int main(int argc, char **argv)
{
ajint begin, end;
AjPSeqall seqall;
AjPSeq seq;
EmbPShow ss;
AjPFile outfile;
AjPStr tablename;
ajint table;
AjPRange uppercase;
AjPRange highlight;
AjBool threeletter;
AjBool numberseq;
AjBool nameseq;
ajint width;
ajint length;
ajint margin;
AjBool description;
ajint offset;
AjBool html;
AjPStr descriptionline;
ajint orfminsize;
AjPTrn trnTable;
AjBool translation;
AjBool reverse;
AjBool cutlist;
AjBool flat;
EmbPMatMatch mm = NULL;
AjPStr *framelist;
AjBool frames[6]; /* frames to be translated 1 to 3, -1 to -3 */
/* stuff for tables and lists of enzymes and hits */
ajint default_mincuts = 1;
ajint default_maxcuts = 2000000000;
AjPTable hittable; /* enzyme hits */
/* stuff lifted from Alan's 'restrict.c' */
AjPStr enzymes = NULL;
ajint mincuts;
ajint maxcuts;
ajint sitelen;
AjBool single;
AjBool blunt;
AjBool sticky;
AjBool ambiguity;
AjBool plasmid;
AjBool commercial;
AjBool limit;
AjBool methyl;
AjPFile enzfile = NULL;
AjPFile equfile = NULL;
AjPFile methfile = NULL;
AjPTable retable = NULL;
ajint hits;
AjPList restrictlist = NULL;
embInit("remap", argc, argv);
seqall = ajAcdGetSeqall("sequence");
outfile = ajAcdGetOutfile("outfile");
tablename = ajAcdGetListSingle("table");
uppercase = ajAcdGetRange("uppercase");
highlight = ajAcdGetRange("highlight");
threeletter = ajAcdGetBoolean("threeletter");
numberseq = ajAcdGetBoolean("number");
width = ajAcdGetInt("width");
length = ajAcdGetInt("length");
margin = ajAcdGetInt("margin");
nameseq = ajAcdGetBoolean("name");
description = ajAcdGetBoolean("description");
offset = ajAcdGetInt("offset");
html = ajAcdGetBoolean("html");
orfminsize = ajAcdGetInt("orfminsize");
translation = ajAcdGetBoolean("translation");
reverse = ajAcdGetBoolean("reverse");
cutlist = ajAcdGetBoolean("cutlist");
flat = ajAcdGetBoolean("flatreformat");
framelist = ajAcdGetList("frame");
/* restriction enzyme stuff */
mincuts = ajAcdGetInt("mincuts");
maxcuts = ajAcdGetInt("maxcuts");
sitelen = ajAcdGetInt("sitelen");
single = ajAcdGetBoolean("single");
blunt = ajAcdGetBoolean("blunt");
sticky = ajAcdGetBoolean("sticky");
ambiguity = ajAcdGetBoolean("ambiguity");
plasmid = ajAcdGetBoolean("plasmid");
commercial = ajAcdGetBoolean("commercial");
limit = ajAcdGetBoolean("limit");
enzymes = ajAcdGetString("enzymes");
methfile = ajAcdGetDatafile("mfile");
methyl = ajAcdGetBoolean("methylation");
if(!blunt && !sticky)
ajFatal("Blunt/Sticky end cutters shouldn't both be disabled.");
/* get the number of the genetic code used */
ajStrToInt(tablename, &table);
trnTable = ajTrnNewI(table);
/* read the local file of enzymes names */
remap_read_file_of_enzyme_names(&enzymes);
/* get the frames to be translated */
remap_GetFrames(framelist, frames);
while(ajSeqallNext(seqall, &seq))
{
/* get begin and end positions */
begin = ajSeqGetBegin(seq)-1;
end = ajSeqGetEnd(seq)-1;
/* do the name and description */
if(nameseq)
{
if(html)
ajFmtPrintF(outfile, "<H2>%S</H2>\n",
ajSeqGetNameS(seq));
else
ajFmtPrintF(outfile, "%S\n", ajSeqGetNameS(seq));
}
if(description)
{
/*
** wrap the description line at the width of the sequence
** plus margin
*/
if(html)
ajFmtPrintF(outfile, "<H3>%S</H3>\n",
ajSeqGetDescS(seq));
else
{
descriptionline = ajStrNew();
ajStrAssignS(&descriptionline, ajSeqGetDescS(seq));
ajStrFmtWrap(&descriptionline, width+margin);
ajFmtPrintF(outfile, "%S\n", descriptionline);
ajStrDel(&descriptionline);
}
}
/* get the restriction cut sites */
/*
** most of this is lifted from the program 'restrict.c' by Alan
** Bleasby
*/
if(single)
maxcuts=mincuts=1;
retable = ajTablestrNew(EQUGUESS);
enzfile = ajDatafileNewInNameC(ENZDATA);
if(!enzfile)
ajFatal("Cannot locate enzyme file. Run REBASEEXTRACT");
if(limit)
{
equfile = ajDatafileNewInNameC(EQUDATA);
if(!equfile)
limit = ajFalse;
else
remap_read_equiv(&equfile, &retable, commercial);
}
ajFileSeek(enzfile, 0L, 0);
restrictlist = ajListNew();
/* search for hits, but don't use mincuts and maxcuts criteria yet */
hits = embPatRestrictMatch(seq, begin+1, end+1, enzfile, methfile,
enzymes, sitelen,plasmid, ambiguity,
default_mincuts, default_maxcuts, blunt,
sticky, commercial, methyl,
restrictlist);
ajDebug("Remap found %d hits\n", hits);
if(hits)
{
/* this bit is lifted from printHits */
embPatRestrictRestrict(restrictlist, hits, !limit,
ajFalse);
if(limit)
remap_RestrictPreferred(restrictlist,retable);
}
ajFileClose(&enzfile);
ajFileClose(&methfile);
/*
** Remove those violating the mincuts and maxcuts
** criteria, but save them in hittable for printing out later.
** Keep a count of how many hits each enzyme gets in hittable.
*/
hittable = ajTablestrNewCase(TABLEGUESS);
remap_RemoveMinMax(restrictlist, hittable, mincuts, maxcuts);
/* make the Show Object */
ss = embShowNew(seq, begin, end, width, length, margin, html, offset);
if(html)
ajFmtPrintF(outfile, "<PRE>");
/* create the format to display */
embShowAddBlank(ss);
embShowAddRE(ss, 1, restrictlist, plasmid, flat);
embShowAddSeq(ss, numberseq, threeletter, uppercase, highlight);
if(!numberseq)
embShowAddTicknum(ss);
embShowAddTicks(ss);
if(reverse)
{
embShowAddComp(ss, numberseq);
embShowAddRE(ss, -1, restrictlist, plasmid, flat);
}
if(translation)
{
if(reverse)
embShowAddBlank(ss);
if(frames[0])
embShowAddTran(ss, trnTable, 1, threeletter,
numberseq, NULL, orfminsize,
AJFALSE, AJFALSE, AJFALSE, AJFALSE);
if(frames[1])
embShowAddTran(ss, trnTable, 2, threeletter,
numberseq, NULL, orfminsize,
AJFALSE, AJFALSE, AJFALSE, AJFALSE);
if(frames[2])
embShowAddTran(ss, trnTable, 3, threeletter,
numberseq, NULL, orfminsize,
AJFALSE, AJFALSE, AJFALSE, AJFALSE);
if(reverse)
{
embShowAddTicks(ss);
if(frames[5])
embShowAddTran(ss, trnTable, -3, threeletter,
numberseq, NULL, orfminsize,
AJFALSE, AJFALSE, AJFALSE, AJFALSE);
if(frames[4])
embShowAddTran(ss, trnTable, -2, threeletter,
numberseq, NULL, orfminsize,
AJFALSE, AJFALSE, AJFALSE, AJFALSE);
if(frames[3])
embShowAddTran(ss, trnTable, -1, threeletter,
numberseq, NULL, orfminsize,
AJFALSE, AJFALSE, AJFALSE, AJFALSE);
}
}
embShowPrint(outfile, ss);
/* display a list of the Enzymes that cut and don't cut */
if(cutlist)
{
remap_CutList(outfile, hittable,
limit, html, mincuts, maxcuts);
remap_NoCutList(outfile, hittable, html, enzymes, blunt,
sticky, sitelen, commercial, ambiguity,
limit, retable);
}
/* add a gratuitous newline at the end of the sequence */
ajFmtPrintF(outfile, "\n");
/* tidy up */
embShowDel(&ss);
while(ajListPop(restrictlist,(void **)&mm))
embMatMatchDel(&mm);
ajListFree(&restrictlist);
remap_DelTable(&hittable);
ajTablestrFree(&retable);
}
ajTrnDel(&trnTable);
ajSeqallDel(&seqall);
ajSeqDel(&seq);
ajFileClose(&outfile);
ajStrDel(&tablename);
ajStrDel(&enzymes);
ajStrDelarray(&framelist);
ajRangeDel(&uppercase);
ajRangeDel(&highlight);
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;
}
static AjPList silent_mismatch(const AjPStr sstr, AjPList relist,
AjPStr* tailstr,
const AjPStr sname, ajint RStotal, ajint begin,
ajint radj,AjBool rev, ajint end,
AjBool tshow)
{
PSilent res;
AjPList results;
AjPStr str; /*holds RS patterns*/
AjPStr tstr;
AjBool dummy = ajFalse; /*need a bool for ajPatClassify*/
ajint mm; /*number of mismatches*/
ajint hits; /*number of pattern matches found*/
ajint i;
ajint aw;
ajint start;
AjPList patlist = NULL; /*a list for pattern matches.
a list of ..*/
EmbPMatMatch match; /*..AjMatMatch structures*/
PRinfo rlp = NULL;
AjPStr pep = NULL; /*string to hold protein*/
AjPTrn table = NULL; /*object to hold translation table*/
str = ajStrNew();
tstr = ajStrNew();
pep = ajStrNew();
table = ajTrnNewI(0); /*0 stands for standard DNA-
see fuzztran.acd*/
results = ajListNew();
start = 1;
if(!rev) /* forward strand */
{
ajTrnSeqFrameS(table,sstr,1,&pep); /*1 stands for frame number*/
if(tshow)
{
silent_fmt_sequence("TRANSLATED SEQUENCE",
pep,tailstr,start,ajFalse);
}
}
else /* reverse strand */
{
ajStrAssignC(&tstr,ajStrGetPtr(sstr)+(end-begin+1)%3);
ajTrnSeqFrameS(table,tstr,1,&pep);
if(tshow)
{
silent_fmt_sequence("REVERSE TRANSLATED SEQUENCE",
pep,tailstr,start,ajFalse);
}
}
for(aw=0;aw<RStotal;aw++) /* loop to read in restriction
enzyme patterns */
{
/* Pop off the first RE */
ajListPop(relist,(void **)&rlp);
/* ignore unknown cut sites and zero cutters */
if(*ajStrGetPtr(rlp->site)=='?'||!rlp->ncuts)
{
ajListPushAppend(relist,(void*)rlp);
continue;
}
ajStrFmtUpper(&rlp->site); /* convert all RS to upper case */
ajStrAssignS(&str,rlp->site); /* str now holds RS patterns */
patlist = ajListNew();
if(!embPatClassify(str,&str,
&dummy,&dummy,&dummy,&dummy,&dummy,&dummy,0))
continue;
mm = 0;
hits=embPatBruteForce(sstr,str,ajFalse,ajFalse,patlist,begin+1,mm,
sname);
if(hits)
while(ajListPop(patlist,(void**)&match))
embMatMatchDel(&match);
else
{
mm = 1;
hits = embPatBruteForce(sstr,str,ajFalse,ajFalse,patlist,
begin+1,mm,sname);
if(hits)
for(i=0;i<hits;++i)
{
ajListPop(patlist,(void**)&match);
res = silent_checktrans(sstr,match,rlp,begin,radj,
rev,end);
if (res)
ajListPushAppend(results,(void *)res);
embMatMatchDel(&match);
}
}
ajListPushAppend(relist,(void *)rlp);
ajListFree(&patlist);
}
ajStrDel(&str);
ajStrDel(&tstr);
ajStrDel(&pep);
ajTrnDel(&table);
return (results);
}
static PSilent silent_checktrans(const AjPStr seq,const EmbPMatMatch match,
const PRinfo rlp, ajint begin, ajint radj,
AjBool rev, ajint end)
{
PSilent ret;
const char *p = NULL;
const char *q = NULL;
const char *s = NULL;
char *t;
const char *u;
ajint matchpos;
ajint framep;
ajint count;
AjPTrn table = NULL;
AjPStr s1 = NULL;
AjPStr s2 = NULL;
char c;
char rc;
ajint min = INT_MAX; /* Reverse sense intentional! */
ajint max = -INT_MAX;
ajint fpos;
ajint rpos;
ajint x;
AjPStr tstr = NULL;
matchpos = match->start;
fpos = matchpos;
rpos=radj-fpos-match->len;
tstr = ajStrNewS(seq);
t = ajStrGetuniquePtr(&tstr);
p = t+fpos-(begin+1);
u = q = ajStrGetPtr(rlp->site);
/* Test here for whether cut site is within sequence substring */
if(rlp->ncuts==4)
{
min = AJMIN(rlp->cut1,rlp->cut2);
max = AJMAX(rlp->cut3,rlp->cut4);
}
else if(rlp->ncuts==2)
{
min = AJMIN(rlp->cut1,rlp->cut2);
max = AJMAX(rlp->cut1,rlp->cut2);
}
else
{
ajWarn("Possibly corrupt RE file");
ajStrDel(&tstr);
return NULL;
}
if(!rev) /* forward strand */
{
if(matchpos+min<0||matchpos+max>end+1)
{
/*Cut site not in sequence range*/
ajStrDel(&tstr);
return NULL;
}
}
else /* reverse strand */
{
if(radj-matchpos-1-min>end+1||radj-matchpos-1-max<begin)
{
/*Cut site not in sequence range*/
ajStrDel(&tstr);
return NULL;
}
}
count=0;
while(ajBaseAlphaToBin(*q++) & ajBaseAlphaToBin(*p++))
++count;
/* Changed base postion */
x = fpos+count-(begin+1);
/* Where the frame starts on the reverse strand */
framep = (end-begin+1)%3;
c = t[x];
rc = u[count];
if(!rev) /* forward strand */
s = t+x-x%3;
else /* reverse strand */
s = t+x-(x-framep)%3;
table = ajTrnNewI(0);
/* translates codon pointed to by s (original seq) */
s1 = ajStrNewK(ajTrnCodonC(table,s));
t[x] = rc;
/* translates codon pointed to by s (mutated base from RS pattern */
s2 = ajStrNewK(ajTrnCodonC(table,s));
t[x] = c; /* changes mutated base in seq back to original base */
AJNEW(ret);
ret->obase = c;
ret->nbase = rc;
ret->code = ajStrNewC(ajStrGetPtr(rlp->code));
ret->site = ajStrNewC(ajStrGetPtr(rlp->site));
ret->seqaa = ajStrNewC(ajStrGetPtr(s1));
ret->reaa = ajStrNewC(ajStrGetPtr(s2));
if(ajStrMatchS(s1,s2))
ret->issilent = ajTrue;
else
ret->issilent = ajFalse;
if(!rev)
{
ret->match = matchpos;
ret->base = matchpos+count;
}
else
{
ret->match = rpos;
ret->base = rpos+match->len-1-count;
}
ajStrDel(&tstr);
ajStrDel(&s1);
ajStrDel(&s2);
ajTrnDel(&table);
return ret;
}
AjPSeqout get_orf(
AjPSeqout seqout,
AjPSeqall seqall,
AjPStr tablestr,
ajuint minsize,
ajuint maxsize,
AjPStr findstr,
AjBool methionine,
AjBool circular,
AjBool reverse,
ajint around
) {
ajint table;
ajint find;
AjPSeq seq = NULL;
AjPTrn trnTable;
AjPStr sseq = NULL; /* sequence string */
/* ORF number to append to name of sequence to create unique name */
ajint orf_no;
AjBool sense; /* ajTrue = forward sense */
ajint len;
/* initialise the translation table */
ajStrToInt(tablestr, &table);
trnTable = ajTrnNewI(table);
/* what sort of ORF are we looking for */
ajStrToInt(findstr, &find);
/*
** get the minimum size converted to protein length if storing
** protein sequences
*/
if (find == P_STOP2STOP || find == P_START2STOP || find == AROUND_START) {
minsize /= 3;
maxsize /= 3;
}
while (ajSeqallNext(seqall, &seq)) {
orf_no = 1; /* number of the next ORF */
sense = ajTrue; /* forward sense initially */
/* get the length of the sequence */
len = ajSeqGetLen(seq);
/*
** if the sequence is circular, append it to itself to triple its
** length so can deal easily with wrapped ORFs, but don't update
** len
*/
if (circular) {
ajStrAssignS(&sseq, ajSeqGetSeqS(seq));
ajStrAppendS(&sseq, ajSeqGetSeqS(seq));
ajStrAppendS(&sseq, ajSeqGetSeqS(seq));
ajSeqAssignSeqS(seq, sseq);
}
/* find the ORFs */
getorf_FindORFs(seq, len, trnTable, minsize, maxsize, seqout, sense,
circular, find, &orf_no, methionine, around, &record);
/* now reverse complement the sequence and do it again */
if (reverse) {
sense = ajFalse;
ajSeqReverseForce(seq);
getorf_FindORFs(seq, len, trnTable, minsize, maxsize, seqout, sense,
circular, find, &orf_no, methionine,
around);
}
}
ajTrnDel(&trnTable);
ajSeqDel(&seq);
ajStrDel(&sseq);
}