')
ajStrAssignC(&m2str,"own");
ma_matrix = ajAcdGetInfile("mamatrixfile");
gapc = ajAcdGetFloat("gapopen");
gapv = ajAcdGetFloat("gapextend");
endgaps = ajAcdGetBoolean("endgaps");
norgap = ajAcdGetBoolean("norgap");
nohgap = ajAcdGetBoolean("nohgap");
gap_dist = ajAcdGetInt("gapdist");
hgapres = ajAcdGetString("hgapres");
maxdiv = ajAcdGetInt("maxdiv");
tmp = ajStrNewC("fasta");
/*
** Start by writing sequences into a unique temporary file
** get file pointer to unique file
*/
fil_file = ajSeqoutNew();
tmpFilename = emma_getUniqueFileName();
if(!ajSeqoutOpenFilename( fil_file, tmpFilename))
embExitBad();
/* Set output format to fasta */
ajSeqoutSetFormatS( fil_file, tmp);
while(ajSeqallNext(seqall, &seq))
{
/*
** Check sequences are all of the same type
** Still to be done
** Write out sequences
*/
if (!nb)
are_prot = ajSeqIsProt(seq);
ajSeqoutWriteSeq(fil_file, seq);
++nb;
}
ajSeqoutClose(fil_file);
if(nb < 2)
ajFatal("Multiple alignments need at least two sequences");
/* Generate clustalw command line */
cmd = ajStrNewS(ajAcdGetpathC("clustalw"));
/* add tmp file containing sequences */
ajStrAppendC(&cmd, " -infile=");
ajStrAppendS(&cmd, tmpFilename);
/* add out file name */
tmp_aln_outfile = emma_getUniqueFileName();
ajStrAppendC(&cmd, " -outfile=");
ajStrAppendS(&cmd, tmp_aln_outfile);
/* calculating just the nj tree or doing full alignment */
if(only_dend)
ajStrAppendC(&cmd, " -tree");
else
if(!use_dend)
ajStrAppendC(&cmd, " -align");
/* Set sequence type from information from acd file */
if(are_prot)
ajStrAppendC(&cmd, " -type=protein");
else
ajStrAppendC(&cmd, " -type=dna");
/*
** set output to MSF format - will read in this file later and output
** user requested format
*/
ajStrAppendC(&cmd, " -output=");
ajStrAppendC(&cmd, "gcg");
/* If going to do pairwise alignment */
if(!use_dend)
{
/* add fast pairwise alignments*/
if(!do_slow)
{
ajStrAppendC(&cmd, " -quicktree");
ajStrAppendC(&cmd, " -ktuple=");
ajStrFromInt(&tmp, ktup);
ajStrAppendS(&cmd, tmp);
ajStrAppendC(&cmd, " -window=");
ajStrFromInt(&tmp, window);
ajStrAppendS(&cmd, tmp);
if(nopercent)
ajStrAppendC(&cmd, " -score=percent");
else
ajStrAppendC(&cmd, " -score=absolute");
ajStrAppendC(&cmd, " -topdiags=");
ajStrFromInt(&tmp, topdiags);
ajStrAppendS(&cmd, tmp);
ajStrAppendC(&cmd, " -pairgap=");
ajStrFromInt(&tmp, gapw);
ajStrAppendS(&cmd, tmp);
}
else
{
if(pairwise_matrix)
{
if(are_prot)
ajStrAppendC(&cmd, " -pwmatrix=");
else
ajStrAppendC(&cmd, " -pwdnamatrix=");
ajStrAppendS(&cmd, ajFileGetPrintnameS(pairwise_matrix));
}
else
{
if(are_prot)
{
ajStrAppendC(&cmd, " -pwmatrix=");
ajStrAppendS(&cmd, pwmstr);
}
else
{
ajStrAppendC(&cmd, " -pwdnamatrix=");
ajStrAppendS(&cmd, pwdstr);
}
}
ajStrAppendC(&cmd, " -pwgapopen=");
ajStrFromFloat(&tmp, pw_gapc, 3);
ajStrAppendS(&cmd, tmp);
ajStrAppendC(&cmd, " -pwgapext=");
ajStrFromFloat(&tmp, pw_gapv, 3);
ajStrAppendS(&cmd, tmp);
}
}
/* Multiple alignments */
/* using existing tree or generating new tree? */
if(use_dend)
{
ajStrAppendC(&cmd, " -usetree=");
ajStrAppendS(&cmd, dend_filename);
}
else
{
/* use tmp file to hold dend file, will read back in later */
tmp_dendfilename = emma_getUniqueFileName();
ajStrAppendC(&cmd, " -newtree=");
ajStrAppendS(&cmd, tmp_dendfilename);
}
if(ma_matrix)
{
if(are_prot)
ajStrAppendC(&cmd, " -matrix=");
else
ajStrAppendC(&cmd, " -pwmatrix=");
ajStrAppendS(&cmd, ajFileGetPrintnameS(ma_matrix));
}
else
{
if(are_prot)
{
ajStrAppendC(&cmd, " -matrix=");
ajStrAppendS(&cmd, m1str);
}
else
{
ajStrAppendC(&cmd, " -dnamatrix=");
ajStrAppendS(&cmd, m2str);
}
}
ajStrAppendC(&cmd, " -gapopen=");
ajStrFromFloat(&tmp, gapc, 3);
ajStrAppendS(&cmd, tmp);
ajStrAppendC(&cmd, " -gapext=");
ajStrFromFloat(&tmp, gapv, 3);
ajStrAppendS(&cmd, tmp);
ajStrAppendC(&cmd, " -gapdist=");
ajStrFromInt(&tmp, gap_dist);
ajStrAppendS(&cmd, tmp);
ajStrAppendC(&cmd, " -hgapresidues=");
ajStrAppendS(&cmd, hgapres);
if(!endgaps)
ajStrAppendC(&cmd, " -endgaps");
if(norgap)
ajStrAppendC(&cmd, " -nopgap");
if(nohgap)
ajStrAppendC(&cmd, " -nohgap");
ajStrAppendC(&cmd, " -maxdiv=");
ajStrFromInt(&tmp, maxdiv);
ajStrAppendS(&cmd, tmp);
/* run clustalw */
/* ajFmtError("..%s..\n\n", ajStrGetPtr( cmd)); */
ajDebug("Executing '%S'\n", cmd);
ajSysExecS(cmd);
/* produce alignment file only if one was produced */
if(!only_dend)
{
/* read in tmp alignment output file to output through EMBOSS output */
seqin = ajSeqinNew();
/*
** add the Usa format to the start of the filename to tell EMBOSS
** format of file
*/
ajStrInsertC(&tmp_aln_outfile, 0, "msf::");
ajSeqinUsa(&seqin, tmp_aln_outfile);
seqset = ajSeqsetNew();
if(ajSeqsetRead(seqset, seqin))
{
ajSeqoutWriteSet(seqout, seqset);
ajSeqoutClose(seqout);
ajSeqinDel(&seqin);
/* remove the Usa from the start of the string */
ajStrCutStart(&tmp_aln_outfile, 5);
}
else
ajFmtError("Problem writing out EMBOSS alignment file\n");
}
/* read in new tmp dend file (if produced) to output through EMBOSS */
if(tmp_dendfilename!=NULL)
{
tmp_dendfile = ajFileNewInNameS( tmp_dendfilename);
if(tmp_dendfile!=NULL){
while(ajReadlineTrim(tmp_dendfile, &line))
ajFmtPrintF(dend_outfile, "%s\n", ajStrGetPtr( line));
ajFileClose(&tmp_dendfile);
ajSysFileUnlinkS(tmp_dendfilename);
}
}
ajSysFileUnlinkS(tmpFilename);
if(!only_dend)
ajSysFileUnlinkS(tmp_aln_outfile);
ajStrDel(&pw_matrix);
ajStrDel(&matrix);
ajStrDel(&pw_dna_matrix);
ajStrDel(&dna_matrix);
ajStrDel(&tmp_dendfilename);
ajStrDel(&dend_filename);
ajStrDel(&tmp_aln_outfile);
ajStrDel(&pwmstr);
ajStrDel(&pwdstr);
ajStrDel(&m1str);
ajStrDel(&m2str);
ajStrDel(&hgapres);
ajStrDel(&cmd);
ajStrDel(&tmp);
ajStrDel(&tmpFilename);
ajStrDel(&line);
ajFileClose(&dend_outfile);
ajFileClose(&tmp_dendfile);
ajFileClose(&dend_file);
ajFileClose(&pairwise_matrix);
ajFileClose(&ma_matrix);
ajSeqallDel(&seqall);
ajSeqsetDel(&seqset);
ajSeqDel(&seq);
ajSeqoutDel(&seqout);
ajSeqoutDel(&fil_file);
ajSeqinDel(&seqin);
embExit();
return 0;
}
AjPPatlistRegex ajPatlistRegexRead (const AjPStr patspec,
const AjPStr patname,
const AjPStr fmt,
ajuint type, AjBool upper, AjBool lower)
{
AjPPatlistRegex patlist = NULL;
AjPStr line = NULL;
AjPStr pat = NULL;
AjPStr name = NULL;
AjPFilebuff infile = NULL;
AjPStr patstr = NULL;
ajuint ifmt;
ajuint npat = 0;
AjPStr namestr = NULL;
ajStrAssignS(&namestr, patname);
ajStrAssignEmptyC(&namestr, "regex");
ajStrAssignS(&patstr, patspec);
patlist = ajPatlistRegexNewType(type);
ifmt = patternRegexFormat(fmt);
if(ajStrGetCharFirst(patspec) == '@')
{
ajStrCutStart(&patstr, 1);
infile = ajFilebuffNewNameS(patstr);
if(!infile)
{
ajErr("Unable to open regular expression file '%S'", patstr);
return NULL;
}
line = ajStrNew();
pat = ajStrNew();
name = ajStrNew();
if(!ifmt)
{
ajBuffreadLineTrim(infile,&line);
if(ajStrPrefixC(line, ">"))
ifmt = 2;
else
ifmt = 1;
ajFilebuffReset(infile);
}
switch(ifmt)
{
case 1:
while (ajBuffreadLineTrim(infile,&line))
{
npat++;
ajStrAppendS (&pat,line);
if(lower)
ajStrFmtLower(&pat);
if(upper)
ajStrFmtUpper(&pat);
ajFmtPrintS(&name, "%S%u", namestr, npat);
ajPatternRegexNewList(patlist,name,pat);
ajStrSetClear(&pat);
}
break;
default:
while (ajBuffreadLineTrim(infile,&line))
{
if (ajStrFindC(line,">")>-1)
{
npat++;
if (ajStrGetLen(name))
{
if(lower)
ajStrFmtLower(&pat);
if(upper)
ajStrFmtUpper(&pat);
ajPatternRegexNewList(patlist,name,pat);
ajStrSetClear(&name);
ajStrSetClear(&pat);
}
ajStrCutStart(&line,1);
ajStrAssignS (&name,line);
if(!ajStrGetLen(name))
ajFmtPrintS(&name, "%S%u", namestr, npat);
}
else
ajStrAppendS (&pat,line);
}
ajStrAssignEmptyS(&name, patname);
ajPatternRegexNewList(patlist,name,pat);
ajStrSetClear(&pat);
break;
}
ajFilebuffDel(&infile);
}
else
{
ajStrAssignS(&pat, patspec);
if(lower)
ajStrFmtLower(&pat);
if(upper)
ajStrFmtUpper(&pat);
ajStrAssignS(&name, namestr);
ajPatternRegexNewList(patlist,name,pat);
}
ajStrDel(&name);
ajStrDel(&namestr);
ajStrDel(&patstr);
ajStrDel(&line);
ajStrDel(&pat);
return patlist;
}
AjPPatlistSeq ajPatlistSeqRead (const AjPStr patspec,
const AjPStr patname,
const AjPStr fmt,
AjBool protein, ajuint mismatches)
{
AjPPatlistSeq patlist = NULL;
AjPStr line = NULL;
AjPStr name = NULL;
AjPFilebuff infile = NULL;
AjPRegexp mismreg = NULL;
AjPStr patstr = NULL;
AjPStr pat = NULL;
ajuint mismatch = 0;
ajint ifmt = 0;
ajuint npat = 0;
AjPStr namestr = NULL;
ajStrAssignS(&namestr, patname);
ajStrAssignEmptyC(&namestr, "pattern");
ajStrAssignS(&patstr, patspec);
patlist = ajPatlistSeqNewType(protein);
ifmt = patternSeqFormat(fmt);
ajDebug("ajPatlistSeqRead patspec: '%S' patname: '%S' "
"protein: %B mismatches: %d\n",
patspec, patname, protein, mismatches);
if(ajStrGetCharFirst(patstr) == '@')
{
ajStrCutStart(&patstr, 1);
infile = ajFilebuffNewNameS(patstr);
if(!infile)
{
ajErr("Unable to open pattern file '%S'", patstr);
return NULL;
}
line = ajStrNew();
name = ajStrNew();
if(!ifmt)
{
ajBuffreadLineTrim(infile,&line);
if(ajStrPrefixC(line, ">"))
ifmt = 2;
else
ifmt = 1;
ajFilebuffReset(infile);
}
switch(ifmt)
{
case 1:
while (ajBuffreadLineTrim(infile,&line))
{
npat++;
ajStrAppendS (&pat,line);
ajFmtPrintS(&name, "%S%u", namestr, npat);
ajPatternSeqNewList(patlist,name,pat,mismatches);
ajStrSetClear(&pat);
}
break;
default:
mismreg = ajRegCompC("<mismatch=(\\d+)>");
while (ajBuffreadLineTrim(infile,&line))
{
if (ajStrGetCharFirst(line) == '>')
{
if (ajStrGetLen(name))
{
ajPatternSeqNewList(patlist,name,pat,
mismatch);
ajStrSetClear(&name);
ajStrSetClear(&pat);
mismatch=mismatches;
}
ajStrCutStart(&line,1);
if (ajRegExec(mismreg,line))
{
ajRegSubI(mismreg,1,&name);
ajStrToUint(name,&mismatch);
ajStrTruncateLen(&line,ajRegOffset(mismreg));
ajStrTrimWhiteEnd(&line);
}
ajStrAssignS (&name,line);
ajStrAssignEmptyS(&name, patname);
}
else
ajStrAppendS (&pat,line);
}
ajStrAssignEmptyS(&name, patname);
ajPatternSeqNewList(patlist,name,pat,mismatch);
ajRegFree(&mismreg);
break;
}
ajFilebuffDel(&infile);
}
else
{
ajStrAssignS(&name, namestr);
ajPatternSeqNewList(patlist,name,patstr,mismatches);
}
ajStrDel(&name);
ajStrDel(&line);
ajStrDel(&pat);
ajStrDel(&namestr);
ajStrDel(&patstr);
return patlist;
}
int main(int argc, char **argv)
{
AjPSeqall nucseq; /* input nucleic sequences */
AjPSeqset protseq; /* input aligned protein sequences */
AjPSeqout seqout;
AjPSeq nseq; /* next nucleic sequence to align */
const AjPSeq pseq; /* next protein sequence use in alignment */
AjPTrn trnTable;
AjPSeq pep; /* translation of nseq */
AjPStr tablelist;
ajint table;
AjPSeqset outseqset; /* set of aligned nucleic sequences */
ajint proteinseqcount = 0;
AjPStr degapstr = NULL;
/* used to check if it matches with START removed */
AjPStr degapstr2 = NULL;
AjPStr codon = NULL; /* holds temporary codon to check if is START */
char aa; /* translated putative START codon */
ajint type; /* returned type of the putative START codon */
/* start position of guide protein in translation */
ajlong pos = 0;
AjPSeq newseq = NULL; /* output aligned nucleic sequence */
ajint frame;
embInit("tranalign", argc, argv);
nucseq = ajAcdGetSeqall("asequence");
protseq = ajAcdGetSeqset("bsequence");
tablelist = ajAcdGetListSingle("table");
seqout = ajAcdGetSeqoutset("outseq");
outseqset = ajSeqsetNew();
degapstr = ajStrNew();
/* initialise the translation table */
ajStrToInt(tablelist, &table);
trnTable = ajTrnNewI(table);
ajSeqsetFill(protseq);
while(ajSeqallNext(nucseq, &nseq))
{
if((pseq = ajSeqsetGetseqSeq(protseq, proteinseqcount++)) == NULL)
ajErr("No guide protein sequence available for "
"nucleic sequence %S",
ajSeqGetNameS(nseq));
ajDebug("Aligning %S and %S\n",
ajSeqGetNameS(nseq), ajSeqGetNameS(pseq));
/* get copy of pseq string with no gaps */
ajStrAssignS(°apstr, ajSeqGetSeqS(pseq));
ajStrRemoveGap(°apstr);
/*
** for each translation frame look for subset of pep that
** matches pseq
*/
for(frame = 1; frame <4; frame++)
{
ajDebug("trying frame %d\n", frame);
pep = ajTrnSeqOrig(trnTable, nseq, frame);
degapstr2 = ajStrNew();
ajStrAssignRef(°apstr2, degapstr);
pos = ajStrFindCaseS(ajSeqGetSeqS(pep), degapstr);
/*
** we might have a START codon that should be translated as 'M'
** we need to check if there is a match after a possible START
** codon
*/
if(pos == -1 && ajStrGetLen(degapstr) > 1 &&
(ajStrGetPtr(degapstr)[0] == 'M' ||
ajStrGetPtr(degapstr)[0] == 'm'))
{
/* see if pep minus the first character is a match */
ajStrCutStart(°apstr2, 1);
pos = ajStrFindCaseS(ajSeqGetSeqS(pep), degapstr2);
/*
** pos is >= 1 if we have a match that is after the first
** residue
*/
if(pos >= 1)
{
/* point back at the putative START Methionine */
pos--;
/* test if first codon is a START */
codon = ajStrNew();
ajStrAssignSubS(&codon, ajSeqGetSeqS(nseq),
(pos*3)+frame-1, (pos*3)+frame+2);
type = ajTrnCodonstrTypeS(trnTable, codon, &aa);
if(type != 1)
{
/* first codon is not a valid START, force a mismatch */
pos = -1;
}
ajStrDel(&codon);
}
else
{
/* force 'pos == 0' to be treated as a mismatch */
pos = -1;
}
}
ajStrDel(°apstr2);
ajSeqDel(&pep);
if(pos != -1)
break;
}
if(pos == -1)
ajErr("Guide protein sequence %S not found in nucleic sequence %S",
ajSeqGetNameS(pseq), ajSeqGetNameS(nseq));
else
{
ajDebug("got a match with frame=%d\n", frame);
/* extract the coding region of nseq with gaps */
newseq = ajSeqNew();
ajSeqSetNuc(newseq);
ajSeqAssignNameS(newseq, ajSeqGetNameS(nseq));
ajSeqAssignDescS(newseq, ajSeqGetDescS(nseq));
tranalign_AddGaps(newseq, nseq, pseq, (pos*3)+frame-1);
/* output the gapped nucleic sequence */
ajSeqsetApp(outseqset, newseq);
ajSeqDel(&newseq);
}
ajStrRemoveWhiteExcess(°apstr);
}
ajSeqoutWriteSet(seqout, outseqset);
ajSeqoutClose(seqout);
ajTrnDel(&trnTable);
ajSeqsetDel(&outseqset);
ajStrDel(°apstr);
ajStrDel(°apstr2);
ajSeqallDel(&nucseq);
ajSeqDel(&nseq);
ajSeqoutDel(&seqout);
ajSeqsetDel(&protseq);
ajStrDel(&tablelist);
embExit();
return 0;
}
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;
}
static AjBool dbxflat_ParseFastq(EmbPBtreeEntry entry, AjPFile inf)
{
AjPStr line = NULL;
ajlong pos = 0L;
ajuint seqlen = 0;
ajuint qlen = 0;
AjPStr tmpfd = NULL;
AjPStr str = NULL;
AjPStr de = NULL;
AjBool ok;
if(!dbxflat_wrdexp)
dbxflat_wrdexp = ajRegCompC("([A-Za-z0-9.:=]+)");
line = ajStrNewC("");
pos = ajFileResetPos(inf);
if(!ajReadlineTrim(inf,&line))
{
ajStrDel(&line);
return ajFalse;
}
/* first line of entry */
if(!ajStrPrefixC(line,"@"))
return ajFalse;
entry->fpos = pos;
ajStrCutStart(&line, 1);
ajStrExtractFirst(line, &de, &entry->id);
if(desfield && ajStrGetLen(de))
{
while(ajRegExec(dbxflat_wrdexp,de))
{
ajRegSubI(dbxflat_wrdexp, 1, &tmpfd);
str = ajStrNew();
ajStrAssignS(&str,tmpfd);
ajListPush(desfield->data,(void *)str);
ajRegPost(dbxflat_wrdexp, &de);
}
}
/* now read sequence */
ok = ajReadlineTrim(inf,&line);
while(ok && !ajStrPrefixC(line, "+"))
{
ajStrRemoveWhite(&line);
seqlen += MAJSTRGETLEN(line);
ok = ajReadlineTrim(inf,&line);
}
if(!ok)
return ajFalse;
ok = ajReadlineTrim(inf,&line);
while(ok)
{
qlen += MAJSTRGETLEN(line);
if(qlen < seqlen)
ok = ajReadlineTrim(inf,&line);
else
ok = ajFalse;
}
ajStrDel(&de);
ajStrDel(&tmpfd);
ajStrDel(&line);
return ajTrue;
}
