static void splitter_ProcessChunk (AjPSeqout seqout, const AjPSeq seq,
ajuint start, ajuint end, const AjPStr name,
AjBool feature)
{
AjPStr str;
AjPFeattable new_feattable = NULL;
AjPSeq subseq;
ajDebug("splitter_ProcessChunk %d..%d '%S' %B\n",
start, end, name, feature);
str = ajStrNew();
subseq = ajSeqNew();
new_feattable = ajFeattableNew(name);
subseq->Fttable = new_feattable;
ajFeattableSetNuc(new_feattable);
ajStrAssignSubC(&str,ajSeqGetSeqC(seq),start,end);
ajSeqAssignSeqS(subseq,str);
if(feature)
splitter_AddSubSeqFeat(subseq->Fttable,start,end,seq);
ajSeqAssignNameS(subseq, name);
splitter_write(seqout,subseq,seq);
ajStrDel(&str);
ajSeqDel(&subseq);
return;
}
static void extractfeat_WriteOut(AjPSeqout seqout, AjPStr *featstr,
AjBool compall, AjBool sense, ajint firstpos,
ajint lastpos, ajint before, ajint after,
const AjPSeq seq, AjBool remote,
const AjPStr type,
AjBool featinname, const AjPStr describestr)
{
AjPSeq newseq = NULL;
AjPStr name = NULL; /* new name of the sequence */
AjPStr value = NULL; /* string value of start or end position */
AjPStr desc = NULL; /* sequence description */
AjBool forward = sense;
if(compall)
forward = ajFalse;
ajDebug("WriteOut %S_%d_%d [%S] %d all:%B fwd:%B remote:%B\n",
ajSeqGetNameS(seq), firstpos+1, lastpos+1, type,
ajStrGetLen(*featstr), compall, sense, remote);
/* see if there is a sequence to be written out */
if(!ajStrGetLen(*featstr))
{
ajWarn("feature %S_%d_%d [%S] "
"not written out because it has zero length\n",
ajSeqGetNameS(seq), firstpos+1, lastpos+1, type);
ajDebug("feature not written out because it has length=0 "
"(probably first time round)\n");
return;
}
/* see if must abort because there were Remote IDs in the features */
if(remote)
{
ajWarn("feature not written out because it has Remote IDs\n");
ajDebug("feature not written out because it has Remote IDs\n");
return;
}
ajDebug("feature = %d bases\n", ajStrGetLen(*featstr));
/* featstr may be edited, so it is a AjPStr* */
extractfeat_BeforeAfter (seq, featstr, firstpos, lastpos, before,
after, forward);
ajDebug("feature+before/after = %d bases\n", ajStrGetLen(*featstr));
/* set the extracted sequence */
newseq = ajSeqNew();
ajSeqAssignSeqS(newseq, *featstr);
/* create a nice name for the new sequence */
name = ajStrNew();
ajStrAppendS(&name, ajSeqGetNameS(seq));
ajStrAppendC(&name, "_");
value = ajStrNew();
ajStrFromInt(&value, firstpos+1);
ajStrAppendS(&name, value);
ajStrAppendC(&name, "_");
ajStrFromInt(&value, lastpos+1);
ajStrAppendS(&name, value);
/* add the type of feature to the name, if required */
if(featinname)
{
ajStrAppendC(&name, "_");
ajStrAppendS(&name, type);
}
ajSeqAssignNameS(newseq, name);
/* set the sequence description with the 'type' added */
desc = ajStrNew();
ajStrAppendC(&desc, "[");
ajStrAppendS(&desc, type);
ajStrAppendC(&desc, "] ");
if(ajStrGetLen(describestr))
ajStrAppendS(&desc, describestr);
ajStrAppendS(&desc, ajSeqGetDescS(seq));
ajSeqAssignDescS(newseq, desc);
/* set the type */
if(ajSeqIsNuc(seq))
ajSeqSetNuc(newseq);
else
ajSeqSetProt(newseq);
/* write the new sequence */
ajSeqoutWriteSeq(seqout, newseq);
ajSeqDel(&newseq);
ajStrDel(&name);
ajStrDel(&value);
ajStrDel(&desc);
return;
}
int main(int argc, char **argv)
{
AjPSeqset seqset = NULL;
AjPStr refseq; /* input name/number of reference sequence */
ajint nrefseq; /* numeric reference sequence */
AjPMatrix matrix; /* scoring matrix structure */
ajint **sub; /* integer scoring matrix */
AjPSeqCvt cvt = 0; /* conversion table for scoring matrix */
float identity;
ajint ident;
float fplural;
AjPStr cons;
AjPSeq consensus;
const AjPSeq ref;
const AjPSeq seq;
ajuint i;
AjBool html;
AjBool doheader;
AjBool dousa;
AjBool doname;
AjBool doseqlength;
AjBool doalignlength;
AjBool dogaps;
AjBool dogapcount;
AjBool doidcount;
AjBool dosimcount;
AjBool dodifcount;
AjBool dochange;
AjBool dodesc;
AjBool dowt;
ajint seqlength;
ajint alignlength;
ajint gaps;
ajint gapcount;
ajint idcount;
ajint simcount;
ajint difcount;
float change;
AjPFile outfile;
const AjPStr usa;
const AjPStr name;
AjPStr altusa; /* default name when the real name
is not known */
AjPStr altname;
AjPStr xxx = NULL;
embInit("infoalign", argc, argv);
seqset = ajAcdGetSeqset("sequence");
refseq = ajAcdGetString("refseq");
matrix = ajAcdGetMatrix("matrix");
ajSeqsetFill(seqset);
outfile = ajAcdGetOutfile("outfile");
html = ajAcdGetBoolean("html");
doheader = ajAcdGetBoolean("heading");
dousa = ajAcdGetBoolean("usa");
doname = ajAcdGetBoolean("name");
doseqlength = ajAcdGetBoolean("seqlength");
doalignlength = ajAcdGetBoolean("alignlength");
dogaps = ajAcdGetBoolean("gaps");
dogapcount = ajAcdGetBoolean("gapcount");
doidcount = ajAcdGetBoolean("idcount");
dosimcount = ajAcdGetBoolean("simcount");
dodifcount = ajAcdGetBoolean("diffcount");
dochange = ajAcdGetBoolean("change");
dodesc = ajAcdGetBoolean("description");
dowt = ajAcdGetBoolean("weight");
/* consensus parameters */
fplural = ajAcdGetFloat("plurality");
identity = ajAcdGetFloat("identity");
cons = ajStrNew();
consensus = ajSeqNew();
altusa = ajStrNewC("-");
altname = ajStrNewC("-");
/* get conversion table and scoring matrix */
cvt = ajMatrixGetCvt(matrix);
sub = ajMatrixGetMatrix(matrix);
/* get the number of the reference sequence */
nrefseq = infoalign_Getrefseq(refseq, seqset);
/* change the % plurality to the fraction of absolute total weight */
fplural = ajSeqsetGetTotweight(seqset) * fplural / 100;
/*
** change the % identity to the number of identical sequences at a
** position required for consensus
*/
ident = ajSeqsetGetSize(seqset) * (ajint)identity / 100;
/* get the consensus sequence */
embConsCalc(seqset, matrix, ajSeqsetGetSize(seqset), ajSeqsetGetLen(seqset),
fplural, 0.0, ident, ajFalse, &cons);
ajSeqAssignSeqS(consensus, cons);
ajSeqAssignNameS(consensus,(xxx=ajStrNewC("Consensus")));
/* get the reference sequence */
if(nrefseq == -1)
ref = consensus;
else
ref = ajSeqsetGetseqSeq(seqset, nrefseq);
/* start the HTML table */
if(html)
ajFmtPrintF(outfile,"<table border cellpadding=4 bgcolor="
"\"#FFFFF0\">\n");
/* print the header information */
if(doheader)
{
/* start the HTML table title line and output the Name header */
if(html)
ajFmtPrintF(outfile, "<tr>");
else
ajFmtPrintF(outfile, "%s", "# ");
if(dousa)
{
if(html)
ajFmtPrintF(outfile, "<th>USA</th>");
else
ajFmtPrintF(outfile, "%-16s", "USA");
}
if(doname)
{
if(html)
ajFmtPrintF(outfile, "<th>Name</th>");
else
ajFmtPrintF(outfile, "%-12s", "Name");
}
if(doseqlength)
{
if(html)
ajFmtPrintF(outfile, "<th>Sequence Length</th>");
else
ajFmtPrintF(outfile, "SeqLen\t");
}
if(doalignlength)
{
if(html)
ajFmtPrintF(outfile, "<th>Aligned Length</th>");
else
ajFmtPrintF(outfile, "AlignLen\t");
}
if(dogaps)
{
if(html)
ajFmtPrintF(outfile, "<th>Gaps</th>");
else
ajFmtPrintF(outfile, "Gaps\t");
}
if(dogapcount)
{
if(html)
ajFmtPrintF(outfile, "<th>Gap Length</th>");
else
ajFmtPrintF(outfile, "GapLen\t");
}
if(doidcount)
{
if(html)
ajFmtPrintF(outfile, "<th>Identity</th>");
else
ajFmtPrintF(outfile, "Ident\t");
}
if(dosimcount)
{
if(html)
ajFmtPrintF(outfile, "<th>Similarity</th>");
else
ajFmtPrintF(outfile, "Similar\t");
}
if(dodifcount)
{
if(html)
ajFmtPrintF(outfile, "<th>Difference</th>");
else
ajFmtPrintF(outfile, "Differ\t");
}
if(dochange)
{
if(html)
ajFmtPrintF(outfile, "<th>%% Change</th>");
else
ajFmtPrintF(outfile, "%% Change\t");
}
if(dowt)
{
if(html)
ajFmtPrintF(outfile, "<th>Weight</th>");
else
ajFmtPrintF(outfile, "Weight\t");
}
if(dodesc)
{
if(html)
ajFmtPrintF(outfile, "<th>Description</th>");
else
ajFmtPrintF(outfile, "Description");
}
/* end the HTML table title line */
if(html)
ajFmtPrintF(outfile, "</tr>\n");
else
ajFmtPrintF(outfile, "\n");
}
for(i=0; i<ajSeqsetGetSize(seqset); i++)
{
seq = ajSeqsetGetseqSeq(seqset, i);
/* get the usa ('-' if unknown) */
usa = ajSeqGetUsaS(seq);
if(ajStrGetLen(usa) == 0)
usa = altusa;
/* get the name ('-' if unknown) */
name = ajSeqGetNameS(seq);
if(ajStrGetLen(name) == 0)
name = altname;
/* get the stats from the comparison to the reference sequence */
infoalign_Compare(ref, seq, sub, cvt, &seqlength, &alignlength,
&gaps, &gapcount, &idcount, &simcount,
&difcount, &change);
/* start table line */
if(html)
ajFmtPrintF(outfile, "<tr>");
if(dousa)
infoalign_OutputStr(outfile, usa, html,
(dodesc || dowt || dochange ||
dodifcount || dosimcount ||
doidcount || dogapcount || dogaps ||
doseqlength || doalignlength || doname), 18);
if(doname)
infoalign_OutputStr(outfile, name, html,
(dodesc || dowt || dochange ||
dodifcount || dosimcount ||
doidcount || dogapcount || dogaps ||
doseqlength || doalignlength), 14);
if(doseqlength)
infoalign_OutputInt(outfile, seqlength, html,
(dodesc || dowt || dochange ||
dodifcount || dosimcount ||
doidcount || dogapcount ||
dogaps || doalignlength));
if(doalignlength)
infoalign_OutputInt(outfile, alignlength, html,
(dodesc || dowt || dochange ||
dodifcount || dosimcount ||
doidcount || dogapcount || dogaps));
if(dogaps)
infoalign_OutputInt(outfile, gaps, html,
(dodesc || dowt || dochange ||
dodifcount || dosimcount ||
doidcount || dogapcount));
if(dogapcount)
infoalign_OutputInt(outfile, gapcount, html,
(dodesc || dowt || dochange ||
dodifcount || dosimcount ||
doidcount));
if(doidcount)
infoalign_OutputInt(outfile, idcount, html,
(dodesc || dowt || dochange ||
dodifcount || dosimcount));
if(dosimcount)
infoalign_OutputInt(outfile, simcount, html,
(dodesc || dowt || dochange ||
dodifcount));
if(dodifcount)
infoalign_OutputInt(outfile, difcount, html,
(dodesc || dowt || dochange));
if(dochange)
infoalign_OutputFloat(outfile, change, html, (dodesc || dowt) );
if(dowt)
infoalign_OutputFloat(outfile, ajSeqsetGetseqWeight(seqset,i), html,
dodesc);
if(dodesc)
infoalign_OutputStr(outfile, ajSeqGetDescS(seq), html, ajFalse,
NOLIMIT);
/* end table line */
if(html)
ajFmtPrintF(outfile, "</tr>\n");
else
ajFmtPrintF(outfile, "\n");
}
/* end the HTML table */
if(html)
ajFmtPrintF(outfile, "</table>\n");
ajFileClose(&outfile);
/* tidy up */
ajStrDel(&altusa);
ajStrDel(&altname);
ajStrDel(&xxx);
ajSeqDel(&consensus);
ajSeqsetDel(&seqset);
ajStrDel(&refseq);
ajMatrixDel(&matrix);
ajStrDel(&cons);
embExit();
return 0;
}
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;
}
int main(int argc, char **argv)
{
AjPSeqout seqout;
AjPSeq seq;
AjPStr name = NULL;
AjPStr desc = NULL;
AjPStr temp = NULL;
AjBool append;
embInit("descseq", argc, argv);
seqout = ajAcdGetSeqout("outseq");
seq = ajAcdGetSeq("sequence");
append = ajAcdGetBoolean("append");
name = ajAcdGetString("name");
desc = ajAcdGetString("description");
/* if appending, then do this */
if(append)
{
/* is there a name? */
if(ajStrGetLen(name))
{
ajStrAssignS(&temp, ajSeqGetNameS(seq));
ajStrAppendS(&temp, name);
ajSeqAssignNameS(seq, temp);
}
/* is there a description? */
if(ajStrGetLen(desc))
{
ajStrAssignS(&temp, ajSeqGetDescS(seq));
ajStrAppendS(&temp, desc);
ajSeqAssignDescS(seq, temp);
}
/* otherwise, just overwrite the existing values */
}
else
{
/* is there a name? */
if(ajStrGetLen(name))
ajSeqAssignNameS(seq, name);
/* is there a description? */
if(ajStrGetLen(desc))
ajSeqAssignDescS(seq, desc);
}
ajSeqoutWriteSeq(seqout, seq);
ajSeqoutClose(seqout);
ajSeqDel(&seq);
ajSeqoutDel(&seqout);
ajStrDel(&name);
ajStrDel(&desc);
ajStrDel(&temp);
embExit();
return 0;
}
