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;
}
/* @prog seqnr **************************************************************
**
** Removes redundancy from DHF files (domain hits files) or other files of
** sequences.
**
****************************************************************************/
int main(int argc, char **argv)
{
/* Variable declarations */
AjPList in = NULL; /* Names of domain hits files (input). */
AjPStr inname = NULL; /* Full name of the current DHF file. */
AjPFile inf = NULL; /* Current DHF file. */
EmbPHitlist infhits = NULL; /* Hitlist from DHF file */
AjBool dosing = ajFalse; /* Filter using singlet sequences. */
AjPDir singlets = NULL; /* Singlets (input). */
AjBool dosets = ajFalse; /* Filter using sets of sequences. */
AjPDir insets = NULL; /* Sets (input). */
AjPStr mode = NULL; /* Mode of operation */
ajint moden = 0; /* Mode 1: single threshold for redundancy
removal, 2: lower and upper thresholds
for redundancy removal. */
float thresh = 0.0; /* Threshold for non-redundancy. */
float threshlow = 0.0; /* Threshold (lower limit). */
float threshup = 0.0; /* Threshold (upper limit). */
AjPMatrixf matrix = NULL; /* Substitution matrix. */
float gapopen = 0.0; /* Gap insertion penalty. */
float gapextend = 0.0; /* Gap extension penalty. */
AjPDirout out = NULL; /* Domain hits files (output). */
AjPFile outf = NULL; /* Current DHF file (output). */
AjBool dored = ajFalse; /* True if redundant hits are output. */
AjPDirout outred = NULL; /* DHF files for redundant hits (output).*/
AjPFile redf = NULL; /* Current DHF file redundancy (output). */
AjPStr outname = NULL; /* Name of output file (re-used). */
AjPFile logf = NULL; /* Log file pointer. */
AjBool ok = ajFalse; /* Housekeeping. */
AjPSeqset seqset = NULL; /* Seqset (re-used). */
AjPSeqin seqin = NULL; /* Seqin (re-used). */
AjPList seq_list = NULL; /* Main list for redundancy removal. */
EmbPDmxNrseq seq_tmp = NULL; /* Temp. pointer for making seq_list. */
ajint seq_siz = 0; /* Size of seq_list. */
AjPUint keep = NULL; /* 1: Sequence in seq_list was classed as
non-redundant, 0: redundant. */
AjPUint nokeep = NULL; /* Inversion of keep array. */
ajint nseqnr = 0; /* No. non-redundant seqs. in seq_list. */
AjPStr filtername= NULL; /* Name of filter file (re-used). */
AjPFile filterf = NULL; /* Current filter file. */
EmbPHitlist hitlist = NULL; /* Hitlist from input file (re-used). */
AjPScopalg scopalg = NULL; /* Scopalg from input file. */
ajint x = 0; /* Housekeeping. */
/* Read data from acd. */
embInitPV("seqnr",argc,argv,"DOMSEARCH",VERSION);
in = ajAcdGetDirlist("dhfinpath");
dosing = ajAcdGetToggle("dosing");
singlets = ajAcdGetDirectory("singletsdir");
dosets = ajAcdGetToggle("dosets");
insets = ajAcdGetDirectory("insetsdir");
mode = ajAcdGetListSingle("mode");
thresh = ajAcdGetFloat("thresh");
threshlow = ajAcdGetFloat("threshlow");
threshup = ajAcdGetFloat("threshup");
matrix = ajAcdGetMatrixf("matrix");
gapopen = ajAcdGetFloat("gapopen");
gapextend = ajAcdGetFloat("gapextend");
out = ajAcdGetOutdir("dhfoutdir");
dored = ajAcdGetToggle("dored");
outred = ajAcdGetOutdir("redoutdir");
logf = ajAcdGetOutfile("logfile");
/* Housekeeping. */
filtername = ajStrNew();
outname = ajStrNew();
if(!(ajStrToInt(mode, &moden)))
ajFatal("Could not parse ACD node option");
/* Process each DHF (input) in turn. */
while(ajListPop(in,(void **)&inname))
{
ajFmtPrint("Processing %S\n", inname);
ajFmtPrintF(logf, "//\n%S\n", inname);
seq_list = ajListNew();
keep = ajUintNew();
nokeep = ajUintNew();
/**********************************/
/* Open DHF file */
/**********************************/
if((inf = ajFileNewInNameS(inname)) == NULL)
ajFatal("Could not open DHF file %S", inname);
/* Read DHF file. */
ok = ajFalse;
if(!(infhits = embHitlistReadFasta(inf)))
{
ajWarn("embHitlistReadFasta call failed in seqnr");
ajFmtPrintF(logf, "embHitlistReadFasta call failed in seqnr\n");
/* Read sequence set instead. */
seqset = ajSeqsetNew();
seqin = ajSeqinNew();
ajSeqinUsa(&seqin, inname);
if(!(ajSeqsetRead(seqset, seqin)))
ajFatal("SeqsetRead failed in seqsearch_psialigned");
if(ajSeqsetGetSize(seqset))
ok = ajTrue;
}
else
if(infhits->N)
ok = ajTrue;
/* Close DHF file. */
ajFileClose(&inf);
/* Process empty DHF files (should never occur). */
if(!ok)
{
ajWarn("Empty input file %S\n", inname);
ajFmtPrintF(logf, "Empty input file %S\n", inname);
if(infhits)
embHitlistDel(&infhits);
if(seqset)
ajSeqsetDel(&seqset);
if(seqin)
ajSeqinDel(&seqin);
continue;
}
/* 1. Create list of sequences from the main input directory.. */
if(infhits)
{
for(x=0; x<infhits->N; x++)
{
AJNEW0(seq_tmp);
seq_tmp->Seq = ajSeqNew();
ajStrAssignS(&seq_tmp->Seq->Acc,infhits->hits[x]->Acc);
ajStrAssignS(&seq_tmp->Seq->Seq,infhits->hits[x]->Seq);
ajListPushAppend(seq_list,seq_tmp);
}
}
else
{
for(x=0;x<ajSeqsetGetSize(seqset);x++)
{
AJNEW0(seq_tmp);
seq_tmp->Seq = ajSeqNew();
ajStrAssignS(&seq_tmp->Seq->Acc, ajSeqsetGetseqAccS(seqset, x));
ajStrAssignS(&seq_tmp->Seq->Seq, ajSeqsetGetseqSeqS(seqset, x));
ajListPushAppend(seq_list,seq_tmp);
}
ajSeqsetDel(&seqset);
ajSeqinDel(&seqin);
}
/**********************************/
/* Open singlets filter file */
/**********************************/
if(dosing)
{
/* Open singlets file. */
ajStrAssignS(&filtername, inname);
ajFilenameTrimPathExt(&filtername);
ajStrInsertS(&filtername, 0, ajDirGetPath(singlets));
ajStrAppendC(&filtername, ".");
ajStrAppendS(&filtername, ajDirGetExt(singlets));
if((filterf = ajFileNewInNameS(filtername)) == NULL)
{
ajWarn("Could not open DHF file %S",
filtername);
ajFmtPrint("Could not open singlets filter file %S",
filtername);
}
else
{
/* Read DHF file. */
ok = ajFalse;
if(!(hitlist = embHitlistReadFasta(filterf)))
{
ajWarn("embHitlistReadFasta call failed in seqnr");
ajFmtPrintF(logf,
"embHitlistReadFasta call failed in seqnr\n");
/* Read sequence set instead. */
seqset = ajSeqsetNew();
seqin = ajSeqinNew();
ajSeqinUsa(&seqin, inname);
if(!(ajSeqsetRead(seqset, seqin)))
ajFatal("SeqsetRead failed in seqnr");
if(ajSeqsetGetSize(seqset))
ok = ajTrue;
}
else
if(hitlist->N)
ok = ajTrue;
/* Close DHF file. */
ajFileClose(&filterf);
/* Process empty DHF files (should never occur). */
if(!ok)
{
ajWarn("Empty singlets filter file %S\n", filtername);
ajFmtPrintF(logf, "Empty singlets filter file %S\n",
filtername);
/* No continue this time. */
}
/* 2. Add sequences from filter directories to List but mark
them up (they are considered in the redundancy calculation
but never appear in the output files). */
if(hitlist)
{
for(x=0; x<hitlist->N; x++)
{
AJNEW0(seq_tmp);
seq_tmp->Seq = ajSeqNew();
seq_tmp->Garbage = ajTrue;
ajStrAssignS(&seq_tmp->Seq->Acc,hitlist->hits[x]->Acc);
ajStrAssignS(&seq_tmp->Seq->Seq,hitlist->hits[x]->Seq);
ajListPushAppend(seq_list,seq_tmp);
}
embHitlistDel(&hitlist);
}
else
{
for(x=0;x<ajSeqsetGetSize(seqset);x++)
{
AJNEW0(seq_tmp);
seq_tmp->Seq = ajSeqNew();
seq_tmp->Garbage = ajTrue;
ajStrAssignS(&seq_tmp->Seq->Acc,
ajSeqsetGetseqAccS(seqset, x));
ajStrAssignS(&seq_tmp->Seq->Seq,
ajSeqsetGetseqSeqS(seqset, x));
ajListPushAppend(seq_list,seq_tmp);
}
ajSeqsetDel(&seqset);
ajSeqinDel(&seqin);
}
}
}
/**********************************/
/* Open sets filter file */
/**********************************/
if(dosets)
{
/* Open sets file. */
ajStrAssignS(&filtername, inname);
ajFilenameTrimPathExt(&filtername);
ajStrInsertS(&filtername, 0, ajDirGetPath(insets));
ajStrAppendC(&filtername, ".");
ajStrAppendS(&filtername, ajDirGetExt(insets));
if((filterf = ajFileNewInNameS(filtername)) == NULL)
{
ajWarn("Could not open DAF file %S", filtername);
ajFmtPrint("Could not open sets filter file %S", filtername);
}
else
{
/* Read DAF file. */
ok = ajFalse;
if(!(ajDmxScopalgRead(filterf, &scopalg)))
{
ajWarn("ajDmxScopalgRead call failed in seqnr");
ajFmtPrintF(logf,
"ajDmxScopalgRead call failed in seqnr\n");
/* Read sequence set instead. */
seqset = ajSeqsetNew();
seqin = ajSeqinNew();
ajSeqinUsa(&seqin, inname);
if(!(ajSeqsetRead(seqset, seqin)))
ajFatal("SeqsetRead failed in seqnr");
if(ajSeqsetGetSize(seqset))
ok = ajTrue;
}
else
if(scopalg->N)
ok = ajTrue;
/* Close DHF file. */
ajFileClose(&filterf);
/* Process empty DHF files (should never occur). */
if(!ok)
{
ajWarn("Empty sets filter file %S\n",
filtername);
ajFmtPrintF(logf, "Empty sets filter file %S\n",
filtername);
/* No continue this time. */
}
/* 2. Add sequences from filter directories to List but mark
them up (they are considered in the redundancy calculation
but never appear in the output files).. */
if(scopalg)
{
for(x=0; x<scopalg->N; x++)
{
AJNEW0(seq_tmp);
seq_tmp->Seq = ajSeqNew();
seq_tmp->Garbage = ajTrue;
ajStrAssignS(&seq_tmp->Seq->Acc,scopalg->Codes[x]);
ajStrAssignS(&seq_tmp->Seq->Seq,scopalg->Seqs[x]);
/* Remove gap char's & whitespace. */
ajStrRemoveGap(&seq_tmp->Seq->Seq);
ajListPushAppend(seq_list,seq_tmp);
}
ajDmxScopalgDel(&scopalg);
}
else
{
for(x=0;x<ajSeqsetGetSize(seqset);x++)
{
AJNEW0(seq_tmp);
seq_tmp->Seq = ajSeqNew();
seq_tmp->Garbage = ajTrue;
ajStrAssignS(&seq_tmp->Seq->Acc,
ajSeqsetGetseqAccS(seqset, x));
ajStrAssignS(&seq_tmp->Seq->Seq,
ajSeqsetGetseqSeqS(seqset, x));
ajListPushAppend(seq_list,seq_tmp);
}
ajSeqsetDel(&seqset);
ajSeqinDel(&seqin);
}
}
}
/* 4. Identify redundant domains.. */
if(moden == 1)
{
if((!embDmxSeqNR(seq_list, &keep, &nseqnr, matrix, gapopen,
gapextend, thresh, ajTrue)))
ajFatal("embDmxSeqNR failure in seqnr");
}
else
{
if((!embDmxSeqNRRange(seq_list, &keep, &nseqnr, matrix, gapopen,
gapextend, threshlow, threshup, ajTrue)))
ajFatal("embDmxSeqNR failure in seqnr");
}
seq_siz = ajListGetLength(seq_list);
for(x=0; x<seq_siz; x++)
if(ajUintGet(keep, x) == 1)
ajUintPut(&nokeep, x, 0);
else
ajUintPut(&nokeep, x, 1);
/* Create output files. */
ajStrAssignS(&outname, inname);
ajFilenameTrimPathExt(&outname);
outf = ajFileNewOutNameDirS(outname, out);
if(dored)
redf = ajFileNewOutNameDirS(outname, outred);
/* 5. Write non-redundant domains to main output directory.
6. If specified, write redundant domains to output directory. */
embHitlistWriteSubsetFasta(outf, infhits, keep);
if(dored)
embHitlistWriteSubsetFasta(redf, infhits, nokeep);
embHitlistDel(&infhits);
ajFileClose(&outf);
ajFileClose(&redf);
ajStrDel(&inname);
while(ajListPop(seq_list, (void **) &seq_tmp))
{
ajSeqDel(&seq_tmp->Seq);
AJFREE(seq_tmp);
}
ajListFree(&seq_list);
ajUintDel(&keep);
ajUintDel(&nokeep);
}
/* Tidy up. */
ajListFree(&in);
if(singlets)
ajDirDel(&singlets);
if(insets)
ajDirDel(&insets);
ajDiroutDel(&out);
if(outred)
ajDiroutDel(&outred);
ajFileClose(&logf);
ajMatrixfDel(&matrix);
ajStrDel(&filtername);
ajStrDel(&outname);
ajStrDel(&mode);
embExit();
return 0;
}
