static AjBool dbxflat_ParseEmbl(EmbPBtreeEntry entry, AjPFile inf)
{
AjPStr line = NULL;
ajlong pos = 0L;
line = ajStrNewC("");
while(!ajStrPrefixC(line,"//"))
{
pos = ajFileResetPos(inf);
if(!ajReadlineTrim(inf,&line))
{
ajStrDel(&line);
return ajFalse;
}
if(ajStrPrefixC(line,"ID"))
{
entry->fpos = pos;
ajFmtScanS(line,"%*S%S",&entry->id);
ajStrTrimEndC(&entry->id, ";");
/*
++global;
printf("%d. %s\n",global,ajStrGetPtr(entry->id));
*/
if(entry->do_sv)
embBtreeEmblSV(line,entry->sv);
}
if(entry->do_sv)
if(ajStrPrefixC(line,"SV") ||
ajStrPrefixC(line,"IV")) /* emblcds database format */
embBtreeEmblAC(line,entry->sv);
if(entry->do_accession)
if(ajStrPrefixC(line,"AC") ||
ajStrPrefixC(line,"PA")) /* emblcds database format */
embBtreeEmblAC(line,entry->ac);
if(entry->do_keyword)
if(ajStrPrefixC(line,"KW"))
embBtreeEmblKW(line,entry->kw,entry->kwlen);
if(entry->do_description)
if(ajStrPrefixC(line,"DE"))
embBtreeEmblDE(line,entry->de,entry->delen);
if(entry->do_taxonomy)
if(ajStrPrefixC(line,"OC") || ajStrPrefixC(line,"OS"))
embBtreeEmblTX(line,entry->tx,entry->txlen);
}
ajStrDel(&line);
return ajTrue;
}
static AjBool dbxflat_ParseEmbl(EmbPBtreeEntry entry, AjPFile inf)
{
AjPStr line = NULL;
ajlong pos = 0L;
line = ajStrNewC("");
while(!ajStrPrefixC(line,"//"))
{
pos = ajFileResetPos(inf);
if(!ajReadlineTrim(inf,&line))
{
ajStrDel(&line);
return ajFalse;
}
if(ajStrPrefixC(line,"ID"))
{
entry->fpos = pos;
ajFmtScanS(line,"%*S%S",&entry->id);
ajStrTrimEndC(&entry->id, ";");
/*
++global;
printf("%d. %s\n",global,ajStrGetPtr(entry->id));
*/
if(svfield)
embBtreeEmblSV(line,svfield->data);
}
if(svfield)
if(ajStrPrefixC(line,"SV") ||
ajStrPrefixC(line,"IV")) /* emblcds database format */
embBtreeEmblAC(line,svfield->data);
if(accfield)
if(ajStrPrefixC(line,"AC") ||
ajStrPrefixC(line,"PA")) /* emblcds database format */
embBtreeEmblAC(line,accfield->data);
if(keyfield)
if(ajStrPrefixC(line,"KW"))
embBtreeEmblKW(line,keyfield->data,keyfield->len);
if(desfield)
if(ajStrPrefixC(line,"DE"))
embBtreeEmblDE(line,desfield->data,desfield->len);
if(orgfield)
if(ajStrPrefixC(line,"OC") || ajStrPrefixC(line,"OS"))
embBtreeEmblTX(line,orgfield->data,orgfield->len);
}
ajStrDel(&line);
return ajTrue;
}
static void acdrelations_readtypefile
(AjPFile inf,
PKtype *T)
{
AjPStr line = NULL;
PKtypedat dattmp = NULL;
AjPList datlist = NULL;
if(!T)
ajFatal("Null arg error 1 in acdrelations_readtypefile");
if(!inf)
ajFatal("Null arg error 3 in acdrelations_readtypefile");
/* Allocate memory */
line = ajStrNew();
datlist = ajListNew();
/* Read data from file */
while(ajReadline(inf,&line))
{
/* Discard comment lines */
if(ajStrPrefixC(line,"#"))
continue;
/* Create object for holding line */
dattmp = ajKtypedatNew();
/* Tokenise line delimited by '|'
Parse first token (value of knowntype: attribute) */
ajStrAssignS(&dattmp->ktype, ajStrParseC(line, "|"));
ajStrRemoveSetC(&dattmp->ktype, "_");
ajStrRemoveWhite(&dattmp->ktype);
/* Parse second token (ACD datatype) */
ajStrAssignS(&dattmp->acdtype, ajStrParseC(NULL, "|"));
/* Parse third token (EDAM relations: value ) */
ajStrAssignS(&dattmp->edam, ajStrParseC(NULL, "|"));
/* Push line onto list */
ajListPushAppend(datlist, dattmp);
}
/* Write PKtype structure */
((*T)->n) = ajListToarray(datlist, (void***) &((*T)->dat));
/* Free memory */
ajStrDel(&line);
ajListFree(&datlist);
return;
}
int main(int argc, char *argv[])
{
embInitPV("genret", argc, argv, "GEMBASSY", "1.0.3");
AjPSeqall seqall;
AjPSeq seq = NULL;
AjPStr inseq = NULL;
AjPStr gene = NULL;
AjPStr access = NULL;
AjBool accid = ajTrue;
AjPStr argument = NULL;
AjPFile outfile = NULL;
AjPStr seqid = NULL;
AjPStr restid = NULL;
AjBool valid = ajFalse;
AjBool isseq = ajFalse;
AjBool isgbk = ajFalse;
AjPFilebuff buff = NULL;
AjPFile tmpfile = NULL;
AjPStr tmpname = NULL;
AjPStr regexstr = NULL;
AjPStrTok token = NULL;
AjPRegexp regex = NULL;
AjPStr url = NULL;
AjPStr base = NULL;
AjPStr head = NULL;
AjPStr line = NULL;
seqall = ajAcdGetSeqall("sequence");
access = ajAcdGetString("access");
gene = ajAcdGetString("gene");
argument = ajAcdGetString("argument");
accid = ajAcdGetBoolean("accid");
outfile = ajAcdGetOutfile("outfile");
if(
ajStrMatchC(access, "translation") ||
ajStrMatchC(access, "get_exon") ||
ajStrMatchC(access, "get_exons") ||
ajStrMatchC(access, "get_cdsseq") ||
ajStrMatchC(access, "get_gbkseq") ||
ajStrMatchC(access, "get_geneseq") ||
ajStrMatchC(access, "get_intron") ||
ajStrMatchC(access, "getseq") ||
ajStrMatchC(access, "seq") ||
ajStrMatchC(access, "around_startcodon") ||
ajStrMatchC(access, "around_stopcodon") ||
ajStrMatchC(access, "before_startcodon") ||
ajStrMatchC(access, "before_stopcodon") ||
ajStrMatchC(access, "after_startcodon") ||
ajStrMatchC(access, "after_stopcodon")
)
{
isseq = ajTrue;
}
else if(ajStrMatchC(access, "annotate") ||
ajStrMatchC(access, "output"))
{
isgbk = ajTrue;
}
else
{
ajFmtPrintF(outfile, "gene,%S\n", access);
}
base = ajStrNewC("rest.g-language.org");
ajStrExchangeCC(&argument, " ", "/");
ajStrExchangeCC(&argument, ",", "/");
ajStrExchangeCC(&argument, "\t", "/");
ajStrExchangeCC(&argument, "\r", "/");
ajStrExchangeCC(&argument, "\n", "/");
if(ajStrMatchC(gene, "*"))
{
ajStrInsertK(&gene, 0, '.');
}
if(ajStrPrefixC(gene, "@") || ajStrPrefixC(gene, "list::"))
{
ajStrExchangeCC(&gene, "@", "");
ajStrExchangeCC(&gene, "list::", "");
ajStrAssignS(&tmpname, gene);
tmpfile = ajFileNewInNameS(tmpname);
if(!tmpfile)
{
ajDie("List file (%S) open error\n", tmpname);
}
gene = ajStrNew();
while(ajReadline(tmpfile, &line))
{
ajStrAppendS(&gene, line);
}
ajFileClose(&tmpfile);
ajStrDel(&tmpname);
ajStrDel(&line);
}
tmpname = ajStrNew();
gAssignUniqueName(&tmpname);
while(ajSeqallNext(seqall, &seq))
{
inseq = ajStrNew();
if(!accid)
{
if(gFormatGenbank(seq, &inseq))
{
tmpfile = ajFileNewOutNameS(tmpname);
if(!tmpfile)
{
ajDie("Output file (%S) open error\n", tmpname);
}
ajFmtPrintF(tmpfile, "%S", inseq);
ajFileClose(&tmpfile);
ajFmtPrintS(&url, "http://%S/upload/upl.pl", base);
gFilePostSS(url, tmpname, &restid);
ajStrDel(&url);
ajSysFileUnlinkS(tmpname);
}
else
{
ajWarn("Sequence does not have features\n"
"Proceeding with sequence accession ID\n");
accid = ajTrue;
}
}
ajStrAssignS(&seqid, ajSeqGetAccS(seq));
if(ajStrGetLen(seqid) == 0)
{
ajStrAssignS(&seqid, ajSeqGetNameS(seq));
}
if(ajStrGetLen(seqid) == 0)
{
ajWarn("No valid header information\n");
}
if(accid)
{
ajStrAssignS(&restid, seqid);
if(ajStrGetLen(seqid) == 0)
{
ajDie("Cannot proceed without header with -accid\n");
}
if(!gValID(seqid))
{
ajDie("Invalid accession ID:%S, exiting\n", seqid);
}
}
url = ajStrNew();
if(isgbk)
{
ajFmtPrintS(&url, "http://%S/%S/%S", base, restid, access);
}
else
{
ajFmtPrintS(&url, "http://%S/%S/*/%S/%S", base, restid, access, argument);
}
if(!gFilebuffURLS(url, &buff))
{
ajDie("GET error from %S\n", url);
}
while(ajBuffreadLine(buff, &line))
{
if(isgbk){
ajFmtPrintF(outfile, "%S", line);
continue;
}
ajStrRemoveLastNewline(&line);
regex = ajRegCompC("^>");
if(ajRegExec(regex, line))
{
head = ajStrNew();
ajStrAssignS(&head, line);
ajStrTrimStartC(&head, ">");
valid = ajFalse;
token = ajStrTokenNewC(ajStrNewS(gene), " ,\t\r\n");
while(ajStrTokenNextParse(token, ®exstr))
{
if(ajStrGetLen(regexstr))
{
regex = ajRegComp(regexstr);
if(ajRegExec(regex, line))
{
valid = ajTrue;
if(ajStrIsAlnum(regexstr))
{
ajStrExchangeSC(&gene, regexstr, "");
}
}
ajRegFree(®ex);
}
}
}
else
{
if(valid)
{
if(isseq)
{
ajStrFmtWrap(&line, 60);
ajFmtPrintF(outfile, ">%S\n%S\n", head, line);
}
else
{
ajFmtPrintF(outfile, "%S,%S\n", head, line);
}
valid = ajFalse;
}
}
}
ajFileClose(&outfile);
ajStrDel(&restid);
ajStrDel(&seqid);
ajStrDel(&inseq);
}
ajSeqallDel(&seqall);
ajSeqDel(&seq);
ajStrDel(&access);
ajStrDel(&gene);
embExit();
}
/* @prog ssematch ***********************************************************
**
** Searches a DCF file (domain classification file) for secondary structure
** matches.
**
****************************************************************************/
int main(int argc, char **argv)
{
/* Variables declarations */
AjPFile dcfin = NULL; /* Domain classification file */
AjPFile ssin = NULL; /* Secondary structure input file*/
AjPMatrixf matrix = NULL; /* Substitution matrix */
AjPFile out_ss = NULL; /* For ss top matches*/
AjPFile out_se = NULL; /* For se top matches*/
AjPFile outfile = NULL; /* Output file*/
AjPFile logf = NULL; /* Log file */
float gapopen_sss = 0.0; /* Gap insertion penalty */
float gapopen_sse = 0.0;
float gapopen = 0.0;
float gapextend_sss = 0.0; /* Gap extension penalty */
float gapextend_sse = 0.0;
float gapextend = 0.0;
ajint max_hits = 0; /* number of top alignments to display*/
ajint mode = 0;
ajint x = 0;
AjPScop temp_scop = NULL; /* scop object pointer*/
AjPList scop_list = NULL; /* list of scop objects for entire domain
classification file */
AjIList iter = NULL;
AjPStr msg = NULL; /* Pointer to String used for messages */
AjPStr line = NULL;
AjPStr qse = NULL; /* query secondary structure elements*/
AjPStr qss = NULL; /* query secondary structure (by residue)*/
AjPSeq q3se = NULL; /* query secondary structure elements, 3-letter
code*/
AjPSeq q3ss = NULL; /* query secondary structure
(by residue), 3-letter code*/
AjPSeq query = NULL;
/* Read data from acd */
embInitPV("ssematch",argc,argv,"DOMAINATRIX",VERSION);
dcfin = ajAcdGetInfile("dcfinfile");
ssin = ajAcdGetInfile("ssinfile");
max_hits = ajAcdGetInt("maxhits");
matrix = ajAcdGetMatrixf("datafile");
gapopen_sss = ajAcdGetFloat("rgapopen");
gapextend_sss = ajAcdGetFloat("rgapextend");
gapopen_sse = ajAcdGetFloat("egapopen");
gapextend_sse = ajAcdGetFloat("egapextend");
out_ss = ajAcdGetOutfile("outssfile");
out_se = ajAcdGetOutfile("outsefile");
logf = ajAcdGetOutfile("logfile");
/* Create list of scop objects for entire input domain classification file. */
scop_list = ajListNew();
while((temp_scop = (ajScopReadCNew(dcfin, "*"))))
ajListPushAppend(scop_list,temp_scop);
/* Error handing if domain classification file was empty. */
if(!(ajListGetLength(scop_list)))
{
ajWarn("Empty list from scop input file\n");
ajFileClose(&dcfin);
ajFileClose(&ssin);
ajMatrixfDel(&matrix);
ajFileClose(&out_ss);
ajFileClose(&out_se);
ajFileClose(&logf);
while(ajListPop(scop_list, (void *) &temp_scop))
ajScopDel(&temp_scop);
ajListFree(&scop_list);
ajListIterDel(&iter);
ajExit();
return 1;
}
/* Error handling in case of empty query file. */
if(ssin == NULL)
{
ajWarn("Empty secondary structure query file\n");
ajFileClose(&dcfin);
ajFileClose(&ssin);
ajMatrixfDel(&matrix);
ajFileClose(&out_ss);
ajFileClose(&out_se);
ajFileClose(&logf);
while(ajListPop(scop_list, (void *) &temp_scop))
ajScopDel(&temp_scop);
ajListFree(&scop_list);
ajListIterDel(&iter);
ajExit();
return 1;
}
/* Assign sequences in query file to sequence objects. */
qse = ajStrNew();
qss = ajStrNew();
while(ajReadlineTrim(ssin,&line))
{
/* SE string */
if(ajStrPrefixC(line,"SE"))
{
ajFmtScanS(line, "%*s %S", &qse);
/* Convert this string to 3-letter code & then convert to AjPSeq
object. */
q3se = ssematch_convertbases(qse);
}
/* SS string */
else if(ajStrPrefixC(line,"SS"))
{
while((ajReadlineTrim(ssin,&line)) && !ajStrPrefixC(line,"XX"))
ajStrAppendS(&qss,line);
ajStrRemoveWhite(&qss);
/* Convert this string to 3-letter code & then to AjPSeq object. */
q3ss = ssematch_convertbases(qss);
}
}
/* For se & then for ss, modes 0 & 1. */
for(mode = 0; mode <= 1; mode++)
{
/* Assign arguments for alignment function. */
if (mode == 0)
{
query = q3se;
gapopen = gapopen_sse;
gapextend = gapextend_sse;
outfile = out_se;
}
else if(mode == 1)
{
query = q3ss;
gapopen = gapopen_sss;
gapextend = gapextend_sss;
outfile = out_ss;
}
/* Iterate through list of scop objects & calculate alignment scores. */
iter=ajListIterNew(scop_list);
while((temp_scop=(AjPScop)ajListIterGet(iter)))
{
/* The function extracts the se (mode 0) or ss (mode 1) subject
sequences from the scop object, performs a Needleman-Wunsch
global alignment with the query sequence & allocates the score
to the Score element of the scop object*/
if(!(ssematch_NWScore(temp_scop , query, mode, matrix, gapopen, gapextend)))
{
ajFmtPrintF(logf, "%-15s\n", "ALIGNMENT");
ajFmtPrintF(logf, "Could not align sequence in scop domain %S\n ",
temp_scop->Entry);
ajFmtPrintS(&msg, "Could not align sequence in scop domain %S\n ",
temp_scop->Entry);
ajWarn(ajStrGetPtr(msg));
continue;
}
}
ajListIterDel(&iter);
temp_scop = NULL;
/* Sort list of Scop objects by Score */
ajListSort(scop_list, ssematch_CompScoreInv);
iter=ajListIterNew(scop_list);
/* Write top-scoring hits to outfile. */
for(x=0; x < max_hits; x++ )
{
temp_scop=(AjPScop)ajListIterGet(iter);
/* Print score to output file. */
ajFmtPrintF(outfile, "XX ALIGNMENT SCORE %.3f\nXX\n", temp_scop->Score);
/* Could also write alignment - later modification. */
if(!ajScopWrite(outfile, temp_scop))
ajFatal("Could not write output file %S\n", outfile);
}
ajListIterDel(&iter);
temp_scop = NULL;
}
/* Memoryt management. */
ajFileClose(&dcfin);
ajFileClose(&ssin);
ajMatrixfDel(&matrix);
ajFileClose(&out_ss);
ajFileClose(&out_se);
ajFileClose(&logf);
while(ajListPop(scop_list, (void *) &temp_scop))
ajScopDel(&temp_scop);
ajListFree(&scop_list);
ajStrDel(&msg);
ajStrDel(&line);
ajStrDel(&qse);
ajStrDel(&qss);
ajSeqDel(&q3se);
ajSeqDel(&q3ss);
ajExit();
return 0;
}
static void domainalign_ProcessStampFile(AjPStr in,
AjPStr out,
AjPDomain domain,
ajint noden,
AjPFile logf)
{
AjPFile outf = NULL; /* Output file pointer. */
AjPFile inf = NULL; /* Input file pointer. */
AjPStr temp1 = NULL; /* Temporary string. */
AjPStr temp2 = NULL; /* Temporary string. */
AjPStr temp3 = NULL; /* Temporary string. */
AjPStr line = NULL; /* Line of text from input file. */
ajint blk = 1; /* Count of the current block in the input file.
Block 1 is the numbering and protein sequences,
Block 2 is the secondary structure,
Block 3 is the Very/Less/Post similar records*/
AjBool ok = ajFalse;
/* Initialise strings. */
line = ajStrNew();
temp1 = ajStrNew();
temp2 = ajStrNew();
temp3 = ajStrNew();
/* Open input and output files. */
if(!(inf=ajFileNewInNameS(in)))
ajFatal("Could not open input file in domainalign_ProcessStampFile");
/* Start of code for reading input file.
Ignore everything up to first line beginning with 'Number'. */
while((ajReadlineTrim(inf,&line)))
{
/* ajFileReadLine will trim the tailing \n. */
if((ajStrGetCharPos(line, 1)=='\0'))
{
ok = ajTrue;
break;
}
}
/* Read rest of input file. */
if(ok)
{
/* Write DOMAIN classification records to file. */
if(!(outf=ajFileNewOutNameS(out)))
ajFatal("Could not open output file in domainalign_ProcessStampFile");
if((domain->Type == ajSCOP))
{
ajFmtPrintF(outf,"# TY SCOP\n# XX\n");
ajFmtPrintF(outf,"# CL %S",domain->Scop->Class);
ajFmtPrintSplit(outf,domain->Scop->Fold,"\n# XX\n# FO ",
75," \t\n\r");
ajFmtPrintSplit(outf,domain->Scop->Superfamily,"# XX\n# SF ",
75," \t\n\r");
ajFmtPrintSplit(outf,domain->Scop->Family,"# XX\n# FA ",
75," \t\n\r");
ajFmtPrintF(outf,"# XX\n");
}
else
{
ajFmtPrintF(outf,"# TY CATH\n# XX\n");
ajFmtPrintF(outf,"# CL %S",domain->Cath->Class);
ajFmtPrintSplit(outf,domain->Cath->Architecture,"\n# XX\n# AR ",
75," \t\n\r");
ajFmtPrintSplit(outf,domain->Cath->Topology,"# XX\n# TP ",75,
" \t\n\r");
ajFmtPrintSplit(outf,domain->Cath->Superfamily,"# XX\n# SF ",75,
" \t\n\r");
ajFmtPrintF(outf,"# XX\n");
}
if((domain->Type == ajSCOP))
{
if(noden==1)
ajFmtPrintF(outf,"# SI %d\n# XX",domain->Scop->Sunid_Class);
else if(noden==2)
ajFmtPrintF(outf,"# SI %d\n# XX",domain->Scop->Sunid_Fold);
else if(noden==3)
ajFmtPrintF(outf,"# SI %d\n# XX",
domain->Scop->Sunid_Superfamily);
else if(noden==4)
ajFmtPrintF(outf,"# SI %d\n# XX",
domain->Scop->Sunid_Family);
else
ajFatal("Node number error in domainalign_ProcessStampFile");
}
else
{
if(noden==5)
ajFmtPrintF(outf,"# SI %d\n# XX", domain->Cath->Class_Id);
else if(noden==6)
ajFmtPrintF(outf,"# SI %d\n# XX", domain->Cath->Arch_Id);
else if(noden==7)
ajFmtPrintF(outf,"# SI %d\n# XX",domain->Cath->Topology_Id);
else if(noden==8)
ajFmtPrintF(outf,"# SI %d\n# XX",
domain->Cath->Superfamily_Id);
else if(noden==9)
ajFmtPrintF(outf,"# SI %d\n# XX",domain->Cath->Family_Id);
else
ajFatal("Node number error in domainalign_ProcessStampFile");
}
while((ajReadlineTrim(inf,&line)))
{
/* Increment counter for block of file. */
if((ajStrGetCharPos(line, 1)=='\0'))
{
blk++;
if(blk==4)
blk=1;
continue;
}
/* Block of numbering line and protein sequences. */
if(blk==1)
{
/* Print the number line out as it is. */
if(ajStrPrefixC(line,"Number"))
ajFmtPrintF(outf,"\n# %7s %S\n"," ", line);
else
{
/* Read only the 7 characters
of the domain identifier
code in. */
ajFmtScanS(line, "%S", &temp1);
ajStrAssignSubS(&temp2, temp1, 0, 6);
/* Read the sequence. */
ajStrAssignSubS(&temp3, line, 13, 69);
ajStrExchangeSetCC(&temp3, " ", "X");
ajFmtPrintF(logf, "Replaced ' ' in STAMP alignment "
"with 'X'\n");
ajStrFmtUpper(&temp3);
/* Write domain id code and sequence out. */
ajFmtPrintF(outf,"%-15S%7d %S%7d\n",
temp2, 0, temp3, 0);
}
}
/* Secondary structure filled with '????' (unwanted). */
else if(blk==2)
{
continue;
}
/* Similarity lines. */
else
{
if(ajStrPrefixC(line,"Post"))
{
/* Read the sequence. */
ajStrAssignSubS(&temp3, line, 13, 69);
/* Write post similar line out. */
ajFmtPrintF(outf,"%-15s%7s %S\n","# Post_similar", " ",
temp3);
}
/* Ignore Very and Less similar lines. */
else continue;
}
}
}
else /* ok == ajFalse. */
{
ajWarn("\n***********************************************\n"
"* STAMP was called but output file was EMPTY! *\n"
"* NO OUTPUT FILE GENERATED FOR THIS NODE. *\n"
"***********************************************\n");
ajFmtPrintF(logf, "STAMP called but output file empty. "
"No output file for this node!");
}
/* Clean up and close input and output files. */
ajFileClose(&outf);
ajFileClose(&inf);
ajStrDel(&line);
ajStrDel(&temp1);
ajStrDel(&temp2);
ajStrDel(&temp3);
/* All done. */
return;
}
/* @funcstatic domainalign_ProcessTcoffeeFile *********************************
**
** Parses tcoffee output.
**
** @param [r] in [AjPStr] Name of TCOFFEE input file
** @param [r] align [AjPStr] Name of sequence alignment file for output
** @param [r] domain [AjPDomain] Domain being aligned
** @param [r] noden [ajint] Node-level of alignment**
** @param [r] logf [AjPFile] Log file
**
** @return [void] True on success
** @@
****************************************************************************/
static void domainalign_ProcessTcoffeeFile(AjPStr in,
AjPStr align,
AjPDomain domain,
ajint noden,
AjPFile logf)
{
AjPFile outf = NULL; /* Output file pointer. */
AjPFile inf = NULL; /* Input file pointer. */
AjPStr temp1 = NULL; /* Temporary string. */
AjPStr temp2 = NULL; /* Temporary string. */
AjPStr temp3 = NULL; /* Temporary string. */
AjPStr line = NULL; /* Line of text from input file. */
/* Initialise strings. */
line = ajStrNew();
temp1 = ajStrNew();
temp2 = ajStrNew();
temp3 = ajStrNew();
/* Open input and output files. */
if(!(inf=ajFileNewInNameS(in)))
ajFatal("Could not open input file in domainalign_ProcessTcoffeeFile");
if(!(outf=ajFileNewOutNameS(align)))
ajFatal("Could not open output file in domainalign_ProcessTcoffeeFile");
/*Write DOMAIN classification records to file*/
if((domain->Type == ajSCOP))
{
ajFmtPrintF(outf,"# TY SCOP\n# XX\n");
ajFmtPrintF(outf,"# CL %S",domain->Scop->Class);
ajFmtPrintSplit(outf,domain->Scop->Fold,"\n# XX\n# FO ",75," \t\n\r");
ajFmtPrintSplit(outf,domain->Scop->Superfamily,"# XX\n# SF ",75," \t\n\r");
ajFmtPrintSplit(outf,domain->Scop->Family,"# XX\n# FA ",75," \t\n\r");
ajFmtPrintF(outf,"# XX\n");
}
else
{
ajFmtPrintF(outf,"# TY CATH\n# XX\n");
ajFmtPrintF(outf,"# CL %S",domain->Cath->Class);
ajFmtPrintSplit(outf,domain->Cath->Architecture,"\n# XX\n# AR ",75," \t\n\r");
ajFmtPrintSplit(outf,domain->Cath->Topology,"# XX\n# TP ",75," \t\n\r");
ajFmtPrintSplit(outf,domain->Cath->Superfamily,"# XX\n# SF ",75," \t\n\r");
ajFmtPrintF(outf,"# XX\n");
}
if((domain->Type == ajSCOP))
{
if(noden==1)
ajFmtPrintF(outf,"# SI %d\n# XX\n",domain->Scop->Sunid_Class);
else if(noden==2)
ajFmtPrintF(outf,"# SI %d\n# XX\n",domain->Scop->Sunid_Fold);
else if(noden==3)
ajFmtPrintF(outf,"# SI %d\n# XX\n",domain->Scop->Sunid_Superfamily);
else if(noden==4)
ajFmtPrintF(outf,"# SI %d\n# XX\n", domain->Scop->Sunid_Family);
else
ajFatal("Node number error in domainalign_ProcessStampFile");
}
else
{
if(noden==5)
ajFmtPrintF(outf,"# SI %d\n# XX\n", domain->Cath->Class_Id);
else if(noden==6)
ajFmtPrintF(outf,"# SI %d\n# XX\n", domain->Cath->Arch_Id);
else if(noden==7)
ajFmtPrintF(outf,"# SI %d\n# XX\n",domain->Cath->Topology_Id);
else if(noden==8)
ajFmtPrintF(outf,"# SI %d\n# XX\n",domain->Cath->Superfamily_Id);
else if(noden==9)
ajFmtPrintF(outf,"# SI %d\n# XX\n",domain->Cath->Family_Id);
else
ajFatal("Node number error in domainalign_ProcessStampFile");
}
/* Start of code for reading input file. */
/*Ignore everything up to first line beginning with 'Number'*/
while((ajReadlineTrim(inf,&line)))
/* ajFileReadLine will trim the tailing \n. */
if((ajStrGetCharPos(line, 1)=='\0'))
break;
/* Read rest of input file. */
while((ajReadlineTrim(inf,&line)))
{
if((ajStrGetCharPos(line, 1)=='\0'))
continue;
/* Print the number line out as it is. */
else if(ajStrPrefixC(line,"CLUSTAL"))
continue;
else if(ajStrPrefixC(line," "))
ajFmtPrintF(outf,"\n");
/* write out a block of protein sequences. */
else
{
/* Read only the 7 characters of the domain identifier code in. */
ajFmtScanS(line, "%S %S", &temp1,&temp3);
ajStrAssignSubS(&temp2, temp1, 0, 6);
/* Read the sequence
ajStrAssignSubS(&temp3, line, 13, 69);
ajStrExchangeSetCC(&temp3, " ", "X");
ajStrFmtUpper(&temp3);*/
/* Write domain id code and sequence out. */
ajFmtPrintF(outf,"%-13S%S\n",temp2, temp3);
}
}
/* Clean up and close input and output files. */
ajFileClose(&outf);
ajFileClose(&inf);
ajStrDel(&line);
ajStrDel(&temp1);
ajStrDel(&temp2);
ajStrDel(&temp3);
/* All done. */
return;
}
int main(ajint argc, char **argv)
{
AjPList ccfin = NULL; /* List of CCF (input) files. */
AjPDir pdbin = NULL; /* Path of pdb input files. */
AjPStr pdbprefix = NULL; /* Prefix of pdb input files. */
AjPStr pdb_name = NULL; /* Full name (path/name/extension) of
pdb format input file. */
AjPDirout ccfout = NULL; /* Path of coordinate output file. */
AjPStr randomname = NULL; /* Name for temp file tempf. */
AjPStr ccf_this = NULL;
AjPStr exec = NULL;
AjPStr naccess_str = NULL;
AjPStr line = NULL;
AjPStr syscmd = NULL; /* Command line arguments. */
AjPStr *mode = NULL; /* Mode of operation from acd. */
AjPFile errf = NULL; /* pdbplus error file pointer. */
AjPFile serrf = NULL; /* stride error file pointer. */
AjPFile nerrf = NULL; /* stride error file pointer. */
AjPFile tempf = NULL; /* Temp file for holding STRIDE output. */
AjPFile ccf_inf = NULL; /* Protein coordinate input file. */
AjPFile ccf_outf = NULL; /* Protein coordinate output file. */
AjIList iter = NULL;
AjBool done_naccess= ajFalse;
AjBool done_stride = ajFalse;
AjBool found = ajFalse;
AjPResidue temp_res = NULL; /* Pointer to Residue object. */
AjPPdb pdb_old = NULL; /* Pointer to PDB object - without new
stride elements. */
AjPPdb pdb = NULL; /* Pointer to PDB object. */
ajint idn = 0; /* Chain identifier as a number (1,2,...) */
ajint chain_num = 0; /* Chain identifier index (0,1,...). */
ajint tS = 0; /* User-defined threshold size for SSEs. */
ajint nostride = 0; /* No. times stride failed */
ajint nonaccess = 0; /* No. times naccess failed */
ajint nofile = 0; /* No. times of file error */
/* Variables for each item that will be parsed from the ASG line. */
AjPStr res = NULL; /* Residue id from STRIDE ASG line (ALA etc). */
AjPStr res_num = NULL; /* PDB residue number from STRIDE ASG line. */
char pcid = ' '; /* Protein chain identifier from STRIDE or
NACESS output (A,B, etc). */
char ss = ' '; /* One-letter secondary structure code from
STRIDE ASG line. */
float ph = 0.0; /* Phi angle from STRIDE ASG line. */
float ps = 0.0; /* Psi angle from STRIDE ASG line. */
float sa = 0.0; /* Residue solvent accessible area from STRIDE
ASG line. */
float f1 = 0;
float f2 = 0;
float f3 = 0;
float f4 = 0;
float f5 = 0;
float f6 = 0;
float f7 = 0;
float f8 = 0;
float f9 = 0;
float f10 = 0;
/* Allocate strings; this section is used for variables that are
allocated once only. */
pdb_name = ajStrNew();
res = ajStrNew();
res_num = ajStrNew();
randomname = ajStrNew();
syscmd = ajStrNew();
line = ajStrNew();
naccess_str = ajStrNew();
exec = ajStrNew();
/* Read data from acd. */
embInitPV("pdbplus",argc,argv,"STRUCTURE",VERSION);
ccfin = ajAcdGetDirlist("ccfinpath");
pdbin = ajAcdGetDirectory("pdbindir");
pdbprefix = ajAcdGetString("pdbprefix");
ccfout = ajAcdGetOutdir("ccfoutdir");
mode = ajAcdGetList("mode");
errf = ajAcdGetOutfile("logfile");
if(ajStrGetCharFirst(*mode) != '2')
serrf = ajAcdGetOutfile("slogfile");
if(ajStrGetCharFirst(*mode) != '1')
nerrf = ajAcdGetOutfile("nlogfile");
tS = ajAcdGetInt("thresholdsize");
ajRandomSeed();
ajFilenameSetTempname(&randomname);
/*
** Start of main application loop.
** Process each PDB/ protein coordinate file (EMBL format) in turn.
*/
while(ajListPop(ccfin,(void **)&ccf_this))
{
/* Open protein coordinate file. If it cannot be opened, write a
message to the error file, delete ccf_this and continue. */
if((ccf_inf = ajFileNewInNameS(ccf_this)) == NULL)
{
ajWarn("%s%S\n//\n",
"clean coordinate file not found: ", ccf_this);
ajFmtPrintF(errf, "%s%S\n//\n",
"clean coordinate file not found: ", ccf_this);
ajStrDel(&ccf_this);
nofile++;
continue;
}
ajFmtPrint("Processing %S\n", ccf_this);
fflush(stdout);
/* Parse protein coordinate data (from clean format file) into
AjPPdb object. ajPdbReadAllModelsNew will create the AjPPdb object. */
if(!(pdb_old=ajPdbReadAllModelsNew(ccf_inf)))
{
ajWarn("ERROR Clean coordinate file read"
"error: %S\n//\n", ccf_this);
ajFmtPrintF(errf, "ERROR Clean coordinate file read"
"error: %S\n//\n", ccf_this);
ajFileClose(&ccf_inf);
ajStrDel(&ccf_this);
nofile++;
continue;
}
ajFileClose(&ccf_inf);
ajPdbCopy(&pdb, pdb_old);
ajPdbDel(&pdb_old);
/* Construct name of corresponding PDB file.
NACCESS does *not* generate an output file if the path is './' e.g.
naccess ./1rbp.ent , therefore replace './' with null. */
ajStrAssignS(&pdb_name, ajDirGetPath(pdbin));
if(ajStrMatchC(pdb_name, "./") || ajStrMatchC(pdb_name, "."))
ajStrAssignC(&pdb_name, "");
ajStrAppendS(&pdb_name, pdbprefix);
ajStrFmtLower(&pdb->Pdb);
ajStrAppendS(&pdb_name, pdb->Pdb);
ajStrAppendC(&pdb_name, ".");
ajStrAppendS(&pdb_name, ajDirGetExt(pdbin));
/* Check corresponding PDB file exists for reading using ajFileStat. */
if(!(ajFilenameExistsRead(pdb_name)))
{
ajFmtPrintF(errf, "%s%S\n//\n", "PDB file not found: ", pdb_name);
ajWarn("%s%S\n//\n", "PDB file not found: ", pdb_name);
ajStrDel(&ccf_this);
ajPdbDel(&pdb);
nofile++;
continue;
}
if(ajStrGetCharFirst(*mode) != '2')
{
/*
** Create a string containing the STRIDE command line (it needs
** PDB file name & name of temp output file).
** Call STRIDE by using ajSystem.
*/
ajFmtPrintS(&syscmd, "%S %S -f%S >> %s 2>&1",
ajAcdGetpathC("stride"),
pdb_name, randomname, ajFileGetNameC(serrf));
ajFmtPrint("%S %S -f%S >> %s 2>&1\n",
ajAcdGetpathC("stride"),
pdb_name, randomname,ajFileGetNameC(serrf));
system(ajStrGetPtr(syscmd));
/* Open the stride output file */
if (((tempf = ajFileNewInNameS(randomname)) == NULL))
{
ajWarn("%s%S\n//\n",
"no stride output for: ", pdb_name);
ajFmtPrintF(errf, "%s%S\n//\n",
"no stride output for: ", pdb_name);
nostride++;
ajStrDel(&ccf_this);
ajPdbDel(&pdb);
continue;
}
else
ajFmtPrintF(errf, "%s%S\n//\n",
"stride output for: ", pdb_name);
done_stride = ajFalse;
/* Parse STRIDE output from temp output file a line at a time. */
while(ajReadlineTrim(tempf,&line))
{
if(ajStrPrefixC(line,"ASG"))
{
ajFmtScanS(line, "%*S %S %c %S %*d %c %*S %f %f %f %*S",
&res, &pcid, &res_num, &ss, &ph, &ps, &sa);
/*
** Populate pdbplus object with the data from this parsed
** line. This means first identifying the chain, then
** finding the residue.
*/
/* Determine the chain number. ajDmxPdbplusChain does not
recognise '-', so change '-' to '.' */
if (pcid == '-')
pcid = '.';
/* Get chain number from the chain identifier. */
if(!ajPdbChnidToNum(pcid, pdb, &idn))
{
ajWarn("Could not convert chain id %c to chain"
" number in pdb file %S\n//\n",
pcid, pdb_name);
ajFmtPrintF(errf, "Could not convert chain id %c "
"to chain number in pdb file %S\n//\n",
pcid, pdb_name);
continue;
}
/*
** The chain number that will get written starts at 1, but
** we want an index into an array which must start at 0,
** so subtract 1 from the chain number to get the index.
*/
chain_num = idn-1;
/*
** Iiterate through the list of residues in the Pdb object,
** found switches to true when first residue corresponding
** to the line is found.
*/
/* iter = ajListIterNewread(pdb->Chains[chain_num]->Atoms); */
iter = ajListIterNewread(pdb->Chains[chain_num]->Residues);
found = ajFalse;
while((temp_res = (AjPResidue)ajListIterGet(iter)))
{
/* If we have found the residue we want */
if((ajStrMatchS(res_num, temp_res->Pdb) &&
ajStrMatchS(res, temp_res->Id3)))
{
done_stride = ajTrue;
found = ajTrue;
temp_res->eStrideType = ss;
temp_res->Phi = ph;
temp_res->Psi = ps;
temp_res->Area = sa;
}
/* If the matching residue has been processed
move on to next ASG line, next residue. */
else if(found == ajTrue)
break;
else
/* Matching residue not found yet. */
continue;
}
ajListIterDel(&iter);
} /* End of if ASG loop. */
} /* End of while line loop. */
if(done_stride)
ajFmtPrintF(errf, "%s%S\n//\n",
"stride data for: ", pdb_name);
else
{
ajFmtPrintF(errf, "%s%S\n//\n",
"no stride data for: ", pdb_name);
ajWarn("%s%S\n//\n", "no stride data for: ", pdb_name);
nostride++;
}
/* Close STRIDE temp file. & tidy up. */
ajFileClose(&tempf);
/* Remove temporary file (stride output file). */
ajFmtPrintS(&exec, "rm %S", randomname);
ajSysSystem(exec);
/*
** Calculate element serial numbers (eStrideNum)& ammend residue
** objects, count no's of elements and ammend chain object
** (numHelices, num Strands).
*/
pdbplus_sort(pdb, tS);
}
if(ajStrGetCharFirst(*mode) != '1')
{
/*
** Create a string containing the NACCESS command line (it needs
** PDB file name & name of temp output file) & call NACCESS.
** If e.g. /data/structure/pdbfred.ent was parsed and the program
** was run from /stuff, then /stuff/fred.asa and /stuff/fred.rsa
** would be written. These must be deleted once parsed (only
** use the .rsa file here).
*/
ajFmtPrintS(&syscmd, "%S %S >> %s 2>&1",
ajAcdGetpathC("naccess"), pdb_name,
ajFileGetNameC(nerrf));
ajFmtPrint("%S %S >> %s 2>&1\n",
ajAcdGetpathC("naccess"), pdb_name,
ajFileGetNameC(nerrf));
system(ajStrGetPtr(syscmd));
ajStrAssignS(&naccess_str, pdbprefix);
ajStrAppendS(&naccess_str, pdb->Pdb);
ajStrAppendC(&naccess_str, ".rsa");
/* Open the NACCESS output file. */
if (((tempf = ajFileNewInNameS(naccess_str)) == NULL))
{
ajFmtPrintF(errf, "%s%S\n//\n",
"no naccess output for: ", pdb_name);
ajWarn("%s%S\n//\n", "no naccess output for: ", pdb_name);
nonaccess++;
ajStrDel(&ccf_this);
ajPdbDel(&pdb);
continue;
}
else
ajFmtPrintF(errf, "%s%S\n//\n",
"naccess output for: ", pdb_name);
done_naccess = ajFalse;
/* Parse NACCESS output from temp output file a line at a time. */
while(ajReadlineTrim(tempf,&line))
{
if(ajStrPrefixC(line,"RES"))
{
/* Read data from lines. */
if((pcid = line->Ptr[8]) == ' ')
ajFmtScanS(line, "%*S %S %S %f %f %f "
"%f %f %f %f %f %f %f",
&res, &res_num, &f1, &f2, &f3, &f4, &f5,
&f6, &f7, &f8, &f9, &f10);
else
ajFmtScanS(line, "%*S %S %*c %S %f %f "
"%f %f %f %f %f %f %f %f",
&res, &res_num, &f1, &f2, &f3, &f4, &f5,
&f6, &f7, &f8, &f9, &f10);
/* Identify the chain, then finding all the residues
corresponding to the residue. */
/* Get the chain number from the chain identifier. */
if(!ajPdbChnidToNum(pcid, pdb, &idn))
{
ajWarn("Could not convert chain id %c to chain"
" number in pdb file %S\n//\n",
pcid, pdb_name);
ajFmtPrintF(errf, "Could not convert chain id"
" %c to chain number in pdb file %S\n//\n",
pcid, pdb_name);
continue;
}
/*
** Chain number will start at 1, but we want an index
** into an array which must start at 0, so subtract 1
** from the chain number to get the index.
*/
chain_num = idn-1;
/*
** Iiterate through the list of residues in the Pdb object,
** temp_res is an AjPResidue used to point to the current
** residue.
** ajBool found switches to true when first residue
** corresponding to the line is found.
*/
iter = ajListIterNewread(pdb->Chains[chain_num]->Residues);
found = ajFalse;
while((temp_res = (AjPResidue)ajListIterGet(iter)))
{
/* If we have found the residue we want, write the residue
object. */
if((ajStrMatchS(res_num, temp_res->Pdb) &&
ajStrMatchS(res, temp_res->Id3)))
{
found = ajTrue;
done_naccess = ajTrue;
temp_res->all_abs = f1;
temp_res->all_rel = f2;
temp_res->side_abs = f3;
temp_res->side_rel = f4;
temp_res->main_abs = f5;
temp_res->main_rel = f6;
temp_res->npol_abs = f7;
temp_res->npol_rel = f8;
temp_res->pol_abs = f9;
temp_res->pol_rel = f10;
}
/* If the matching residues have all been processed.
move on to next ASG line, next residue. */
else if(found == ajTrue)
break;
else
/* Matching residues not found yet, move on to next
residue. */
continue;
}
ajListIterDel(&iter);
}
}
if(done_naccess)
ajFmtPrintF(errf, "%s%S\n//\n",
"naccess data for: ", pdb_name);
else
{
ajFmtPrintF(errf, "%s%S\n//\n",
"no naccess data for: ", pdb_name);
ajWarn("%s%S\n//\n", "no naccess data for: ", pdb_name);
nonaccess++;
}
/* Remove temporary file (naccess output files). */
ajFileClose(&tempf);
ajFmtPrintS(&exec, "rm %S", naccess_str);
ajSysSystem(exec);
ajStrAssignS(&naccess_str, pdbprefix);
ajStrAppendS(&naccess_str, pdb->Pdb);
ajStrAppendC(&naccess_str, ".asa");
ajFmtPrintS(&exec, "rm %S", naccess_str);
ajSysSystem(exec);
ajStrAssignS(&naccess_str, pdbprefix);
ajStrAppendS(&naccess_str, pdb->Pdb);
ajStrAppendC(&naccess_str, ".log");
ajFmtPrintS(&exec, "rm %S", naccess_str);
ajSysSystem(exec);
}
/* Open CCF (output) file. */
ccf_outf = ajFileNewOutNameDirS(pdb->Pdb, ccfout);
/* Write AjPPdb object to the output file in clean format. */
if(!ajPdbWriteAll(ccf_outf, pdb))
{
ajWarn("%s%S\n//\n","Could not write results file for: ",
pdb->Pdb);
ajFmtPrintF(errf,"%s%S\n//\n","Could not write results file for ",
pdb->Pdb);
}
ajFileClose(&ccf_outf);
ajPdbDel(&pdb);
ajStrDel(&ccf_this);
} /* End of main application loop. */
ajFmtPrint("STRIDE failures: %d\n", nostride);
ajFmtPrint("NACCESS failures: %d\n", nonaccess);
ajFmtPrintF(errf, "\n\nSTRIDE failures: %d\nNACCESS failures: %d\n",
nostride, nonaccess);
ajListFree(&ccfin);
ajDirDel(&pdbin);
ajStrDel(&pdbprefix);
ajStrDel(&pdb_name);
ajDiroutDel(&ccfout);
ajStrDel(&res);
ajStrDel(&res_num);
ajStrDel(&randomname);
ajStrDel(&line);
ajStrDel(&naccess_str);
ajStrDel(&exec);
ajStrDel(&syscmd);
ajFileClose(&errf);
if(ajStrGetCharFirst(*mode) != '2')
ajFileClose(&serrf);
if(ajStrGetCharFirst(*mode) != '1')
ajFileClose(&nerrf);
ajStrDel(&mode[0]);
AJFREE(mode);
ajExit();
return 0;
}
/* @funcstatic seqwords_TermsRead *********************************************
**
** Read the next Terms object from a file in embl-like format. The search
** terms are modified with a leading and trailing space.
**
** @param [r] inf [AjPFile] Input file stream
** @param [w] thys [AjPTerms*] Terms object
**
** @return [AjBool] True on succcess
** @@
*****************************************************************************/
static AjBool seqwords_TermsRead(AjPFile inf,
AjPTerms *thys)
{
AjPStr line =NULL; /* Line of text. */
AjPStr temp =NULL;
AjPList list_terms =NULL; /* List of keywords for a scop node*/
AjBool ok =ajFalse;
AjPStr type = NULL;
/* Memory management */
(*thys)=seqwords_TermsNew();
list_terms = ajListstrNew();
line = ajStrNew();
type = ajStrNew();
/* Read first line. */
ok = ajReadlineTrim(inf,&line);
while(ok && !ajStrPrefixC(line,"//"))
{
if(ajStrPrefixC(line,"XX"))
{
ok = ajReadlineTrim(inf,&line);
continue;
}
else if(ajStrPrefixC(line,"TY"))
{
ajFmtScanS(line, "%*s %S", &type);
if(ajStrMatchC(type, "SCOP"))
(*thys)->Type = ajSCOP;
else if(ajStrMatchC(type, "CATH"))
(*thys)->Type = ajCATH;
}
else if(ajStrPrefixC(line,"CL"))
{
ajStrAssignC(&(*thys)->Class,ajStrGetPtr(line)+3);
ajStrRemoveWhiteExcess(&(*thys)->Class);
}
else if(ajStrPrefixC(line,"AR"))
{
ajStrAssignC(&(*thys)->Architecture,ajStrGetPtr(line)+3);
ajStrRemoveWhiteExcess(&(*thys)->Architecture);
}
else if(ajStrPrefixC(line,"TP"))
{
ajStrAssignC(&(*thys)->Topology,ajStrGetPtr(line)+3);
ajStrRemoveWhiteExcess(&(*thys)->Topology);
}
else if(ajStrPrefixC(line,"FO"))
{
ajStrAssignC(&(*thys)->Fold,ajStrGetPtr(line)+3);
while(ajReadlineTrim(inf,&line))
{
if(ajStrPrefixC(line,"XX"))
break;
ajStrAppendC(&(*thys)->Fold,ajStrGetPtr(line)+3);
}
ajStrRemoveWhiteExcess(&(*thys)->Fold);
}
else if(ajStrPrefixC(line,"SF"))
{
ajStrAssignC(&(*thys)->Superfamily,ajStrGetPtr(line)+3);
while(ajReadlineTrim(inf,&line))
{
if(ajStrPrefixC(line,"XX"))
break;
ajStrAppendC(&(*thys)->Superfamily,ajStrGetPtr(line)+3);
}
ajStrRemoveWhiteExcess(&(*thys)->Superfamily);
}
else if(ajStrPrefixC(line,"FA"))
{
ajStrAssignC(&(*thys)->Family,ajStrGetPtr(line)+3);
while(ajReadlineTrim(inf,&line))
{
if(ajStrPrefixC(line,"XX"))
break;
ajStrAppendC(&(*thys)->Family,ajStrGetPtr(line)+3);
}
ajStrRemoveWhiteExcess(&(*thys)->Family);
}
else if(ajStrPrefixC(line,"TE"))
{
/* Copy and clean up term. */
temp = ajStrNew();
ajStrAssignC(&temp,ajStrGetPtr(line)+3);
ajStrRemoveWhiteExcess(&temp);
/* Append a leading and trailing space to search term*/
ajStrAppendK(&temp, ' ');
ajStrInsertC(&temp, 0, " ");
/* Add the current term to the list. */
ajListstrPush(list_terms,temp);
}
ok = ajReadlineTrim(inf,&line);
}
if(!ok)
{
/* Clean up. */
ajListstrFree(&list_terms);
ajStrDel(&line);
/* Return. */
return ajFalse;
}
/* Convert the AjPList of terms to array of AjPSeq's. */
if(!((*thys)->N=ajListstrToarray((AjPList)list_terms,&(*thys)->Keywords)))
ajWarn("Zero sized list of terms passed into seqwords_TermsRead");
/* Clean up. Free the list (not the nodes!). */
ajListstrFree(&list_terms);
ajStrDel(&line);
ajStrDel(&type);
return ajTrue;
}
AjPPatlistRegex ajPatlistRegexRead (const AjPStr patspec,
const AjPStr patname,
const AjPStr fmt,
ajuint type, AjBool upper, AjBool lower)
{
AjPPatlistRegex patlist = NULL;
AjPStr line = NULL;
AjPStr pat = NULL;
AjPStr name = NULL;
AjPFilebuff infile = NULL;
AjPStr patstr = NULL;
ajuint ifmt;
ajuint npat = 0;
AjPStr namestr = NULL;
ajStrAssignS(&namestr, patname);
ajStrAssignEmptyC(&namestr, "regex");
ajStrAssignS(&patstr, patspec);
patlist = ajPatlistRegexNewType(type);
ifmt = patternRegexFormat(fmt);
if(ajStrGetCharFirst(patspec) == '@')
{
ajStrCutStart(&patstr, 1);
infile = ajFilebuffNewNameS(patstr);
if(!infile)
{
ajErr("Unable to open regular expression file '%S'", patstr);
return NULL;
}
line = ajStrNew();
pat = ajStrNew();
name = ajStrNew();
if(!ifmt)
{
ajBuffreadLineTrim(infile,&line);
if(ajStrPrefixC(line, ">"))
ifmt = 2;
else
ifmt = 1;
ajFilebuffReset(infile);
}
switch(ifmt)
{
case 1:
while (ajBuffreadLineTrim(infile,&line))
{
npat++;
ajStrAppendS (&pat,line);
if(lower)
ajStrFmtLower(&pat);
if(upper)
ajStrFmtUpper(&pat);
ajFmtPrintS(&name, "%S%u", namestr, npat);
ajPatternRegexNewList(patlist,name,pat);
ajStrSetClear(&pat);
}
break;
default:
while (ajBuffreadLineTrim(infile,&line))
{
if (ajStrFindC(line,">")>-1)
{
npat++;
if (ajStrGetLen(name))
{
if(lower)
ajStrFmtLower(&pat);
if(upper)
ajStrFmtUpper(&pat);
ajPatternRegexNewList(patlist,name,pat);
ajStrSetClear(&name);
ajStrSetClear(&pat);
}
ajStrCutStart(&line,1);
ajStrAssignS (&name,line);
if(!ajStrGetLen(name))
ajFmtPrintS(&name, "%S%u", namestr, npat);
}
else
ajStrAppendS (&pat,line);
}
ajStrAssignEmptyS(&name, patname);
ajPatternRegexNewList(patlist,name,pat);
ajStrSetClear(&pat);
break;
}
ajFilebuffDel(&infile);
}
else
{
ajStrAssignS(&pat, patspec);
if(lower)
ajStrFmtLower(&pat);
if(upper)
ajStrFmtUpper(&pat);
ajStrAssignS(&name, namestr);
ajPatternRegexNewList(patlist,name,pat);
}
ajStrDel(&name);
ajStrDel(&namestr);
ajStrDel(&patstr);
ajStrDel(&line);
ajStrDel(&pat);
return patlist;
}
static AjBool dbxflat_ParseGenbank(EmbPBtreeEntry entry, AjPFile inf)
{
AjPStr line = NULL;
ajlong pos = 0L;
AjBool ret = ajTrue;
AjPStr sumline = NULL;
line = ajStrNewC("");
sumline = ajStrNew();
while(!ajStrPrefixC(line,"//") && ret)
{
if(ajStrPrefixC(line,"LOCUS"))
{
entry->fpos = pos;
ajFmtScanS(line,"%*S%S",&entry->id);
}
if(entry->do_sv)
if(ajStrPrefixC(line,"VERSION"))
embBtreeGenBankAC(line,entry->sv);
if(entry->do_accession)
if(ajStrPrefixC(line,"ACCESSION"))
embBtreeGenBankAC(line,entry->ac);
if(entry->do_keyword)
if(ajStrPrefixC(line,"KEYWORDS"))
{
ajStrAssignS(&sumline,line);
ret = ajReadlineTrim(inf,&line);
while(ret && *MAJSTRGETPTR(line)==' ')
{
ajStrAppendS(&sumline,line);
ret = ajReadlineTrim(inf,&line);
}
ajStrRemoveWhiteExcess(&sumline);
embBtreeGenBankKW(sumline,entry->kw,entry->kwlen);
continue;
}
if(entry->do_description)
if(ajStrPrefixC(line,"DEFINITION"))
{
ajStrAssignS(&sumline,line);
ret = ajReadlineTrim(inf,&line);
while(ret && *MAJSTRGETPTR(line)==' ')
{
ajStrAppendS(&sumline,line);
ret = ajReadlineTrim(inf,&line);
}
ajStrRemoveWhiteExcess(&sumline);
embBtreeGenBankDE(sumline,entry->de,entry->delen);
continue;
}
if(entry->do_taxonomy)
if(ajStrPrefixC(line,"SOURCE"))
{
ret = ajReadlineTrim(inf,&line);
ajStrAppendC(&line,";");
while(ret && *MAJSTRGETPTR(line)==' ')
{
ajStrAppendS(&sumline,line);
ret = ajReadlineTrim(inf,&line);
}
ajStrRemoveWhiteExcess(&sumline);
embBtreeGenBankTX(sumline,entry->tx,entry->txlen);
continue;
}
pos = ajFileResetPos(inf);
if(!ajReadlineTrim(inf,&line))
ret = ajFalse;
}
ajStrDel(&line);
ajStrDel(&sumline);
return ret;
}
EmbPPropAmino* embPropEaminoRead(AjPFile mfptr)
{
AjPStr line = NULL;
AjPStr token = NULL;
AjBool firstline;
const char *p;
ajuint i;
ajint n;
EmbPPropAmino *ret;
line = ajStrNew();
token = ajStrNew();
firstline = ajTrue;
AJCNEW0(ret,EMBPROPSIZE);
for(i=0; i < EMBPROPSIZE; ++i)
AJNEW0(ret[i]);
while(ajReadline(mfptr, &line))
{
ajStrRemoveWhiteExcess(&line);
p = ajStrGetPtr(line);
if(*p=='#' || *p=='!' || !*p)
continue;
if(firstline)
{
if(!ajStrPrefixC(line,"aa"))
ajFatal("Incorrect (old?) format amino data file");
firstline = ajFalse;
continue;
}
ajFmtScanS(line,"%S",&token);
ajStrFmtUpper(&token);
if(ajStrGetLen(token) != 1)
ajFatal("Amino file line doesn't begin with a single character");
i = ajBasecodeToInt((ajint) *ajStrGetPtr(token));
if(i == 27)
ajFatal("Amino file line doesn't begin with a single A->Z (%S)",
line);
n = ajFmtScanS(line,"%*s%d%d%d%d%d%d%f%d%d%d",
&ret[i]->tiny,
&ret[i]->sm_all,
&ret[i]->aliphatic,
&ret[i]->aromatic,
&ret[i]->nonpolar,
&ret[i]->polar,
&ret[i]->charge,
&ret[i]->pve,
&ret[i]->nve,
&ret[i]->extcoeff);
if(n!= 10)
ajFatal("Only %d columns in amino file - expected %d",n+1,11);
}
ajStrDel(&line);
ajStrDel(&token);
return ret;
}
EmbPPropMolwt* embPropEmolwtRead(AjPFile mfptr)
{
AjPStr line = NULL;
AjPStr token = NULL;
AjBool firstline;
const char *p;
ajuint i;
ajint n;
EmbPPropMolwt *ret;
line = ajStrNew();
token = ajStrNew();
firstline = ajTrue;
AJCNEW0(ret,EMBPROPSIZE+2);
for(i=0; i < EMBPROPSIZE+2; ++i)
AJNEW0(ret[i]);
while(ajReadline(mfptr, &line))
{
ajStrRemoveWhiteExcess(&line);
p = ajStrGetPtr(line);
if(*p=='#' || *p=='!' || !*p)
continue;
if(firstline)
{
if(!ajStrPrefixC(line,"Mol"))
ajFatal("Incorrect format molwt file: '%S'", line);
firstline = ajFalse;
continue;
}
ajFmtScanS(line,"%S",&token);
ajStrFmtUpper(&token);
if(ajStrGetLen(token) != 1)
{
if(ajStrPrefixC(token,"HYDROGEN"))
{
if(ajFmtScanS(line,"%*s%lf%lf",
&ret[EMBPROPHINDEX]->average,
&ret[EMBPROPHINDEX]->mono) != 2)
ajFatal("Bad format hydrogen data line");
}
else if(ajStrPrefixC(token,"OXYGEN"))
{
if(ajFmtScanS(line,"%*s%lf%lf",
&ret[EMBPROPOINDEX]->average,
&ret[EMBPROPOINDEX]->mono) != 2)
ajFatal("Bad format oxygen data line");
}
else if(ajStrPrefixC(token,"WATER"))
{
if(ajFmtScanS(line,"%*s%lf%lf",
&ret[EMBPROPWINDEX]->average,
&ret[EMBPROPWINDEX]->mono) != 2)
ajFatal("Bad format water data line");
}
else
ajFatal("Unknown molwt token %S",token);
continue;
}
i = ajBasecodeToInt((ajint) *ajStrGetPtr(token));
if(i == 27)
ajFatal("Molwt file line doesn't begin with a single A->Z (%S)",
line);
n = ajFmtScanS(line,"%*s%lf%lf",
&ret[i]->average,
&ret[i]->mono);
if(n != 2)
ajFatal("Only %d columns in amino file - expected %d",n,3);
}
ajStrDel(&line);
ajStrDel(&token);
return ret;
}
/* @funcstatic acdrelations_writerelations ************************************
**
** Writes relations: attribute for an ACD data definition
** The relations: values given in knowntypes.standard have highest precedence,
** then the values given in edamtoacd.dat
** Attribute values for a given datatype in edamtoacd.dat are in order of
** increasing precedence, i.e. the last line is highest and the relations:
** value will be used if all conditions are met.
**
** @param [r] outf [AjPFile] ACD output file
** @param [r] acdtype [AjPStr ] ACD datatype, e.g. "align"
** @param [r] strarr [AjPStr*] All ACD attribute lines (whitespace removed)
** for the the current ACD data item (of type
** acdtype). One line per array element.
** @param [r] n [ajint] Size of strarr
** @param [r] P [PEdam] edam object to write
** @param [r] T [PKtype] ktype object to read
** @return [void]
** @@
******************************************************************************/
static void acdrelations_writerelations
(AjPFile outf,
AjPStr acdtype,
AjPStr *strarr,
ajint n,
PEdam P,
PKtype T)
{
ajint i = 0;
ajint j = 0;
ajint k = 0;
ajint nmatch = 0;
AjPStr relations = NULL;
AjPStr ktype = NULL; /* Value of knowntype: attribute */
AjBool done = ajFalse;
AjBool donetype = ajFalse;
AjPStr tmpstr = NULL;
if(!outf || !acdtype || !strarr || !n || !P)
ajFatal("NULL args passed to acdrelations_writerelations");
/* Memory allocation */
relations = ajStrNew();
ktype = ajStrNew();
tmpstr = ajStrNew();
/* Loop through all lines in edamtoacd.dat */
for(i=0; i<P->n ;i++)
{
/* Found matching datatype */
if(ajStrMatchS(acdtype, P->dat[i]->acdtype))
{
/* Copy first relations: string defined for this datatype (default) */
ajStrAssignS(&relations, P->dat[i]->edam);
done = ajTrue;
i++;
/* Check next line in edamtoacd.dat */
for( ; i<P->n; i++)
{
/* Datatype still matches */
if(ajStrMatchS(acdtype, P->dat[i]->acdtype))
{
/* Loop through all required attributes for this datatype */
for(nmatch=0, j=0; j<P->dat[i]->n; j++)
{
/* Loop through all attribute lines for the data defininition */
for(k=0; k<n; k++)
if(ajStrMatchS(P->dat[i]->acdattr[j], strarr[k]))
{
nmatch++;
/* ajFmtPrint("Found match %d: %S:%S\n", nmatch, P->dat[i]->acdattr[j], strarr[k]); */
break;
}
}
/* All attribute values match */
if(nmatch == P->dat[i]->n)
ajStrAssignS(&relations, P->dat[i]->edam);
/* Should never happen */
else if (nmatch > P->dat[i]->n)
ajFatal("Terminal weirdness in acdrelations_writerelations");
}
else
break;
}
break;
}
}
/* Check for match of knowntype: attribute against knowntypes.standard.
These have higher precedence than the rules defined in edamtoacd.dat */
for(donetype=ajFalse, i=0; i<n; i++)
{
if(ajStrPrefixC(strarr[i], "knowntype:"))
{
for(j=0;j<T->n; j++)
{
/* No check is made on the "Type" column in knowntypes.standard
as these are not proper ACD datatype names
To check these add
if(ajStrMatchS(acdtype, T->dat[j]->acdtype)) */
ajFmtPrintS(&tmpstr, "knowntype:\"%S\"", T->dat[j]->ktype);
if(ajStrMatchS(tmpstr, strarr[i]))
{
ajStrAssignS(&relations, T->dat[j]->edam);
donetype=ajTrue;
break;
}
}
if(donetype)
break;
}
else
continue;
}
if(!done)
ajFatal("No matching datatype (%S) in acdrelations_writerelations", acdtype);
/* Write relations: attribute line to file */
ajFmtPrintF(outf, " relations:%S\n", relations);
/* Free memory */
ajStrDel(&relations);
ajStrDel(&ktype);
ajStrDel(&tmpstr);
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;
}
static void jaspextract_copyfiles(AjPStr directory)
{
AjPStr matrixfile = NULL;
AjPList flist = NULL;
AjPStr wild = NULL;
AjPStr entry = NULL;
AjPStr bname = NULL;
AjPStr line = NULL;
AjPStr dest = NULL;
const AjPStr datadir = NULL;
ajuint preflen = 0;
ajuint i = 0;
const char *p = NULL;
AjPFile inf = NULL;
AjPFile outf = NULL;
matrixfile = ajStrNew();
flist = ajListNew();
wild = ajStrNewC("*.pfm");
bname = ajStrNew();
line = ajStrNew();
dest = ajStrNew();
datadir = ajDatafileValuePath();
if(!datadir)
ajFatal("jaspextract: Cannot determine the EMBOSS data directory");
ajFmtPrintS(&matrixfile,"%S%s",directory,MATRIXFILE);
if(!ajFilenameExistsRead(matrixfile))
ajFatal("jaspextract: Directory (%S) doesn't appear to be a JASPAR "
"one\nNo matrix_list.txt file found",directory);
ajFilelistAddPathWild(flist, directory, wild);
while(ajListPop(flist,(void **)&entry))
{
ajStrAssignS(&bname,entry);
ajFilenameTrimPath(&bname);
i = 0;
while(Jprefix[i].Prefix)
{
if(!ajStrPrefixC(bname,Jprefix[i].Prefix))
{
++i;
continue;
}
preflen = strlen(Jprefix[i].Prefix);
p = ajStrGetPtr(bname);
if(p[preflen]>='0' && p[preflen]<='9')
break;
++i;
}
if(!Jprefix[i].Prefix)
{
ajStrDel(&entry);
continue;
}
ajFmtPrintS(&dest,"%S%s%c%S",datadir,Jprefix[i].Directory,SLASH_CHAR,
bname);
outf = ajFileNewOutNameS(dest);
if(!outf)
ajFatal("Cannot open output file %S",dest);
/* Avoid UNIX copy for portability */
inf = ajFileNewInNameS(entry);
if(!inf)
ajFatal("Cannot open input file: %S",entry);
while(ajReadlineTrim(inf,&line))
ajFmtPrintF(outf,"%S\n",line);
ajFileClose(&inf);
ajFileClose(&outf);
ajStrDel(&entry);
}
ajListFree(&flist);
ajStrDel(&wild);
ajStrDel(&dest);
ajStrDel(&line);
ajStrDel(&bname);
ajStrDel(&matrixfile);
return;
}
AjPPatlistSeq ajPatlistSeqRead (const AjPStr patspec,
const AjPStr patname,
const AjPStr fmt,
AjBool protein, ajuint mismatches)
{
AjPPatlistSeq patlist = NULL;
AjPStr line = NULL;
AjPStr name = NULL;
AjPFilebuff infile = NULL;
AjPRegexp mismreg = NULL;
AjPStr patstr = NULL;
AjPStr pat = NULL;
ajuint mismatch = 0;
ajint ifmt = 0;
ajuint npat = 0;
AjPStr namestr = NULL;
ajStrAssignS(&namestr, patname);
ajStrAssignEmptyC(&namestr, "pattern");
ajStrAssignS(&patstr, patspec);
patlist = ajPatlistSeqNewType(protein);
ifmt = patternSeqFormat(fmt);
ajDebug("ajPatlistSeqRead patspec: '%S' patname: '%S' "
"protein: %B mismatches: %d\n",
patspec, patname, protein, mismatches);
if(ajStrGetCharFirst(patstr) == '@')
{
ajStrCutStart(&patstr, 1);
infile = ajFilebuffNewNameS(patstr);
if(!infile)
{
ajErr("Unable to open pattern file '%S'", patstr);
return NULL;
}
line = ajStrNew();
name = ajStrNew();
if(!ifmt)
{
ajBuffreadLineTrim(infile,&line);
if(ajStrPrefixC(line, ">"))
ifmt = 2;
else
ifmt = 1;
ajFilebuffReset(infile);
}
switch(ifmt)
{
case 1:
while (ajBuffreadLineTrim(infile,&line))
{
npat++;
ajStrAppendS (&pat,line);
ajFmtPrintS(&name, "%S%u", namestr, npat);
ajPatternSeqNewList(patlist,name,pat,mismatches);
ajStrSetClear(&pat);
}
break;
default:
mismreg = ajRegCompC("<mismatch=(\\d+)>");
while (ajBuffreadLineTrim(infile,&line))
{
if (ajStrGetCharFirst(line) == '>')
{
if (ajStrGetLen(name))
{
ajPatternSeqNewList(patlist,name,pat,
mismatch);
ajStrSetClear(&name);
ajStrSetClear(&pat);
mismatch=mismatches;
}
ajStrCutStart(&line,1);
if (ajRegExec(mismreg,line))
{
ajRegSubI(mismreg,1,&name);
ajStrToUint(name,&mismatch);
ajStrTruncateLen(&line,ajRegOffset(mismreg));
ajStrTrimWhiteEnd(&line);
}
ajStrAssignS (&name,line);
ajStrAssignEmptyS(&name, patname);
}
else
ajStrAppendS (&pat,line);
}
ajStrAssignEmptyS(&name, patname);
ajPatternSeqNewList(patlist,name,pat,mismatch);
ajRegFree(&mismreg);
break;
}
ajFilebuffDel(&infile);
}
else
{
ajStrAssignS(&name, namestr);
ajPatternSeqNewList(patlist,name,patstr,mismatches);
}
ajStrDel(&name);
ajStrDel(&line);
ajStrDel(&pat);
ajStrDel(&namestr);
ajStrDel(&patstr);
return patlist;
}
static AjBool dbxflat_ParseGenbank(EmbPBtreeEntry entry, AjPFile inf)
{
AjPStr line = NULL;
ajlong pos = 0L;
AjBool ret = ajTrue;
AjPStr sumline = NULL;
line = ajStrNewC("");
sumline = ajStrNew();
while(!ajStrPrefixC(line,"//") && ret)
{
if(ajStrPrefixC(line,"LOCUS"))
{
entry->fpos = pos;
ajFmtScanS(line,"%*S%S",&entry->id);
}
if(svfield)
if(ajStrPrefixC(line,"VERSION"))
embBtreeGenBankAC(line,svfield->data);
if(accfield)
if(ajStrPrefixC(line,"ACCESSION"))
embBtreeGenBankAC(line,accfield->data);
if(keyfield)
if(ajStrPrefixC(line,"KEYWORDS"))
{
ajStrAssignS(&sumline,line);
ret = ajReadlineTrim(inf,&line);
while(ret && *MAJSTRGETPTR(line)==' ')
{
ajStrAppendS(&sumline,line);
ret = ajReadlineTrim(inf,&line);
}
ajStrRemoveWhiteExcess(&sumline);
embBtreeGenBankKW(sumline,keyfield->data,keyfield->len);
continue;
}
if(desfield)
if(ajStrPrefixC(line,"DEFINITION"))
{
ajStrAssignS(&sumline,line);
ret = ajReadlineTrim(inf,&line);
while(ret && *MAJSTRGETPTR(line)==' ')
{
ajStrAppendS(&sumline,line);
ret = ajReadlineTrim(inf,&line);
}
ajStrRemoveWhiteExcess(&sumline);
embBtreeGenBankDE(sumline,desfield->data,desfield->len);
continue;
}
if(orgfield)
if(ajStrPrefixC(line,"SOURCE"))
{
ret = ajReadlineTrim(inf,&line);
ajStrAppendC(&line,";");
while(ret && *MAJSTRGETPTR(line)==' ')
{
ajStrAppendS(&sumline,line);
ret = ajReadlineTrim(inf,&line);
}
ajStrRemoveWhiteExcess(&sumline);
embBtreeGenBankTX(sumline,orgfield->data,orgfield->len);
continue;
}
pos = ajFileResetPos(inf);
if(!ajReadlineTrim(inf,&line))
ret = ajFalse;
}
ajStrDel(&line);
ajStrDel(&sumline);
return ret;
}
int main(int argc, char **argv)
{
AjPFile inf = NULL;
AjPFile inf2 = NULL;
AjPFeattable tab = NULL;
AjPReport report = NULL;
AjPSeq sequence = NULL;
AjPStr redatanew = NULL;
AjPStr str = NULL;
AjPStr regexp = NULL;
AjPStr temp = NULL;
AjPStr text = NULL;
AjPStr docdata = NULL;
AjPStr data = NULL;
AjPStr accession = NULL;
AjPStr name = NULL;
EmbPPatMatch match = NULL;
AjPStr savereg = NULL;
AjPStr fthit = NULL;
AjBool full;
AjBool prune;
ajint i;
ajint number;
ajint start;
ajint end;
ajint length;
ajint zstart;
ajint zend;
const char *p;
ajint seqlength;
AjPStr tmpstr = NULL;
AjPStr tailstr = NULL;
AjPFeature gf;
embInit("patmatmotifs", argc, argv);
ajStrAssignC(&fthit, "SO:0001067");
savereg = ajStrNew();
str = ajStrNew();
regexp = ajStrNew();
temp = ajStrNew();
data = ajStrNew();
accession = ajStrNew();
text = ajStrNew();
name = ajStrNew();
sequence = ajAcdGetSeq("sequence");
report = ajAcdGetReport("outfile");
full = ajAcdGetBoolean("full");
prune = ajAcdGetBoolean("prune");
ajSeqFmtUpper(sequence); /* prosite regexs are all upper case */
tab = ajFeattableNewSeq(sequence);
ajStrAssignC(&tailstr, "");
seqlength = ajStrGetLen(str);
str = ajSeqGetSeqCopyS(sequence);
redatanew = ajStrNewC("PROSITE/prosite.lines");
docdata = ajStrNewC("PROSITE/");
inf = ajDatafileNewInNameS(redatanew);
if(!inf)
ajFatal("Either EMBOSS_DATA undefined or PROSEXTRACT needs running");
ajFmtPrintAppS(&tmpstr, "Full: %B\n", full);
ajFmtPrintAppS(&tmpstr, "Prune: %B\n", prune);
ajFmtPrintAppS(&tmpstr, "Data_file: %F\n", inf);
ajReportSetHeaderS(report, tmpstr);
while(ajReadlineTrim(inf, ®exp))
{
p=ajStrGetPtr(regexp);
if(*p && *p!=' ' && *p!='^')
{
p=ajSysFuncStrtok(p," ");
ajStrAssignC(&name,p);
if(prune)
if(ajStrMatchCaseC(name,"myristyl") ||
ajStrMatchCaseC(name,"asn_glycosylation") ||
ajStrMatchCaseC(name,"camp_phospho_site") ||
ajStrMatchCaseC(name,"pkc_phospho_site") ||
ajStrMatchCaseC(name,"ck2_phospho_site") ||
ajStrMatchCaseC(name,"tyr_phospho_site"))
{
for(i=0;i<4;++i)
ajReadlineTrim(inf, ®exp);
continue;
}
p=ajSysFuncStrtok(NULL," ");
ajStrAssignC(&accession,p);
}
if(ajStrPrefixC(regexp, "^"))
{
p = ajStrGetPtr(regexp);
ajStrAssignC(&temp,p+1);
ajStrAssignC(&savereg,p+1);
match = embPatMatchFind(temp, str, ajFalse, ajFalse);
number = embPatMatchGetNumber(match);
for(i=0; i<number; i++)
{
seqlength = ajStrGetLen(str);
start = 1+embPatMatchGetStart(match, i);
end = 1+embPatMatchGetEnd(match, i);
length = embPatMatchGetLen(match, i);
gf = ajFeatNew(tab, NULL, fthit, start, end,
(float) length, ' ', 0);
ajFmtPrintS(&tmpstr, "*motif %S", name);
ajFeatTagAddSS(gf, NULL, tmpstr);
if(start-5<0)
zstart = 0;
else
zstart = start-5;
if(end+5> seqlength)
zend = end;
else
zend = end+5;
ajStrAssignSubS(&temp, str, zstart, zend);
}
if(full && number)
{
ajStrAssignC(&redatanew,ajStrGetPtr(docdata));
ajStrAppendC(&redatanew,ajStrGetPtr(accession));
inf2 = ajDatafileNewInNameS(redatanew);
if(!inf2)
continue;
/*
** Insert Prosite documentation from files made by
** prosextract.c
*/
ajFmtPrintAppS(&tailstr, "Motif: %S\n", name);
ajFmtPrintAppS(&tailstr, "Count: %d\n\n", number);
while(ajReadlineTrim(inf2, &text))
ajFmtPrintAppS(&tailstr, "%S\n", text);
ajFmtPrintAppS(&tailstr, "\n***************\n\n");
ajFileClose(&inf2);
}
embPatMatchDel(&match);
}
}
ajReportSetTailS(report,tailstr);
ajReportWrite(report, tab, sequence);
ajReportDel(&report);
ajFeattableDel(&tab);
ajStrDel(&temp);
ajStrDel(®exp);
ajStrDel(&savereg);
ajStrDel(&str);
ajStrDel(&data);
ajStrDel(&docdata);
ajStrDel(&text);
ajStrDel(&redatanew);
ajStrDel(&accession);
ajSeqDel(&sequence);
ajStrDel(&tailstr);
ajStrDel(&fthit);
ajStrDel(&name);
ajStrDel(&tmpstr);
ajFeattableDel(&tab);
ajFileClose(&inf);
embExit();
return 0;
}
static AjBool dbxflat_ParseFastq(EmbPBtreeEntry entry, AjPFile inf)
{
AjPStr line = NULL;
ajlong pos = 0L;
ajuint seqlen = 0;
ajuint qlen = 0;
AjPStr tmpfd = NULL;
AjPStr str = NULL;
AjPStr de = NULL;
AjBool ok;
if(!dbxflat_wrdexp)
dbxflat_wrdexp = ajRegCompC("([A-Za-z0-9.:=]+)");
line = ajStrNewC("");
pos = ajFileResetPos(inf);
if(!ajReadlineTrim(inf,&line))
{
ajStrDel(&line);
return ajFalse;
}
/* first line of entry */
if(!ajStrPrefixC(line,"@"))
return ajFalse;
entry->fpos = pos;
ajStrCutStart(&line, 1);
ajStrExtractFirst(line, &de, &entry->id);
if(desfield && ajStrGetLen(de))
{
while(ajRegExec(dbxflat_wrdexp,de))
{
ajRegSubI(dbxflat_wrdexp, 1, &tmpfd);
str = ajStrNew();
ajStrAssignS(&str,tmpfd);
ajListPush(desfield->data,(void *)str);
ajRegPost(dbxflat_wrdexp, &de);
}
}
/* now read sequence */
ok = ajReadlineTrim(inf,&line);
while(ok && !ajStrPrefixC(line, "+"))
{
ajStrRemoveWhite(&line);
seqlen += MAJSTRGETLEN(line);
ok = ajReadlineTrim(inf,&line);
}
if(!ok)
return ajFalse;
ok = ajReadlineTrim(inf,&line);
while(ok)
{
qlen += MAJSTRGETLEN(line);
if(qlen < seqlen)
ok = ajReadlineTrim(inf,&line);
else
ok = ajFalse;
}
ajStrDel(&de);
ajStrDel(&tmpfd);
ajStrDel(&line);
return ajTrue;
}
/* @funcstatic seqwords_keysearch ********************************************
**
** Search swissprot with terms structure and writes a hitlist structure
**
** @param [r] inf [AjPFile] File pointer to swissprot database
** @param [r] terms [AjPTerms] Terms object pointer
** @param [w] hits [EmbPHitlist*] Hitlist object pointer
**
** @return [AjBool] True on success
** @@
******************************************************************************/
static AjBool seqwords_keysearch(AjPFile inf,
AjPTerms terms,
EmbPHitlist *hits)
{
AjPStr line =NULL; /* Line of text. */
AjPStr id =NULL; /* Line of text. */
AjPStr temp =NULL;
ajint s =0; /* Temp. start of hit value. */
ajint e =0; /* Temp. end of hit value. */
AjPInt start =NULL; /* Array of start of hit(s). */
AjPInt end =NULL; /* Array of end of hit(s). */
ajint nhits =0; /* Number of hits. */
ajint x =0;
AjBool foundkw =ajFalse;
AjBool foundft =ajFalse;
/* Check for valid args. */
if(!inf)
return ajFalse;
/* Allocate strings and arrays. */
line = ajStrNew();
id = ajStrNew();
start = ajIntNew();
end = ajIntNew();
/* Start of main loop. */
while((ajReadlineTrim(inf,&line)))
{
/* Parse the AC line. */
if(ajStrPrefixC(line,"AC"))
{
/* Copy accesion number and remove the ';' from the end. */
ajFmtScanS(line, "%*s %S", &id);
ajStrExchangeCC(&id, ";", "\0");
/* Reset flags & no. hits. */
foundkw=ajFalse;
foundft=ajFalse;
nhits=0;
}
/* Search the description and keyword lines with search terms. */
else if((ajStrPrefixC(line,"DE") || (ajStrPrefixC(line,"KW"))))
{
/*
** Search terms have a leading and trailing space to prevent
** them being found as substrings within other words. To
** catch cases where a DE or KW line begins with a search
** term, we must add a leading and trailing space to line.
** We must first remove punctation from the line to be parsed.
*/
ajStrExchangeSetCC(&line, ".,;:", " ");
ajStrAppendK(&line, ' ');
ajStrInsertC(&line, 0, " ");
for (x = 0; x < terms->N; x++)
/* Search term is found. */
if((ajStrFindCaseS(line, terms->Keywords[x])!=-1))
{
foundkw=ajTrue;
break;
}
}
/* Search the feature table line with search terms. */
else if((ajStrPrefixC(line,"FT DOMAIN")))
{
/*
** Search terms have a leading and trailing space to prevent
** them being found as substrings within other words. To
** catch cases where a FT line ends with a search
** term, we must add a trailing space to line
** We must first remove punctation from the line to be parsed.
*/
ajStrExchangeSetCC(&line, ".,;:", " ");
ajStrAppendK(&line, ' ');
for (x = 0; x < terms->N; x++)
if((ajStrFindCaseS(line, terms->Keywords[x])!=-1))
{
/* Search term is found. */
foundft = ajTrue;
nhits++;
/* Assign start and end of hit. */
ajFmtScanS(line, "%*s %*s %d %d", &s, &e);
ajIntPut(&start, nhits-1, s);
ajIntPut(&end, nhits-1, e);
break;
}
}
/* Parse the sequence. */
else if((ajStrPrefixC(line,"SQ") && ((foundkw == ajTrue) ||
(foundft == ajTrue))))
{
/* Allocate memory for temp. sequence. */
temp = ajStrNew();
/* Read the sequence into hitlist structure. */
while((ajReadlineTrim(inf,&line)) && !ajStrPrefixC(line,"//"))
/* Read sequence line into temp. */
ajStrAppendC(&temp,ajStrGetPtr(line)+3);
/* Clean up temp. sequence. */
ajStrRemoveWhite(&temp);
/*Priority is given to domain (rather than full length) sequence.*/
if(foundft)
{
for(x=0;x<nhits;x++)
{
/* Increment counter of hits for subsequent hits*/
(*hits)->N++;
/* Reallocate memory for array of hits in hitlist
structure. */
AJCRESIZE((*hits)->hits, (*hits)->N);
(*hits)->hits[(*hits)->N-1]=embHitNew();
ajStrAssignC(&(*hits)->hits[(*hits)->N-1]->Model,
"KEYWORD");
/* Assign start and end of hit. */
(*hits)->hits[(*hits)->N-1]->Start = ajIntGet(start, x);
(*hits)->hits[(*hits)->N-1]->End = ajIntGet(end, x);
/* Extract sequence within specified range */
ajStrAssignSubS(&(*hits)->hits[(*hits)->N - 1]->Seq, temp,
(*hits)->hits[(*hits)->N - 1]->Start - 1,
(*hits)->hits[(*hits)->N - 1]->End - 1);
/* Put id into structure */
ajStrAssignRef(&(*hits)->hits[(*hits)->N - 1]->Acc, id);
}
}
else
{
/* Increment counter of hits */
(*hits)->N++;
/* Reallocate memory for array of hits in hitlist structure */
AJCRESIZE((*hits)->hits, (*hits)->N);
(*hits)->hits[(*hits)->N-1]=embHitNew();
ajStrAssignC(&(*hits)->hits[(*hits)->N-1]->Model, "KEYWORD");
/* Extract whole sequence */
ajStrAssignRef(&(*hits)->hits[(*hits)->N - 1]->Seq, temp);
(*hits)->hits[(*hits)->N - 1]->Start = 1;
(*hits)->hits[(*hits)->N - 1]->End =
ajStrGetLen((*hits)->hits[(*hits)->N - 1]->Seq);
/* Put id into structure */
ajStrAssignRef(&(*hits)->hits[(*hits)->N - 1]->Acc, id);
}
/* Free temp. sequence */
ajStrDel(&temp);
}
}
/* Clean up */
ajStrDel(&line);
ajStrDel(&id);
ajIntDel(&start);
ajIntDel(&end);
return ajTrue;
}
int main(int argc, char **argv)
{
AjPFile infdat = NULL;
AjPFile infdoc = NULL;
AjPFile outf = NULL;
AjPFile outs = NULL;
AjBool haspattern;
const char *p;
AjPStr line = NULL;
AjPStr text = NULL;
AjPStr dirname = NULL;
AjPStr filename = NULL;
AjPStr id = NULL;
AjPStr ac = NULL;
AjPStr de = NULL;
AjPStr pa = NULL;
AjPStr ps = NULL;
AjPStr fn = NULL;
AjPStr re = NULL;
AjPStr fname = NULL;
AjBool flag;
AjBool isopen;
AjBool goback;
ajlong storepos = 0L;
embInit("prosextract", argc, argv);
dirname = ajAcdGetDirectoryName("prositedir");
line = ajStrNew();
text = ajStrNew();
id = ajStrNew();
ac = ajStrNew();
de = ajStrNew();
pa = ajStrNew();
ps = ajStrNew();
fn=ajStrNew();
ajStrAssignS(&fn,dirname);
ajStrAppendC(&fn,"prosite.dat");
if(!(infdat=ajFileNewInNameS(fn)))
ajFatal("Cannot open file %S",fn);
ajStrDel(&fn);
fn=ajStrNewC("PROSITE/prosite.lines");
outf = ajDatafileNewOutNameS(fn);
ajStrDel(&fn);
haspattern = ajFalse;
while(ajReadlineTrim(infdat, &line) )
{
if(ajStrPrefixC(line, "ID"))
{
if(ajStrSuffixC(line,"PATTERN."))
{
haspattern = ajTrue;
/*save id*/
p = ajStrGetPtr(line);
p = ajSysFuncStrtok(p," \t;");
p = ajSysFuncStrtok(NULL," \t;");
ajStrAssignC(&id,p);
ajFmtPrintF(outf, "%S ", id);
continue;
}
else
{
haspattern = ajFalse;
continue;
}
}
if(!haspattern)
continue;
if(ajStrPrefixC(line, "AC") )
{
p = ajStrGetPtr(line);
p = ajSysFuncStrtok(p, " \t;");
p = ajSysFuncStrtok(NULL, " \t;");
ajStrAssignC(&ac,p);
ajFmtPrintF(outf, "%S\n ", ac);
continue;
}
if(ajStrPrefixC(line, "DE") )
{
p = ajStrGetPtr(line);
p = ajSysFuncStrtok(p, " \t.");
p = ajSysFuncStrtok(NULL, " \t.");
ajStrAssignC(&de,p);
ajFmtPrintF(outf, "%S\n ", de);
continue;
}
if(ajStrPrefixC(line, "PA"))
{
ajStrAssignC(&pa,"");
while(ajStrPrefixC(line,"PA"))
{
p = ajStrGetPtr(line);
p = ajSysFuncStrtok(p, " \t.");
p = ajSysFuncStrtok(NULL, " \t.");
ajStrAppendC(&pa,p);
ajReadlineTrim(infdat, &line);
}
ajFmtPrintF(outf, "%S\n", pa);
re = embPatPrositeToRegExp(pa);
ajFmtPrintF(outf, "^%S\n\n", re);
ajStrDel(&re);
continue;
}
}
/* Finished processing prosite.dat so look at prosite.doc */
fn = ajStrNew();
ajStrAssignS(&fn,dirname);
ajStrAppendC(&fn,"prosite.doc");
if(!(infdoc=ajFileNewInNameS(fn)))
ajFatal("Cannot open file %S",fn);
ajStrDel(&fn);
fname = ajStrNewC("PROSITE/");
flag = ajFalse;
isopen = ajFalse;
goback = ajFalse;
while(ajReadlineTrim(infdoc, &text))
{
if(ajStrPrefixC(text, "{PS") && isopen && !goback)
goback = ajTrue;
if(ajStrPrefixC(text, "{PS") && !isopen)
{
storepos = ajFileResetPos(infdoc);
/* save out the documentation text to acc numbered outfiles . */
p = ajStrGetPtr(text)+1;
p = ajSysFuncStrtok(p, ";");
ajStrAssignS(&filename, fname);
ajStrAppendC(&filename, p);
outs = ajDatafileNewOutNameS(filename);
flag = ajTrue;
isopen = ajTrue;
continue;
}
if(ajStrPrefixC(text, "{BEGIN}") && flag)
{
while(ajReadlineTrim(infdoc, &text))
{
if(ajStrPrefixC(text,"{END}"))
break;
ajFmtPrintF(outs, "%S\n", text);
}
ajFileClose(&outs);
isopen = ajFalse;
if(goback)
{
goback = ajFalse;
ajFileSeek(infdoc,storepos,0);
}
}
}
ajStrDel(&line);
ajStrDel(&text);
ajStrDel(&dirname);
ajStrDel(&filename);
ajStrDel(&id);
ajStrDel(&ac);
ajStrDel(&de);
ajStrDel(&pa);
ajStrDel(&re);
ajStrDel(&ps);
ajStrDel(&fname);
ajFileClose(&infdat);
ajFileClose(&infdoc);
ajFileClose(&outf);
embExit();
return 0;
}
