static AjPTable btwisted_getdinucdata(AjPFile inf)
{
AjPStr valstr = NULL;
AjPStr key = NULL;
AjPStr line = NULL;
AjPStrTok token = NULL;
AjPTable table = NULL;
valstr = ajStrNew();
line = ajStrNew();
table = ajTablestrNewCase(20);
while(ajReadlineTrim(inf,&line))
{
if(*ajStrGetPtr(line)=='#')
continue;
token = ajStrTokenNewC(line," \n\t\r");
key = ajStrNew();
ajStrTokenNextParseC(&token," \n\t\r",&key);
valstr = ajStrNew();
ajStrTokenNextParseC(&token," \n\t\r",&valstr);
ajTablePut(table,(void *)key,(void *) valstr);
ajStrTokenDel(&token);
}
ajStrDel(&line);
return table;
}
static ajint cutgextract_readcodons(AjPFile inf, AjBool allrecords,
ajint *count)
{
static int cutidx[] =
{
42,43,46,41,45,44,26,30,31,29,27,28,48,51,47,50,
52,49,55,56,53,54,36,38,35,37, 4, 6, 3, 5,17,18,
16,15,57,59,60,58,24,25,34,33,39,40,20,19,11,12,
10, 9,63,62, 8, 7,14,13,21,23,22,32,61, 1, 0, 2
};
AjPStr line = NULL;
AjPStr value = NULL;
ajint thiscount[64];
AjPStrTok token = NULL;
ajint i;
ajint n = 0;
ajint nstops = 0;
if(!line)
{
line = ajStrNew();
value = ajStrNew();
}
if(!ajReadlineTrim(inf,&line))
ajFatal("Premature end of file");
token = ajStrTokenNewC(line," \n\t\r");
for(i=0;i<CODONS;++i)
{
ajStrTokenNextParseC(&token," \n\t\r",&value);
ajStrToInt(value,&n);
thiscount[cutidx[i]] = n;
if(i>60)
nstops += n;
}
ajStrDel(&line);
ajStrDel(&value);
ajStrTokenDel(&token);
if(!allrecords)
if(nstops > 1)
return -1;
for(i=0;i<CODONS;++i)
{
count[i] += thiscount[i];
}
return nstops;
}
static void remap_NoCutList(AjPFile outfile, const AjPTable hittable,
AjBool html, const AjPStr enzymes, AjBool blunt,
AjBool sticky, ajuint sitelen, AjBool commercial,
AjBool ambiguity, AjBool limit,
const AjPTable retable)
{
/* for iterating over hittable */
PValue value;
void **keyarray = NULL; /* array for table */
void **valarray = NULL; /* array for table */
ajint i;
/* list of enzymes that cut */
AjPList cutlist;
AjIList citer; /* iterator for cutlist */
AjPStr cutname = NULL;
AjBool found;
/* for parsing value->iso string */
AjPStrTok tok;
char tokens[] = " ,";
AjPStr code = NULL;
const char *p;
/* for reading in enzymes names */
AjPFile enzfile = NULL;
AjPStr *ea;
ajint ne; /* number of enzymes */
AjBool isall = ajTrue;
/* list of enzymes that don't cut */
AjPList nocutlist;
AjIList niter; /* iterator for nocutlist */
AjPStr nocutname = NULL;
/* count of rejected enzymes not matching criteria */
ajint rejected_count = 0;
EmbPPatRestrict enz;
/* for renaming preferred isoschizomers */
AjPList newlist;
/*
**
** Make a list of enzymes('cutlist') that hit
** including the isoschizomer names
**
*/
ajDebug("Make a list of all enzymes that cut\n");
cutlist = ajListstrNew();
nocutlist = ajListstrNew();
ajTableToarrayKeysValues(hittable, &keyarray, &valarray);
for(i = 0; keyarray[i]; i++)
{
value = (PValue) valarray[i];
cutname = ajStrNew();
ajStrAssignRef(&cutname, keyarray[i]);
ajListstrPushAppend(cutlist, cutname);
/* Add to cutlist all isoschizomers of enzymes that cut */
ajDebug("Add to cutlist all isoschizomers of enzymes that cut\n");
/* start token to parse isoschizomers names */
tok = ajStrTokenNewC(value->iso, tokens);
while(ajStrTokenNextParseC(&tok, tokens, &code))
{
cutname = ajStrNew();
ajStrAssignS(&cutname, code);
ajListstrPushAppend(cutlist, cutname);
}
ajStrTokenDel(&tok);
}
ajStrDel(&code);
AJFREE(keyarray);
AJFREE(valarray);
/*
** Read in list of enzymes ('nocutlist') - either all or
** the input enzyme list.
** Exclude those that don't match the selection criteria - count these.
*/
ajDebug("Read in a list of all input enzyme names\n");
ne = 0;
if(!enzymes)
isall = ajTrue;
else
{
/* get input list of enzymes into ea[] */
ne = ajArrCommaList(enzymes, &ea);
if(ajStrMatchCaseC(ea[0], "all"))
isall = ajTrue;
else
{
isall = ajFalse;
for(i=0; i<ne; ++i)
ajStrRemoveWhite(&ea[i]);
}
}
enzfile = ajDatafileNewInNameC(ENZDATA);
/* push all enzyme names without the required criteria onto nocutlist */
enz = embPatRestrictNew();
while(!ajFileIsEof(enzfile))
{
if(!embPatRestrictReadEntry(enz, enzfile))
continue;
/*
** If user entered explicit enzyme list, then check to see if
** this is one of that explicit list
*/
if(!isall)
{
found = AJFALSE;
for(i=0; i<ne; ++i)
if(ajStrMatchCaseS(ea[i], enz->cod))
{
found = AJTRUE;
break;
}
if(!found) /* not in the explicit list */
continue;
ajDebug("RE %S is in the input explicit list of REs\n", enz->cod);
}
/* ignore ncuts==0 as they are unknown */
if(!enz->ncuts)
{
/* number of cut positions */
ajDebug("RE %S has an unknown number of cut positions\n",
enz->cod);
continue;
}
ajDebug("RE %S has a known number of cut sites\n", enz->cod);
if(enz->len < sitelen)
{
/* recognition site length */
ajDebug("RE %S does not have a long enough recognition site\n",
enz->cod);
rejected_count++;
continue;
}
if(!blunt && enz->blunt)
{
/* blunt/sticky */
ajDebug("RE %S is blunt\n", enz->cod);
rejected_count++;
continue;
}
if(!sticky && !enz->blunt)
{
/* blunt/sticky */
ajDebug("RE %S is sticky\n", enz->cod);
rejected_count++;
continue;
}
/* commercially available enzymes have uppercase patterns */
p = ajStrGetPtr(enz->pat);
/*
** The -commercial qualifier is only used if we are searching
** through 'all' of the REBASE database - if we have specified an
** explicit list of enzymes then they are searched for whether or
** not they are commercially available
*/
if((*p >= 'a' && *p <= 'z') && commercial && isall)
{
ajDebug("RE %S is not commercial\n", enz->cod);
rejected_count++;
continue;
}
if(!ambiguity && remap_Ambiguous(enz->pat)) {
ajDebug("RE %S is ambiguous\n", enz->cod);
rejected_count++;
continue;
}
ajDebug("RE %S matches all required criteria\n", enz->cod);
code = ajStrNew();
ajStrAssignS(&code, enz->cod);
ajListstrPushAppend(nocutlist, code);
}
embPatRestrictDel(&enz);
ajFileClose(&enzfile);
for(i=0; i<ne; ++i)
if(ea[i])
ajStrDel(&ea[i]);
if(ne)
AJFREE(ea);
/*
** Change names of enzymes in the non-cutter list
** to that of preferred (prototype)
** enzyme name so that the isoschizomers of cutters
** will be removed from the
** non-cutter list in the next bit.
** Remove duplicate prototype names.
*/
if(limit)
{
newlist = ajListstrNew();
remap_RenamePreferred(nocutlist, retable, newlist);
ajListstrFreeData(&nocutlist);
nocutlist = newlist;
ajListSortUnique(nocutlist, remap_cmpcase, remap_strdel);
}
/*
** Iterate through the list of input enzymes removing those that are in
** the cutlist.
*/
ajDebug("Remove from the nocutlist all enzymes and isoschizomers "
"that cut\n");
/*
** This steps down both lists at the same time, comparing names and
** iterating to the next name in whichever list whose name compares
** alphabetically before the other. Where a match is found, the
** nocutlist item is deleted.
*/
ajListSort(nocutlist, remap_cmpcase);
ajListSort(cutlist, remap_cmpcase);
citer = ajListIterNewread(cutlist);
niter = ajListIterNew(nocutlist);
/*
while((cutname = (AjPStr)ajListIterGet(citer)) != NULL)
ajDebug("dbg cutname = %S\n", cutname);
*/
nocutname = (AjPStr)ajListIterGet(niter);
cutname = (AjPStr)ajListIterGet(citer);
ajDebug("initial cutname, nocutname: '%S' '%S'\n", cutname, nocutname);
while(nocutname != NULL && cutname != NULL)
{
i = ajStrCmpCaseS(cutname, nocutname);
ajDebug("compare cutname, nocutname: %S %S ", cutname, nocutname);
ajDebug("ajStrCmpCase=%d\n", i);
if(i == 0)
{
/* match - so remove from nocutlist */
ajDebug("ajListstrRemove %S\n", nocutname);
ajListstrIterRemove(niter);
nocutname = (AjPStr)ajListIterGet(niter);
/*
** Don't increment the cutname list pointer here
** - there may be more than one entry in the nocutname
** list with the same name because we have converted
** isoschizomers to their preferred name
*/
/* cutname = (AjPStr)ajListIterGet(citer); */
}
else if(i == -1)
/* cutlist name sorts before nocutlist name */
cutname = (AjPStr)ajListIterGet(citer);
else if(i == 1)
/* nocutlist name sorts before cutlist name */
nocutname = (AjPStr)ajListIterGet(niter);
}
ajListIterDel(&citer);
ajListIterDel(&niter);
ajListstrFreeData(&cutlist);
/* Print the resulting list of those that do not cut*/
ajDebug("Print out the list\n");
/* print the title */
if(html)
ajFmtPrintF(outfile, "<H2>");
ajFmtPrintF(outfile, "\n\n# Enzymes that do not cut\n\n");
if(html)
ajFmtPrintF(outfile, "</H2>\n");
if(html)
ajFmtPrintF(outfile, "<PRE>");
/* ajListSort(nocutlist, ajStrVcmp);*/
niter = ajListIterNewread(nocutlist);
i = 0;
while((nocutname = (AjPStr)ajListIterGet(niter)) != NULL)
{
ajFmtPrintF(outfile, "%-10S", nocutname);
/* new line after every 7 names printed */
if(i++ == 7)
{
ajFmtPrintF(outfile, "\n");
i = 0;
}
}
ajListIterDel(&niter);
/* end the output */
ajFmtPrintF(outfile, "\n");
if(html) {ajFmtPrintF(outfile, "</PRE>\n");}
/*
** Print the count of rejected enzymes
** N.B. This is the count of ALL rejected enzymes including all
** isoschizomers
*/
if(html)
ajFmtPrintF(outfile, "<H2>");
ajFmtPrintF(outfile,
"\n\n# No. of cutting enzymes which do not match the\n"
"# SITELEN, BLUNT, STICKY, COMMERCIAL, AMBIGUOUS criteria\n\n");
if(html)
ajFmtPrintF(outfile, "</H2>\n");
ajFmtPrintF(outfile, "%d\n", rejected_count);
ajDebug("Tidy up\n");
ajListstrFreeData(&nocutlist);
ajListstrFreeData(&cutlist);
return;
}
static void remap_RestrictPreferred(const AjPList l, const AjPTable t)
{
AjIList iter = NULL;
EmbPMatMatch m = NULL;
const AjPStr value = NULL;
AjPStr newiso = NULL;
AjBool found; /* name found in isoschizomer list */
/* for parsing value->iso string */
AjPStrTok tok = NULL;
char tokens[] = " ,";
AjPStr code = NULL;
iter = ajListIterNewread(l);
while((m = (EmbPMatMatch)ajListIterGet(iter)))
{
found = ajFalse;
/* get prototype name */
value = ajTableFetchS(t, m->cod);
if(value)
{
ajStrAssignC(&newiso, "");
/* parse isoschizomer names from m->iso */
ajStrTokenDel(&tok);
tok = ajStrTokenNewC(m->iso, tokens);
while(ajStrTokenNextParseC(&tok, tokens, &code))
{
if(ajStrGetLen(newiso) > 0)
ajStrAppendC(&newiso, ",");
/* found the prototype name? */
if(!ajStrCmpCaseS(code, value))
{
ajStrAppendS(&newiso, m->cod);
found = ajTrue;
}
else
ajStrAppendS(&newiso, code);
}
ajStrTokenDel(&tok);
/* if the name was not replaced, then add it in now */
if(!found)
{
if(ajStrGetLen(newiso) > 0)
ajStrAppendC(&newiso, ",");
ajStrAppendS(&newiso, m->cod);
}
ajDebug("RE: %S -> %S iso=%S newiso=%S\n", m->cod, value,
m->iso, newiso);
/* replace the old iso string with the new one */
ajStrAssignS(&m->iso, newiso);
/* rename the enzyme to the prototype name */
ajStrAssignS(&m->cod, value);
}
}
ajListIterDel(&iter);
ajStrDel(&newiso);
ajStrDel(&code);
ajStrTokenDel(&tok);
return;
}
static void eprimer3_report(AjPFile outfile, const AjPStr output,
ajint numreturn, ajint begin)
{
AjPStr line = NULL;
AjPStrTok linetokenhandle;
char eol[] = "\n\r";
AjPStrTok keytokenhandle;
char equals[] = "=";
AjPStr key = NULL;
AjPStr value = NULL;
AjBool gotsequenceid = AJFALSE;
AjPTable table;
linetokenhandle = ajStrTokenNewC(output, eol);
/* get next line of relevant results */
while(ajStrTokenNextParseC(&linetokenhandle, eol, &line))
{
if(!gotsequenceid)
{
/*
** Att the start of another sequence's results?
** Start storing the results in the table.
*/
if(ajStrCmpLenC(line, "PRIMER_SEQUENCE_ID=", 19) == 0)
{
gotsequenceid = AJTRUE;
table = ajTablestrNew(TABLEGUESS);
}
else
continue;
}
else
{
/*
** At the end of this sequence? - marked by a '=' in the primer3
** output - then output the results.
*/
if(ajStrCmpC(line, "=") == 0)
{
gotsequenceid = AJFALSE;
eprimer3_output_report(outfile, table, numreturn, begin);
ajTablestrFree(&table);
continue;
}
}
/*
** store key and value in table and parse values
** when have all of the sequence
** results in the table because the LEFT, RIGHT
** and INTERNAL results for each
** resulting primer are interleaved
*/
keytokenhandle = ajStrTokenNewC(line, equals);
key = ajStrNew();
ajStrTokenNextParse(&keytokenhandle, &key);
value = ajStrNew();
ajStrTokenNextParse(&keytokenhandle, &value);
ajDebug("key=%S\tvalue=%S\n", key, value);
ajTablePut(table,(void *)key, (void *)value);
ajStrTokenDel(&keytokenhandle);
}
ajStrDel(&line);
ajStrTokenDel(&linetokenhandle);
ajTablestrFree(&table);
return;
}
static char* cutgextract_next(AjPFile inf, const AjPStr wildspecies,
AjPStr* pspecies, AjPStr* pdoc)
{
AjPStrTok handle = NULL;
AjPStr token = NULL;
ajint i;
ajint len;
char *p = NULL;
char c;
AjBool done = ajFalse;
if(!cutgextractLine)
cutgextractLine = ajStrNew();
if(!cutgextractOrg)
cutgextractOrg = ajStrNew();
ajStrAssignC(&cutgextractLine,"");
ajStrAssignC(pdoc,"");
while (!done)
{
while(ajStrGetCharFirst(cutgextractLine) != '>')
if(!ajReadlineTrim(inf,&cutgextractLine))
return NULL;
handle = ajStrTokenNewC(cutgextractLine,"\\\n\t\r");
for(i=0;i<7;++i) {
ajStrTokenNextParseC(&handle,"\\\n\t\r",&token);
if(i==5)
{
ajStrAssignC(&cutgextractOrg,"E");
ajStrAppendS(&cutgextractOrg, token);
ajStrAssignS(pspecies, token);
if(ajStrMatchWildS(token,wildspecies))
{
done = ajTrue;
}
}
switch(i)
{
case 0:
ajStrAppendC(pdoc, "#ID ");
ajStrAppendS(pdoc, token);
ajStrAppendC(pdoc, "\n");
break;
case 1:
ajStrAppendC(pdoc, "#AC ");
ajStrAppendS(pdoc, token);
ajStrAppendC(pdoc, "\n");
break;
case 2:
ajStrAppendC(pdoc, "#FT ");
ajStrAppendS(pdoc, token);
ajStrAppendC(pdoc, "\n");
break;
case 3:
ajStrAppendC(pdoc, "#FL ");
ajStrAppendS(pdoc, token);
ajStrAppendC(pdoc, "\n");
break;
case 4:
ajStrAppendC(pdoc, "#PI ");
ajStrAppendS(pdoc, token);
ajStrAppendC(pdoc, "\n");
ajStrAssignS(&cutgextractSavepid, token);
break;
case 5:
ajStrAppendC(pdoc, "#OS ");
ajStrAppendS(pdoc, token);
ajStrAppendC(pdoc, "\n");
break;
case 6:
ajStrAppendC(pdoc, "#DE ");
ajStrAppendS(pdoc, token);
ajStrAppendC(pdoc, "\n");
break;
default:
break;
}
}
ajStrTokenDel(&handle);
ajStrDel(&token);
if(!done)
if(!ajReadlineTrim(inf,&cutgextractLine))
return NULL;
}
len = ajStrGetLen(cutgextractOrg);
p = ajStrGetuniquePtr(&cutgextractOrg);
for(i=0;i<len;++i)
{
c = p[i];
if(c=='/' || c==' ' || c=='.' || c=='\'')
p[i]='_';
}
if(p[strlen(p)-1]=='_')
p[strlen(p)-1]='\0';
ajStrDel(&token);
ajStrTokenDel(&handle);
return p;
}
