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)
{
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)
{
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)
{
AjPSeqout outseq = NULL;
AjPList list = NULL;
AjPSeq seq = NULL;
AjPStr insert = NULL;
AjPStr seqstr = NULL;
AjPStr* seqr = NULL;
AjPFile data = NULL;
ajint start = 0;
ajint length = 0;
ajint amount = 0;
ajint scmax = 0;
ajint extra = 0;
embInit("makeprotseq", argc, argv);
data = ajAcdGetInfile("pepstatsfile");
insert = ajAcdGetString("insert");
start = ajAcdGetInt("start");
length = ajAcdGetInt("length");
amount = ajAcdGetInt("amount");
outseq = ajAcdGetSeqoutall("outseq");
list = ajListstrNew();
/* this is checked by acd
if(amount <=0 || length <= 0)
ajFatal("Amount or length is 0 or less. "
"Unable to create any sequences"); */
/* if insert, make sure sequence is large enough */
if(ajStrGetLen(insert))
{
length -= ajStrGetLen(insert);
/* start= start <= 1 ? 0 : --start; */ /* checked in acd */
start--;
if(length <= 0)
ajFatal("Sequence smaller than inserted part. "
"Unable to create sequences.");
}
/* make the list of AjPStr to be used in sequence creation */
if(data)
{
ajDebug("Distribution datafile '%s' given checking type\n",
ajFileGetPrintnameC(data));
seqstr = ajStrNew();
ajReadlineTrim(data,&seqstr);
if(ajStrFindC(seqstr,"PEPSTATS") == 0)
{
makeprotseq_parse_pepstats(&list,data);
}
else
{
ajWarn("Not pepstats file. Making completely random sequences.");
makeprotseq_default_chars(&list);
}
ajStrDel(&seqstr);
ajFileClose(&data);
}
else
makeprotseq_default_chars(&list);
/* if insert, make sure type is correct */
/* typecheking code is not working, uncomment and test after it is
if(ajStrGetLen(insert))
{
seqstr = ajStrNew();
if(prot)
ajStrAssignC(&seqstr,"pureprotein");
if(!ajSeqTypeCheckS(&insert,seqstr))
ajFatal("Insert not the same sequence type as sequence itself.");
ajStrDel(&seqstr);
} */
/* array allows fast creation of a sequences */
scmax = (ajuint) ajListstrToarray(list,&seqr);
if(!scmax)
ajFatal("No strings in list. No characters to make the sequence.");
ajDebug("Distribution array done.\nscmax '%d', extra '%d', first '%S'\n",
scmax,extra,seqr[0]);
ajRandomSeed();
while(amount-- > 0)
{
seqstr = makeprotseq_random_sequence(seqr,scmax,length);
if(ajStrGetLen(insert))
ajStrInsertS(&seqstr,start,insert);
ajStrFmtLower(&seqstr);
seq = ajSeqNew();
ajSeqAssignSeqS(seq, seqstr);
ajSeqSetProt(seq);
ajSeqoutWriteSeq(outseq, seq);
ajSeqDel(&seq);
ajStrDel(&seqstr);
}
ajSeqoutClose(outseq);
ajSeqoutDel(&outseq);
ajListstrFreeData(&list);
ajStrDel(&insert);
AJFREE(seqr);
embExit();
return 0;
}
int main(ajint argc, char **argv)
{
AjPFile datafile;
AjPFile outf = NULL;
AjPSeqall seqall;
AjPSeq ajseq = NULL;
ajuint i;
ajint verb;
ajint window;
ajint pt;
ajint which;
ajint weighted;
ajint t = 0;
ajint tc = 0;
ajint mode;
ajint min_seg;
const AjPStr seqdes;
float min_P;
struct hept_pref *h;
embInit("newcoils",argc,argv);
window = ajAcdGetInt("window");
weighted = ajAcdGetInt("weighted");
verb = ajAcdGetInt("verb");
mode = ajAcdGetInt("mode");
min_P = ajAcdGetFloat("minp");
min_seg = ajAcdGetInt("minseg");
outf = ajAcdGetOutfile("outfile");
datafile = ajAcdGetDatafile("datafile");
seqall = ajAcdGetSeqall("sequence");
ajseq = ajSeqNew();
h = newcoils_read_matrix(datafile);
if(verb)
{
for(i=0; i<strlen(NCAAs); ++i)
if(NCAAs[i] != '_')
{
pt = (int)(NCAAs[i]-'A');
ajFmtPrintF(outf,"AA %c %4.2f %4.2f %4.2f %4.2f %4.2f %4.2f "
"%4.2f\n",NCAAs[i],h->m[pt][0],h->m[pt][1],
h->m[pt][2],h->m[pt][3],h->m[pt][4],
h->m[pt][5],h->m[pt][6]);
}
for(i=0; i<(ajuint)h->n; ++i)
ajFmtPrintF(outf,"Window %4d %1d %f %f %f %f %f\n",h->f[i].win,
h->f[i].w,h->f[i].m_cc,h->f[i].sd_cc,h->f[i].m_g,
h->f[i].sd_g,h->f[i].sc);
}
/* See if there is a file for the chosen window length/weight scheme */
which = -1;
for(i=0; i<(ajuint)h->n; ++i)
{
if((h->f[i].win == window) && (h->f[i].w == weighted))
{ /* match */
if(verb)
ajFmtPrintF(outf,"Found fitting data for win %4d w %d\n",
window,weighted);
which = i;
}
}
while(ajSeqallNext(seqall, &ajseq))
{
seqdes = ajSeqGetDescS(ajseq);
newcoils_pred_coils(outf,ajSeqGetSeqC(ajseq),ajSeqGetNameC(ajseq),
seqdes,h,window,
which,weighted,mode,min_P,&t,&tc,min_seg);
}
if (outf)
ajFileClose(&outf);
ajSeqDel(&ajseq);
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;
}
