static void remap_RenamePreferred(const AjPList list, const AjPTable table,
AjPList newlist)
{
AjIList iter = NULL;
AjPStr key = NULL;
const AjPStr value = NULL;
AjPStr name = NULL;
iter = ajListIterNewread(list);
while((key = (AjPStr)ajListIterGet(iter)))
{
/*
** If a key-value entry found, write the new value to the new list
** else write the old key name to the new list
*/
value = ajTableFetchS(table, key);
name = ajStrNew();
if(value)
{
ajDebug("Rename: %S renamed to %S\n", key, value);
ajStrAssignS(&name, value);
}
else
{
ajDebug("Rename: %S not found\n", key);
ajStrAssignS(&name, key);
}
ajListstrPushAppend(newlist, name);
}
ajListIterDel(&iter);
return;
}
static void makeprotseq_default_chars(AjPList* list)
{
int i;
int max;
char *chars;
char seqCharProtPure[] = "ACDEFGHIKLMNPQRSTVWYacdefghiklmnpqrstvwy";
int seqCharProtPureLength = 40;
AjPStr tmp;
chars = seqCharProtPure;
max = seqCharProtPureLength;
for(i = 0; i < max; i++)
{
tmp = ajStrNew();
tmp = ajFmtStr("%c",chars[i]);
ajListstrPushAppend(*list,tmp);
}
return;
}
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;
}
/* @funcstatic acdrelations_procacdfile ***************************************
**
** Process ACD file and write new ACD file with new relations: attributes
** added (replaced if necessary).
**
** @param [r] inf [AjPFile] ACD input file
** @param [r] outf [AjPFile] ACD output file
** @param [r] P [PEdam] edam object
** @param [r] T [PKtype] ktype object
** @return [void]
** @@
******************************************************************************/
static void acdrelations_procacdfile
(AjPFile inf,
AjPFile outf,
PEdam P,
PKtype T)
{
AjPStr line = NULL;
AjPStr tok = NULL;
AjPStr acdtype = NULL;
AjPStr strtmp = NULL;
AjPList strlist = NULL;
AjPStr *strarr = NULL;
ajint nstr = 0;
/* Allocate memory */
line = ajStrNew();
tok = ajStrNew();
acdtype = ajStrNew();
strlist = ajListstrNew();
/* Read next line */
while(ajReadline(inf,&line))
{
ajFmtScanS(line, "%S", &tok);
/* Write application definition or section definition out as-is */
if(ajStrMatchC(tok, "application:") ||
ajStrMatchC(tok, "section:"))
{
ajFmtPrintF(outf, "%S", line);
while(ajReadline(inf,&line))
{
ajFmtPrintF(outf, "%S", line);
ajFmtScanS(line, "%S", &tok);
if(ajStrMatchC(tok, "]"))
break;
}
}
/* Write variables, endsection definitions and comments out as-is */
else if(ajStrMatchC(tok, "variable:") ||
ajStrMatchC(tok, "endsection:") ||
ajStrMatchC(tok, "#"))
ajFmtPrintF(outf, "%S", line);
/* Write out blank lines as-is */
else if (!ajFmtScanS(line, "%S", &tok))
ajFmtPrintF(outf, "%S", line);
/* Process data definition */
else
/* First line of data definition */
{
/* Process and write datatype line */
ajFmtPrintF(outf, "%S", line);
ajFmtScanS(line, "%S", &acdtype);
ajStrRemoveSetC(&acdtype, ":");
/* Process subsequent (attribute) lines */
while(ajReadline(inf,&line))
{
strtmp = ajStrNew();
ajStrAssignS(&strtmp, line);
ajStrRemoveWhite(&strtmp);
ajListstrPushAppend(strlist, strtmp);
ajFmtScanS(line, "%S", &tok);
/* Reached end of data definition */
if(ajStrMatchC(tok, "]"))
{
nstr = ajListstrToarray(strlist, &strarr);
/* Write relations: line */
acdrelations_writerelations(outf, acdtype, strarr, nstr, P, T);
AJFREE(strarr);
ajListstrFreeData(&strlist);
strlist = ajListstrNew();
ajFmtPrintF(outf, "%S", line);
break;
}
/* Ignore existing relations: lines */
else if(ajStrMatchC(tok, "relations:"))
continue;
ajFmtPrintF(outf, "%S", line);
}
}
}
/* Free memory */
ajStrDel(&line);
ajStrDel(&tok);
ajStrDel(&acdtype);
ajListstrFreeData(&strlist);
return;
}
static void acdrelations_readdatfile
(AjPFile inf,
PEdam *P)
{
AjPStr line = NULL;
const AjPStr tok = NULL;
const AjPStr subtok = NULL;
AjPStr strtmp = NULL;
AjPList strlist = NULL;
AjPStr acdtype = NULL;
AjPStr relations = NULL;
PEdamdat dattmp = NULL;
AjPList datlist = NULL;
if(!P)
ajFatal("Null arg error 1 in acdrelations_readdatfile");
if(!inf)
ajFatal("Null arg error 3 in acdrelations_readdatfile");
/* Allocate memory */
line = ajStrNew();
acdtype = ajStrNew();
relations = ajStrNew();
datlist = ajListNew();
/* Read data from file */
while(ajReadline(inf,&line))
{
/* Discard comment lines */
if(ajStrPrefixC(line,"#"))
continue;
/* Tokenise line, delimited by '|'.
Parse first token (ACD datatype ) */
ajStrAssignS(&acdtype, ajStrParseC(line, "|"));
/* Parse second token (EDAM relations: value ) */
ajStrAssignS(&relations, ajStrParseC(NULL, "|"));
/* Parse third token (attribute:value strings block) */
tok = ajStrParseC(NULL, "|");
/* Create new string list */
strlist = ajListstrNew();
/* Tokenise third token itself into tokens delimited by ' ' (space)
Parse tokens (individual attribute:value strings)*/
if((subtok=ajStrParseC(tok, ";")))
{
strtmp = ajStrNew();
ajStrAssignS(&strtmp, subtok);
ajStrRemoveWhite(&strtmp);
ajListstrPushAppend(strlist, strtmp);
while((subtok=ajStrParseC(NULL, ";")))
{
strtmp = ajStrNew();
ajStrAssignS(&strtmp, subtok);
ajStrRemoveWhite(&strtmp);
ajListstrPushAppend(strlist, strtmp);
}
}
/* Write PEdamdat structure & push onto list */
dattmp = ajEdamdatNew();
ajStrRemoveWhite(&acdtype);
ajStrAssignS(&dattmp->acdtype, acdtype);
ajStrAssignS(&dattmp->edam, relations);
dattmp->n = ajListstrToarray(strlist, &dattmp->acdattr);
ajListPushAppend(datlist, dattmp);
/* Clear nodes (but not strings) from string list */
ajListstrFree(&strlist);
}
/* Write PEdam structure */
((*P)->n) = ajListToarray(datlist, (void***) &((*P)->dat));
/* Free memory */
ajStrDel(&line);
ajStrDel(&acdtype);
ajStrDel(&relations);
ajListFree(&datlist);
return;
}
