int main(int argc, char **argv)
{
/* ACD data item variables */
AjPFile database = NULL;
/* Housekeeping variables */
AjPStr cmd = NULL;
AjPStr tmp = NULL;
/* ACD file processing */
embInitPV("ehmmindex",argc,argv,"HMMERNEW",VERSION);
database = ajAcdGetInfile("database");
/* MAIN APPLICATION CODE */
/* 1. Housekeeping */
cmd = ajStrNew();
tmp = ajStrNew();
/* 2. Build hmmindex command line */
/* Command line is built in this order:
i. Application name.
ii. HMMER 'options' (in order they appear in ACD file)
iii.HMMER 'options' (that don't appear in ACD file)
iv. HMMER & new parameters.
*/
ajFmtPrintS(&cmd, "%S ", ajAcdGetpathC("hmmindex"));
ajStrAppendC(&cmd, ajFileGetNameC(database));
/* 3. Close ACD files. */
ajFileClose(&database);
/* 4. Call hmmindex */
ajFmtPrint("\n%S\n\n", cmd);
system(ajStrGetPtr(cmd));
/* 5. Exit cleanly */
ajStrDel(&cmd);
ajStrDel(&tmp);
embExit();
return 0;
}
static void assemoutBamIndex(AjPOutfile outf)
{
char *fname = ajCharNewC(ajFileGetNameC(ajOutfileGetFile(outf)));
AjPSeqBamBgzf gzfile = outf->OutData;
ajSeqBamBgzfClose(gzfile);
ajFileClose(&outf->File);
ajBamIndexBuild(fname);
AJFREE(fname);
return;
}
int main(int argc, char **argv)
{
/* ACD data item variables */
AjPSeqset alignfile = NULL;
AjPFile prior = NULL;
AjPFile null = NULL;
AjPFile pam = NULL;
float pamwgt = 0.0;
AjPStr nhmm = NULL;
AjPStr strategy = NULL;
ajint pbswitch = 0;
float archpri = 0.0;
AjBool binary = ajFalse;
AjBool fast = ajFalse;
float gapmax = 0.0;
AjBool hand = ajFalse;
float idlevel = 0.0;
AjBool noeff = ajFalse;
float swentry = 0.0;
float swexit = 0.0;
AjBool verbosity = ajFalse;
AjPStr weighting = NULL;
AjPFile hmmfile = NULL;
AjPFile o = NULL;
AjPFile cfile = NULL;
/* Housekeeping variables */
AjPStr cmd = NULL;
AjPStr rnd1 = NULL;
AjPStr rnd2 = NULL;
AjPStr tmp = NULL;
AjPStr fmt = NULL;
char option;
AjBool fmtok = ajFalse;
AjPStr hmmfilename = NULL;
/* ACD file processing */
embInitPV("ehmmbuild",argc,argv,"HMMERNEW",VERSION);
alignfile = ajAcdGetSeqset("alignfile");
prior = ajAcdGetInfile("prior");
null = ajAcdGetInfile("null");
pam = ajAcdGetInfile("pam");
pamwgt = ajAcdGetFloat("pamwgt");
nhmm = ajAcdGetString("nhmm");
strategy = ajAcdGetListSingle("strategy");
pbswitch = ajAcdGetInt("pbswitch");
archpri = ajAcdGetFloat("archpri");
binary = ajAcdGetBoolean("binary");
fast = ajAcdGetBoolean("fast");
gapmax = ajAcdGetFloat("gapmax");
hand = ajAcdGetBoolean("hand");
idlevel = ajAcdGetFloat("sidlevel");
noeff = ajAcdGetBoolean("noeff");
swentry = ajAcdGetFloat("swentry");
swexit = ajAcdGetFloat("swexit");
verbosity = ajAcdGetBoolean("verbosity");
weighting = ajAcdGetListSingle("weighting");
hmmfile = ajAcdGetOutfile("hmmfile");
o = ajAcdGetOutfile("o");
cfile = ajAcdGetOutfile("cfile");
/* MAIN APPLICATION CODE */
/* 1. Housekeeping */
cmd = ajStrNew();
rnd1 = ajStrNew();
rnd2 = ajStrNew();
tmp = ajStrNew();
fmt = ajStrNew();
hmmfilename = ajStrNew();
ajStrAssignC(&hmmfilename, ajFileGetNameC(hmmfile));
/* 2. Ensure alignfile is in format HMMER can understand. These include
FASTA, GENBANK,EMBL, GCG, PIR, STOCKHOLM, SELEX, MSF,CLUSTAL and PHYLIP.
EMBOSS name definitions are taken from seqInFormatDef in ajseqread.c and
seqOutFormat in ajseqwrite.c */
fmtok=ajFalse;
ajStrAssignS(&fmt, ajSeqsetGetFormat(alignfile));
if(ajStrMatchC(fmt, "fasta") ||
ajStrMatchC(fmt, "genbank") ||
ajStrMatchC(fmt, "embl") ||
ajStrMatchC(fmt, "gcg") ||
ajStrMatchC(fmt, "pir") ||
ajStrMatchC(fmt, "stockholm")||
ajStrMatchC(fmt, "selex") ||
ajStrMatchC(fmt, "msf") ||
ajStrMatchC(fmt, "clustal") ||
ajStrMatchC(fmt, "phylip"))
fmtok = ajTrue;
/* This could be replaced with code to reformat the file. */
if(!fmtok)
ajFatal("Input alignment ('alignfile' ACD option) is not in format "
"HMMER understands. Please use a a file in FASTA, GENBANK, "
"EMBL, GCG, PIR, STOCKHOLM, SELEX, MSF,CLUSTAL or PHYLIP format.");
/* 3. Build hmmbuild command line */
/* Command line is built in this order:
i. Application name.
ii. HMMER 'options' (in order they appear in ACD file)
iii.HMMER 'options' (that don't appear in ACD file)
iv. HMMER & new parameters.
*/
ajStrAssignS(&cmd, ajAcdGetpathC("hmmbuild"));
if(prior)
ajFmtPrintAppS(&cmd, " --prior %s ", ajFileGetNameC(prior));
if(null)
ajFmtPrintS(&cmd, " --null %s ", ajFileGetNameC(null));
if(pam)
ajFmtPrintAppS(&cmd, " --pam %s --pamwgt %f ", ajFileGetNameC(pam), pamwgt);
ajFmtPrintAppS(&cmd, " -n %S ", nhmm);
/* ACD option only allows one selection */
option = ajStrGetCharFirst(strategy);
if(option == 'F')
ajStrAppendC(&cmd, " -f ");
else if(option == 'G')
ajStrAppendC(&cmd, " -g ");
else if(option == 'S')
ajStrAppendC(&cmd, " -s ");
/* else go with default ('D' option in ACD file) */
ajFmtPrintAppS(&cmd, " --pbswitch %d ", pbswitch);
ajFmtPrintAppS(&cmd, " --archpri %f ", archpri);
if(binary)
ajStrAppendC(&cmd, " --binary ");
if(fast)
ajFmtPrintAppS(&cmd, " --fast --gapmax %f ", gapmax);
if(hand)
ajStrAppendC(&cmd, " --hand ");
ajFmtPrintAppS(&cmd, " --idlevel %f ", idlevel);
if(noeff)
ajStrAppendC(&cmd, " --noeff ");
ajFmtPrintAppS(&cmd, " --swentry %f ", swentry);
ajFmtPrintAppS(&cmd, " --swexit %f ", swexit);
if(verbosity)
ajStrAppendC(&cmd, " --verbose ");
/* ACD option only allows one selection */
option = ajStrGetCharFirst(weighting);
if(option == 'B')
ajStrAppendC(&cmd, " --wblosum ");
else if(option == 'G')
ajStrAppendC(&cmd, " --wgsc ");
else if(option == 'K')
ajStrAppendC(&cmd, " --wme ");
else if(option == 'W')
ajStrAppendC(&cmd, " --wpb ");
else if(option == 'V')
ajStrAppendC(&cmd, " --wvoronoi ");
else if(option == 'N')
ajStrAppendC(&cmd, " --wnone ");
if(o)
ajFmtPrintAppS(&cmd, " -o %s ", ajFileGetNameC(o));
if(cfile)
ajFmtPrintAppS(&cmd, " --cfile %s ", ajFileGetNameC(cfile));
/* -A (append) always set but file will be wiped by EMBOSS first unless
append: "Y" is set for "hmmfile" in the ACD file. */
ajStrAppendC(&cmd, " -A -F ");
ajFmtPrintAppS(&cmd, " %S %S", hmmfilename, ajSeqsetGetFilename(alignfile));
/* 4. Close ACD files */
ajSeqsetDel(&alignfile);
ajFileClose(&prior);
ajFileClose(&null);
ajFileClose(&pam);
ajFileClose(&hmmfile);
ajFileClose(&o);
ajFileClose(&cfile);
/* 5. Call hmmbuild */
ajFmtPrint("\n%S\n", cmd);
system(ajStrGetPtr(cmd));
/* 6. Exit cleanly */
ajStrDel(&nhmm);
ajStrDel(&cmd);
ajStrDel(&rnd1);
ajStrDel(&rnd2);
ajStrDel(&tmp);
ajStrDel(&fmt);
ajStrDel(&hmmfilename);
ajStrDel(&strategy);
ajStrDel(&weighting);
embExit();
return 0;
}
int main(int argc, char **argv)
{
/* ACD data item variables */
AjPFile oldhmmfile = NULL;
AjPStr format = NULL;
AjPFile newhmmfile = NULL;
/* Housekeeping variables */
AjPStr cmd = NULL;
char option;
/* ACD file processing */
embInitPV("ehmmconvert",argc,argv,"HMMERNEW",VERSION);
oldhmmfile = ajAcdGetInfile("oldhmmfile");
format = ajAcdGetListSingle("format");
newhmmfile = ajAcdGetOutfile("newhmmfile");
/* MAIN APPLICATION CODE */
cmd = ajStrNew();
/* 1. Build hmmconvert command line */
/* Command line is built in this order:
i. Application name.
ii. HMMER 'options' (in order they appear in ACD file)
iii.HMMER 'options' (that don't appear in ACD file)
iv. HMMER & new parameters.
*/
ajStrAssignS(&cmd, ajAcdGetpathC("hmmconvert"));
/* ACD option only allows one selection */
option = ajStrGetCharFirst(format);
if(option == 'A')
ajStrAppendC(&cmd, " -a ");
else if(option == 'B')
ajStrAppendC(&cmd, " -b ");
else if(option == 'G')
ajStrAppendC(&cmd, " -p ");
else if(option == 'X')
ajStrAppendC(&cmd, " -P ");
/* -A (append) always set but file will be wiped by EMBOSS first unless
append: "Y" is set for "newhmmfile" in the ACD file. */
/* hmmer cannot append to empty file, so only set the -A (append) option
if file was opened for appending to (and therefore was not wiped by
EMBOSS) and is not zero size. */
if(ajFileIsAppend(newhmmfile) && (ajFilenameGetSize(ajFileGetNameS(newhmmfile))!=-1))
ajStrAppendC(&cmd, " -A ");
ajStrAppendC(&cmd, " -F ");
ajFmtPrintAppS(&cmd, " %s %s", ajFileGetNameC(oldhmmfile), ajFileGetNameC(newhmmfile));
/* 2. Close ACD files. */
ajFileClose(&oldhmmfile);
ajFileClose(&newhmmfile);
/* 3. Call hmmconvert */
ajFmtPrint("\n%S\n\n", cmd);
ajSysExecS(cmd);
/* 4. Exit cleanly */
ajStrDel(&cmd);
ajStrDel(&format);
embExit();
return 0;
}
int main(int argc, char **argv)
{
const char *hmmfile; /* file to read HMMs from */
FILE *fp; /* output file handle */
HMMFILE *hmmfp; /* opened hmmfile for reading */
struct plan7_s *hmm; /* HMM to generate from */
int L; /* length of a sequence */
int i; /* counter over sequences */
char *ofile; /* output sequence file */
int nseq; /* number of seqs to sample */
int seed; /* random number generator seed */
int be_quiet; /* TRUE to silence header/footer */
int do_alignment; /* TRUE to output in aligned format */
int do_consensus; /* TRUE to do a single consensus seq */
AjBool ajselex;
AjBool ajcons;
AjPFile inf=NULL;
AjPFile outf=NULL;
AjPStr instr=NULL;
AjPStr outstr=NULL;
#ifdef MEMDEBUG
unsigned long histid1, histid2, orig_size, current_size;
orig_size = malloc_inuse(&histid1);
fprintf(stderr, "[... memory debugging is ON ...]\n");
#endif
/***********************************************
* Parse command line
***********************************************/
nseq = 10;
be_quiet = FALSE;
do_alignment = FALSE;
do_consensus = FALSE;
ofile = NULL;
embInitPV("ohmmemit",argc,argv,"HMMER",VERSION);
ajselex = ajAcdGetBoolean("selex");
ajcons = ajAcdGetBoolean("consensus");
nseq = ajAcdGetInt("number");
seed = ajAcdGetInt("seed");
inf = ajAcdGetInfile("infile");
outf = ajAcdGetOutfile("outfile");
if(!seed)
seed = time ((time_t *) NULL);
if(ajselex)
do_alignment=TRUE;
else
do_alignment=FALSE;
if(ajcons)
do_consensus=TRUE;
else
do_consensus=FALSE;
instr = ajStrNewC((char *)ajFileGetNameC(inf));
outstr = ajStrNewC((char *)ajFileGetNameC(outf));
hmmfile = ajStrGetPtr(instr);
sre_srandom(seed);
if (do_alignment && do_consensus)
ajFatal("Sorry, -selex and -consensus are incompatible.\n");
if (nseq != 10 && do_consensus)
ajWarn("-consensus overrides -number (# of sampled seqs)");
/***********************************************
* Open HMM file (might be in HMMERDB or current directory).
* Read a single HMM from it.
***********************************************/
if ((hmmfp = HMMFileOpen(hmmfile, "HMMERDB")) == NULL)
ajFatal("Failed to open HMM file %s\n", hmmfile);
if (!HMMFileRead(hmmfp, &hmm))
ajFatal("Failed to read any HMMs from %s\n", hmmfile);
HMMFileClose(hmmfp);
if (hmm == NULL)
ajFatal("HMM file %s corrupt or in incorrect format? Parse failed",
hmmfile);
/* Configure the HMM to shut off N,J,C emission: so we
* do a simple single pass through the model.
*/
Plan7NakedConfig(hmm);
Plan7Renormalize(hmm);
/***********************************************
* Open the output file, or stdout
***********************************************/
fp = ajFileGetFileptr(outf);
/***********************************************
* Show the options banner
***********************************************/
be_quiet=TRUE;
if (! be_quiet)
{
printf("HMM file: %s\n", hmmfile);
if (! do_consensus)
{
printf("Number of seqs: %d\n", nseq);
printf("Random seed: %d\n", seed);
}
printf("- - - - - - - - - - - - - - - - - - - - - - - - - "
"- - - - - - -\n\n");
}
/***********************************************
* Do the work.
* If we're generating an alignment, we have to collect
* all our traces, then output. If we're generating unaligned
* sequences, we can emit one at a time.
***********************************************/
if (do_consensus)
{
char *seq;
SQINFO sqinfo; /* info about sequence (name/desc) */
EmitConsensusSequence(hmm, &seq, NULL, &L, NULL);
strcpy(sqinfo.name, "consensus");
sqinfo.len = L;
sqinfo.flags = SQINFO_NAME | SQINFO_LEN;
WriteSeq(fp, kPearson, seq, &sqinfo);
free(seq);
}
else if (do_alignment)
{
struct p7trace_s **tr;
char **dsq;
SQINFO *sqinfo;
char **aseq;
AINFO ainfo;
float *wgt;
dsq = MallocOrDie(sizeof(char *) * nseq);
tr = MallocOrDie(sizeof(struct p7trace_s *) * nseq);
sqinfo = MallocOrDie(sizeof(SQINFO) * nseq);
wgt = MallocOrDie(sizeof(float) * nseq);
FSet(wgt, nseq, 1.0);
for (i = 0; i < nseq; i++)
{
EmitSequence(hmm, &(dsq[i]), &L, &(tr[i]));
sprintf(sqinfo[i].name, "seq%d", i+1);
sqinfo[i].len = L;
sqinfo[i].flags = SQINFO_NAME | SQINFO_LEN;
}
P7Traces2Alignment(dsq, sqinfo, wgt, nseq, hmm->M, tr, FALSE,
&aseq, &ainfo);
/* Output the alignment */
WriteSELEX(fp, aseq, &ainfo, 50);
if (ofile != NULL && !be_quiet)
printf("Alignment saved in file %s\n", ofile);
/* Free memory
*/
for (i = 0; i < nseq; i++)
{
P7FreeTrace(tr[i]);
free(dsq[i]);
}
FreeAlignment(aseq, &ainfo);
free(sqinfo);
free(dsq);
free(wgt);
free(tr);
}
else /* unaligned sequence output */
{
struct p7trace_s *tr;
char *dsq;
char *seq;
SQINFO sqinfo;
for (i = 0; i < nseq; i++)
{
EmitSequence(hmm, &dsq, &L, &tr);
sprintf(sqinfo.name, "seq%d", i+1);
sqinfo.len = L;
sqinfo.flags = SQINFO_NAME | SQINFO_LEN;
seq = DedigitizeSequence(dsq, L);
WriteSeq(fp, kPearson, seq, &sqinfo);
P7FreeTrace(tr);
free(dsq);
free(seq);
}
}
ajFileClose(&outf);
FreePlan7(hmm);
SqdClean();
#ifdef MEMDEBUG
current_size = malloc_inuse(&histid2);
if (current_size != orig_size)
malloc_list(2, histid1, histid2);
else
fprintf(stderr, "[No memory leaks.]\n");
#endif
ajStrDel(&instr);
ajStrDel(&outstr);
ajFileClose(&inf);
ajFileClose(&outf);
embExit();
return 0;
}
int main(int argc, char **argv)
{
/* ACD data item variables */
AjPFile hmmfile = NULL;
AjPSeqall seqfile = NULL;
AjBool nuc = 0;
ajint A = 0;
float E = 0.0;
float T = 0.0;
ajint Z = 0;
AjBool acc = ajFalse;
AjBool compat = ajFalse;
ajint cpu = 0;
AjBool cutga = ajFalse;
AjBool cuttc = ajFalse;
AjBool cutnc = ajFalse;
float dome = 0.0;
float domt = 0.0;
AjBool forward = ajFalse;
AjBool nulltwo = ajFalse;
AjBool pvm = ajFalse;
AjBool xnu = ajFalse;
AjPStr outname = NULL;
/* Housekeeping variables */
AjPStr cmd = NULL;
AjPStr rnd = NULL;
AjPSeqout rndo = NULL;
AjPSeq seq = NULL;
/* ACD file processing */
embInitPV("ehmmpfam",argc,argv,"HMMERNEW",VERSION);
hmmfile = ajAcdGetInfile("hmmfile");
seqfile = ajAcdGetSeqall("seqfile");
nuc = ajAcdGetBoolean("nuc");
A = ajAcdGetInt("A");
E = ajAcdGetFloat("E");
T = ajAcdGetFloat("T");
Z = ajAcdGetInt("Z");
acc = ajAcdGetBoolean("acc");
compat = ajAcdGetBoolean("compat");
cpu = ajAcdGetInt("cpu");
cutga = ajAcdGetBoolean("cutga");
cuttc = ajAcdGetBoolean("cuttc");
cutnc = ajAcdGetBoolean("cutnc");
dome = ajAcdGetFloat("dome");
domt = ajAcdGetFloat("domt");
forward = ajAcdGetBoolean("forward");
nulltwo = ajAcdGetBoolean("nulltwo");
pvm = ajAcdGetBoolean("pvm");
xnu = ajAcdGetBoolean("xnu");
outname = ajAcdGetOutfileName("outfile");
/* MAIN APPLICATION CODE */
/* 1. Housekeeping */
cmd = ajStrNew();
rnd = ajStrNew();
/* 2. Re-write seqfile to a temporary file in a format (fasta) HMMER can understand.
We cannot just pass the name of seqfile to HMMER as the name provided might be a
USA which HMMER would not understand. */
ajFilenameSetTempname(&rnd);
rndo = ajSeqoutNew();
if(!ajSeqoutOpenFilename(rndo, rnd))
ajFatal("Terminal ajSeqFileNewOut failure. Email EMBOSS helpdesk!\n");
ajSeqoutSetFormatC(rndo, "fasta");
while(ajSeqallNext(seqfile, &seq))
ajSeqoutWriteSeq(rndo, seq);
ajSeqoutClose(rndo);
ajSeqoutDel(&rndo);
/* 2. Build hmmpfam command line */
/* Command line is built in this order:
i. Application name.
ii. HMMER 'options' (in order they appear in ACD file)
iii.HMMER 'options' (that don't appear in ACD file)
iv. HMMER & new parameters.
*/
ajStrAssignS(&cmd, ajAcdGetpathC("hmmpfam"));
if(nuc)
ajStrAppendC(&cmd, " -n ");
ajFmtPrintAppS(&cmd, " -A %d -E %f -T %f -Z %d", A, E, T, Z);
if(acc)
ajStrAppendC(&cmd, " --acc ");
if(compat)
ajStrAppendC(&cmd, " --compat ");
if(cpu)
ajFmtPrintAppS(&cmd, " --cpu %d ", cpu);
if(cutga)
ajStrAppendC(&cmd, " --cutga ");
if(cuttc)
ajStrAppendC(&cmd, " --cuttc ");
if(cutnc)
ajStrAppendC(&cmd, " --cutnc ");
ajFmtPrintAppS(&cmd, " --domE %f --domT %f ", dome, domt);
if(forward)
ajStrAppendC(&cmd, " --forward ");
if(nulltwo)
ajStrAppendC(&cmd, " --null2 ");
if(pvm)
ajStrAppendC(&cmd, " --pvm ");
if(xnu)
ajStrAppendC(&cmd, " --xnu ");
/* Note output redirected to outname.
rnd is the name of the rewritten seqfile.
MUST specify FASTA format explicitly. */
ajFmtPrintAppS(&cmd, " --informat FASTA %s %S",
ajFileGetNameC(hmmfile),
rnd);
/* 3. Close ACD files. */
ajFileClose(&hmmfile);
ajSeqallDel(&seqfile);
/* 4. Call hmmpfam. Use C system call instead of ajSystem
so that redirect in cmd works ok. */
ajFmtPrint("\n%S\n\n", cmd);
ajSysExecOutnameAppendS(cmd,outname);
/* 5. Exit cleanly */
ajSysFileUnlinkS(rnd);
ajStrDel(&cmd);
ajStrDel(&rnd);
ajStrDel(&outname);
embExit();
return 0;
}
int main(int argc, char **argv)
{
/* ACD data item variables */
AjPSeqset dataset = NULL;
AjPFile bfile = NULL;
AjPFile plib = NULL;
AjPStr mod = NULL;
ajint nmotifs = 0;
AjBool text = ajFalse;
AjPStr prior = NULL;
float evt = 0.0;
ajint nsites = 0;
ajint minsites = 0;
ajint maxsites = 0;
float wnsites = 0.0;
ajint w = 0;
ajint minw = 0;
ajint maxw = 0;
AjBool nomatrim = ajFalse;
ajint wg = 0;
ajint ws = 0;
AjBool noendgaps = ajFalse;
AjBool revcomp = ajFalse;
AjBool pal = ajFalse;
AjBool nostatus = ajFalse;
ajint maxiter = 0;
float distance = 0.0;
float b = 0.0;
float spfuzz = 0.0;
AjPStr spmap = NULL;
AjPStr cons = NULL;
ajint maxsize = 0;
ajint p = 0;
ajint time = 0;
AjPStr sf = NULL;
ajint heapsize = 64;
AjBool xbranch = ajFalse;
AjBool wbranch = ajFalse;
ajint bfactor = 0;
AjPFile outtext = NULL;
/* Housekeeping variables */
AjPStr cmd = NULL;
AjPStr ssname = NULL;
AjPSeqout outseq = NULL;
AjPStr tmp = NULL;
char option;
/* ACD file processing */
embInitPV("ememetext",argc,argv,"MEME",VERSION);
dataset = ajAcdGetSeqset("dataset");
bfile = ajAcdGetInfile("bfile");
plib = ajAcdGetInfile("plibfile");
mod = ajAcdGetSelectSingle("mod");
nmotifs = ajAcdGetInt("nmotifs");
text = ajAcdGetBoolean("text");
prior = ajAcdGetSelectSingle("prior");
evt = ajAcdGetFloat("evt");
nsites = ajAcdGetInt("nsites");
minsites = ajAcdGetInt("minsites");
maxsites = ajAcdGetInt("maxsites");
wnsites = ajAcdGetFloat("wnsites");
w = ajAcdGetInt("w");
minw = ajAcdGetInt("minw");
maxw = ajAcdGetInt("maxw");
nomatrim = ajAcdGetBoolean("nomatrim");
wg = ajAcdGetInt("wg");
ws = ajAcdGetInt("ws");
noendgaps = ajAcdGetBoolean("noendgaps");
revcomp = ajAcdGetBoolean("revcomp");
pal = ajAcdGetBoolean("pal");
nostatus = ajAcdGetBoolean("nostatus");
maxiter = ajAcdGetInt("maxiter");
distance = ajAcdGetFloat("distance");
b = ajAcdGetFloat("b");
spfuzz = ajAcdGetFloat("spfuzz");
spmap = ajAcdGetSelectSingle("spmap");
cons = ajAcdGetString("cons");
maxsize = ajAcdGetInt("maxsize");
p = ajAcdGetInt("p");
time = ajAcdGetInt("time");
sf = ajAcdGetString("sf");
heapsize = ajAcdGetInt("heapsize");
xbranch = ajAcdGetBoolean("xbranch");
wbranch = ajAcdGetBoolean("wbranch");
bfactor = ajAcdGetInt("bfactor");
outtext = ajAcdGetOutfile("outtext");
outseq = ajAcdGetSeqoutset("outseq");
/* MAIN APPLICATION CODE */
/* 1. Housekeeping */
cmd = ajStrNew();
tmp = ajStrNew();
/* 2. Re-write dataset to a temporary file in a format (fasta) MEME
** can understand.
** Can't just pass the name of dataset to MEME as the name provided
** might be a USA which MEME would not understand.
*/
ssname = ajStrNewS(ajFileGetNameS(outseq->File));
ajSeqoutSetFormatC(outseq, "fasta");
ajSeqoutWriteSet(outseq, dataset);
ajSeqoutClose(outseq);
ajSeqoutDel(&outseq);
/* 3. Build ememe command line */
/* Command line is built in this order:
i. Application name.
ii. Original MEME options (in order they appear in ACD file)
iii.Original MEME options (that don't appear in ACD file)
iv. EMBASSY MEME new qualifiers and parameters.
*/
ajStrAssignS(&cmd, ajAcdGetpathC("meme"));
ajFmtPrintAppS(&cmd, " %S", ssname);
if(bfile)
ajFmtPrintAppS(&cmd, " -bfile %s ", ajFileGetNameC(bfile));
if(plib)
ajFmtPrintAppS(&cmd, " -plib %s ", ajFileGetNameC(plib));
option = ajStrGetCharFirst(mod);
if(option == 'o')
ajStrAppendC(&cmd, " -mod oops ");
else if(option == 'z')
ajStrAppendC(&cmd, " -mod zoops ");
else if(option == 'a')
ajStrAppendC(&cmd, " -mod anr ");
if(nmotifs != 1)
ajFmtPrintAppS(&cmd, " -nmotifs %d ", nmotifs);
if(text)
ajFmtPrintAppS(&cmd, " -text ");
ajFmtPrintAppS(&cmd, " -prior %S ", prior);
if(evt != -1)
ajFmtPrintAppS(&cmd, " -evt %f ", evt);
if(nsites != -1)
ajFmtPrintAppS(&cmd, " -nsites %d ", nsites);
else
{
if(minsites != -1)
ajFmtPrintAppS(&cmd, " -minsites %d ", minsites);
if(maxsites != -1)
ajFmtPrintAppS(&cmd, " -maxsites %d ", maxsites);
}
if(wnsites < 0.7999 || wnsites > .8001)
ajFmtPrintAppS(&cmd, " -wnsites %f ", wnsites);
if(w != -1)
ajFmtPrintAppS(&cmd, " -w %d ", w);
if(minw != 8)
ajFmtPrintAppS(&cmd, " -minw %d ", minw);
if(maxw != 50)
ajFmtPrintAppS(&cmd, " -maxw %d ", maxw);
if(nomatrim)
ajFmtPrintAppS(&cmd, " -nomatrim ");
if(wg != 11)
ajFmtPrintAppS(&cmd, " -wg %d ", wg);
if(ws != 1)
ajFmtPrintAppS(&cmd, " -ws %d ", ws);
if(noendgaps)
ajFmtPrintAppS(&cmd, " -noendgaps ");
if(revcomp)
ajFmtPrintAppS(&cmd, " -revcomp ");
if(pal && ajSeqsetIsNuc(dataset))
ajFmtPrintAppS(&cmd, " -pal ");
if(nostatus)
ajFmtPrintAppS(&cmd, " -nostatus ");
if(maxiter != 50)
ajFmtPrintAppS(&cmd, " -maxiter %d ", maxiter);
if(distance < 0.00099 || distance > 0.00101)
ajFmtPrintAppS(&cmd, " -distance %f ", distance);
if(b != -1)
ajFmtPrintAppS(&cmd, " -b %f ", b);
if(spfuzz != -1)
ajFmtPrintAppS(&cmd, " -spfuzz %f ", spfuzz);
if(!ajStrMatchC(spmap,"default"))
ajFmtPrintAppS(&cmd, " -spmap %S ", spmap);
if(MAJSTRGETLEN(cons))
ajFmtPrintAppS(&cmd, "-cons %S", cons);
if(maxsize != -1)
ajFmtPrintAppS(&cmd, " -maxsize %d ", maxsize);
if(p > 0)
ajFmtPrintAppS(&cmd, " -p %d ", p);
if(time > 0)
ajFmtPrintAppS(&cmd, " -time %d ", time);
if(MAJSTRGETLEN(sf))
ajFmtPrintAppS(&cmd, " -sf %S", sf);
if(heapsize != 64)
ajFmtPrintAppS(&cmd, " -heapsize %d ", heapsize);
if(xbranch)
ajFmtPrintAppS(&cmd, " -x_branch");
if(wbranch)
ajFmtPrintAppS(&cmd, " -w_branch");
if(bfactor != 3)
ajFmtPrintAppS(&cmd, " -bfactor %d ", bfactor);
if(ajSeqsetIsProt(dataset))
ajFmtPrintAppS(&cmd, "-protein ");
else
ajFmtPrintAppS(&cmd, "-dna ");
ajFmtPrintAppS(&cmd, " -text");
ajFmtPrintAppS(&cmd, " > %S ", ajFileGetNameS(outtext));
/* 4. Close files from ACD before calling meme */
ajFileClose(&bfile);
ajFileClose(&plib);
/* 5. Call meme */
/* ajFmtPrint("\n%S\n", cmd); */
system(ajStrGetPtr(cmd));
/* 6. Exit cleanly */
ajSeqsetDel(&dataset);
ajStrDel(&cons);
ajStrDel(&sf);
ajStrDel(&mod);
ajStrDel(&prior);
ajStrDel(&spmap);
ajStrDel(&cmd);
ajStrDel(&ssname);
ajStrDel(&tmp);
ajFileClose(&bfile);
ajFileClose(&plib);
ajFileClose(&outtext);
ajSeqoutDel(&outseq);
embExit();
return 0;
}
static AjBool assemoutWriteBam(AjPFile outf, const AjPAssem assem)
{
AjPSeqBamHeader header = NULL;
AjPAssemContig c = NULL;
AjPSeqBam bam;
AjPAssemRead r = NULL;
AjPAssemContig* contigs = NULL;
AjPAssemTag t = NULL;
AjIList j = NULL;
AjPSeqBamBgzf gzfile = NULL;
AjPStr headertext=NULL;
const AjPStr rgheadertext=NULL;
AjBool ret = ajTrue;
ajint i=0;
ajulong ncontigs=0UL;
if(ajListGetLength(assem->ContigsOrder))
ncontigs = ajListToarray(assem->ContigsOrder, (void***)&contigs);
else
ncontigs = ajTableToarrayValues(assem->Contigs, (void***)&contigs);
AJNEW0(bam);
bam->m_data=10;
AJCNEW0(bam->data, bam->m_data);
gzfile = ajSeqBamBgzfNew(ajFileGetFileptr(outf), "w");
ajFmtPrintS(&headertext, "@HD\tVN:1.3\tSO:%s\n",
ajAssemGetSortorderC(assem));
header = ajSeqBamHeaderNewN((ajuint) ncontigs);
while (contigs[i]) /* contigs */
{
c = contigs[i];
if(ajStrMatchC(c->Name, "*"))
{
i++;
continue;
}
header->target_name[i] = strdup(ajStrGetPtr(c->Name));
header->target_len[i++] = c->Length;
ajFmtPrintAppS(&headertext, "@SQ\tSN:%S\tLN:%d",
c->Name, c->Length);
if(c->URI)
ajFmtPrintAppS(&headertext, "\tUR:%S", c->URI);
if(c->MD5)
ajFmtPrintAppS(&headertext, "\tM5:%S", c->MD5);
if(c->Species)
ajFmtPrintAppS(&headertext, "\tSP:%S", c->Species);
ajFmtPrintAppS(&headertext, "\n");
j = ajListIterNewread(c->Tags);
while (!ajListIterDone(j))
{
t = ajListIterGet(j);
ajFmtPrintAppS(&headertext,
"@CO\t%S %u %u %S\n", t->Name, t->x1, t->y1,
t->Comment);
}
ajListIterDel(&j);
}
rgheadertext = assemSAMGetReadgroupHeaderlines(assem);
if(rgheadertext)
ajStrAppendS(&headertext, rgheadertext);
ajSeqBamHeaderSetTextC(header, ajStrGetPtr(headertext));
ajSeqBamHeaderWrite(gzfile, header);
j = ajListIterNewread(assem->Reads);
while (!ajListIterDone(j)) /* reads */
{
r = ajListIterGet(j);
assemoutWriteBamAlignment(gzfile, r, bam);
}
ajListIterDel(&j);
ajSeqBamBgzfClose(gzfile);
ajSeqBamHeaderDel(&header);
ajStrDel(&headertext);
AJFREE(contigs);
AJFREE(bam->data);
AJFREE(bam);
ajBamIndexBuild(ajFileGetNameC(outf));
return ret;
}
int main(int argc, char **argv)
{
/* ACD data item variables */
AjPFile hmmfile = NULL;
AjPSeqset seqfile = NULL;
AjPFile mapali = NULL;
AjPFile withali = NULL;
AjPAlign o = NULL;
AjBool m = ajFalse;
AjBool q = ajFalse;
/* Housekeeping variables */
AjPStr cmd = NULL;
AjPStr fmt = NULL;
AjBool fmtok = ajFalse;
AjPStr rnd = NULL;
AjPSeqout rndo = NULL;
/* ACD file processing */
embInitPV("ehmmalign",argc,argv,"HMMERNEW",VERSION);
hmmfile = ajAcdGetInfile("hmmfile");
seqfile = ajAcdGetSeqset("seqfile");
mapali = ajAcdGetInfile("mapali");
withali = ajAcdGetInfile("withali");
o = ajAcdGetAlign("o");
m = ajAcdGetBoolean("m");
q = ajAcdGetBoolean("q");
/* MAIN APPLICATION CODE */
/* 1. Housekeeping */
cmd = ajStrNew();
fmt = ajStrNew();
rnd = ajStrNew();
/* 2. Re-write seqfile to a temporary file in a format (fasta) HMMER can understand.
We cannot just pass the name of seqfile to HMMER as the name provided might be a
USA which HMMER would not understand. */
ajFilenameSetTempname(&rnd);
rndo = ajSeqoutNew();
if(!ajSeqoutOpenFilename(rndo, rnd))
ajFatal("Terminal ajSeqFileNewOut failure. Email EMBOSS helpdesk!\n");
ajSeqoutSetFormatC(rndo, "fasta");
ajSeqoutWriteSet(rndo, seqfile);
ajSeqoutClose(rndo);
ajSeqoutDel(&rndo);
/* 3. Build hmmalign command line */
/* Command line is built in this order:
i. Application name.
ii. HMMER 'options' (in order they appear in ACD file)
iii.HMMER 'options' (that don't appear in ACD file)
iv. HMMER & new parameters.
*/
ajFmtPrintS(&cmd, "%S ", ajAcdGetpathC("hmmalign"));
if(mapali)
ajFmtPrintAppS(&cmd, " --mapali %s ", ajFileGetNameC(mapali));
if(withali)
ajFmtPrintAppS(&cmd, " --withali %s ", ajFileGetNameC(withali));
if(m)
ajStrAppendC(&cmd, " -m ");
if(q)
ajStrAppendC(&cmd, " -q ");
/* Ensure output alignment is in user-specified format. */
fmtok=ajTrue;
ajStrAssignS(&fmt, ajAlignGetFormat(o));
/* fasta and a2m are identical formats. */
if(ajStrMatchC(fmt, "fasta"))
ajStrAssignC(&fmt, "A2M");
else if(ajStrMatchC(fmt, "a2m"))
ajStrAssignC(&fmt, "A2M");
else if(ajStrMatchC(fmt, "msf"))
ajStrAssignC(&fmt, "MSF");
else if(ajStrMatchC(fmt, "phylip"))
ajStrAssignC(&fmt, "PHYLIP");
/* hmmer also supports stockholm, SELEX & Clustal output, EMBOSS does not.
EMBOSS supports unknown/multiple/simple and srs output, hmmer does not. */
else
fmtok = ajFalse;
if(!fmtok)
{
/* This could be replaced with code to reformat the file. */
ajWarn("Specified output alignment format ('o' ACD option) is "
"not understood by HMMER. Using stockholm format instead.");
ajStrAssignC(&fmt, "Stockholm");
}
/* rnd is the name of the rewritten seqfile. MUST specify FASTA format explicitly. */
ajFmtPrintAppS(&cmd, " --informat FASTA --outformat %S -o %s %s %S",
fmt,
ajAlignGetFilename(o),
ajFileGetNameC(hmmfile),
rnd);
/* 4. Close ACD files */
ajFileClose(&hmmfile);
ajSeqsetDel(&seqfile);
ajFileClose(&mapali);
ajFileClose(&withali);
ajAlignClose(o);
ajAlignDel(&o);
/* 5. Call hmmalign */
ajFmtPrint("\n%S\n\n", cmd);
ajSysExecS(cmd);
/* 6. Exit cleanly */
ajSysFileUnlinkS(rnd);
ajStrDel(&cmd);
ajStrDel(&fmt);
ajStrDel(&rnd);
embExit();
return 0;
}
int main(int argc, char **argv)
{
const char *hmmfile; /* file to read HMMs from */
HMMFILE *hmmfp; /* opened hmmfile for reading */
const char *seqfile; /* file to read target sequence from */
char **rseq; /* raw, unaligned sequences */
SQINFO *sqinfo; /* info associated with sequences */
char **dsq; /* digitized raw sequences */
int nseq; /* number of sequences */
char **aseq; /* aligned sequences */
AINFO ainfo; /* alignment information */
float *wgt; /* per-sequence weights */
int i;
struct plan7_s *hmm; /* HMM to align to */
struct p7trace_s **tr; /* traces for aligned sequences */
int be_quiet; /* TRUE to suppress verbose banner */
int matchonly; /* TRUE to show only match state syms */
const char *outfile; /* optional alignment output file */
FILE *ofp; /* handle on alignment output file */
AjPFile ajwithali; /* name of additional alignment file to align */
AjPFile ajmapali; /* name of additional alignment file to map */
AjBool ajmatch=ajFalse;
AjPFile outf=NULL;
AjPStr outfname=NULL;
AjPFile inf=NULL;
AjPStr infname=NULL;
AjPSeqset seqset=NULL;
AjPStr ajseqfile=NULL;
char* mapali=NULL;
char* withali=NULL;
#ifdef MEMDEBUG
unsigned long histid1, histid2, orig_size, current_size;
orig_size = malloc_inuse(&histid1);
fprintf(stderr, "[... memory debugging is ON ...]\n");
#endif
/***********************************************
* Parse command line
***********************************************/
matchonly = FALSE;
outfile = NULL;
be_quiet = FALSE;
withali = NULL;
mapali = NULL;
embInitPV("ohmmalign",argc,argv,"HMMER",VERSION);
ajmatch = ajAcdGetBoolean("matchonly");
if(ajmatch)
matchonly=TRUE;
else
matchonly=FALSE;
ajmapali = ajAcdGetInfile("mapalifile");
if (ajmapali)
mapali = ajCharNewS(ajFileGetNameS(ajmapali));
ajFileClose(&ajmapali);
ajwithali = ajAcdGetInfile("withalifile");
if (ajwithali)
withali = ajCharNewS(ajFileGetNameS(ajwithali));
ajFileClose(&ajwithali);
be_quiet=TRUE;
outf = ajAcdGetOutfile("outfile");
outfname = ajStrNewC((char *)ajFileGetNameC(outf));
if(*ajStrGetPtr(outfname)>31)
ajFileClose(&outf);
outfile = ajStrGetPtr(outfname);
inf = ajAcdGetInfile("hmmfile");
infname = ajStrNewC((char *)ajFileGetNameC(inf));
ajFileClose(&inf);
hmmfile = ajStrGetPtr(infname);
seqset = ajAcdGetSeqset("sequences");
ajseqfile = ajStrNewC(ajStrGetPtr(seqset->Filename));
seqfile = ajStrGetPtr(ajseqfile);
/***********************************************
* Open HMM file (might be in HMMERDB or current directory).
* Read a single HMM from it.
*
* Currently hmmalign disallows the J state and
* only allows one domain per sequence. To preserve
* the S/W entry information, the J state is explicitly
* disallowed, rather than calling a Plan7*Config() function.
* this is a workaround in 2.1 for the 2.0.x "yo!" bug.
***********************************************/
if ((hmmfp = HMMFileOpen(hmmfile, "HMMERDB")) == NULL)
ajFatal("Failed to open HMM file %s\n", hmmfile);
if (!HMMFileRead(hmmfp, &hmm))
ajFatal("Failed to read any HMMs from %s\n", hmmfile);
HMMFileClose(hmmfp);
if (hmm == NULL)
ajFatal("HMM file %s corrupt or in incorrect format? Parse failed", hmmfile);
hmm->xt[XTE][MOVE] = 1.; /* only 1 domain/sequence ("global" alignment) */
hmm->xt[XTE][LOOP] = 0.;
P7Logoddsify(hmm, TRUE);
/* do we have the map we might need? */
if (mapali != NULL && ! (hmm->flags & PLAN7_MAP))
ajFatal("HMMER: HMM file %s has no map; you can't use --mapali.", hmmfile);
/***********************************************
* Open sequence file in current directory.
* Read all seqs from it.
***********************************************/
/*
if (! SeqfileFormat(seqfile, &format, NULL))
switch (squid_errno) {
case SQERR_NOFILE:
ajFatal("Sequence file %s could not be opened for reading", seqfile);
case SQERR_FORMAT:
default:
ajFatal("Failed to determine format of sequence file %s", seqfile);
}
if (! ReadMultipleRseqs(seqfile, format, &rseq, &sqinfo, &nseq))
ajFatal("Failed to read any sequences from file %s", seqfile);
*/
emboss_rseqs(seqset,&rseq,&sqinfo,&nseq);
/***********************************************
* Show the banner
***********************************************/
be_quiet=TRUE;
if (! be_quiet)
{
/* Banner(stdout, banner); */
printf( "HMM file: %s\n", hmmfile);
printf( "Sequence file: %s\n", seqfile);
printf("- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -\n\n");
}
/***********************************************
* Do the work
***********************************************/
/* Allocations and initializations.
*/
dsq = MallocOrDie(sizeof(char *) * nseq);
tr = MallocOrDie(sizeof(struct p7trace_s *) * nseq);
/* Align each sequence to the model, collect traces
*/
for (i = 0; i < nseq; i++)
{
dsq[i] = DigitizeSequence(rseq[i], sqinfo[i].len);
if (P7ViterbiSize(sqinfo[i].len, hmm->M) <= RAMLIMIT)
(void) P7Viterbi(dsq[i], sqinfo[i].len, hmm, &(tr[i]));
else
(void) P7SmallViterbi(dsq[i], sqinfo[i].len, hmm, &(tr[i]));
}
/* Include an aligned alignment, if desired.
*/
if (mapali != NULL)
include_alignment(mapali, hmm, TRUE, &rseq, &dsq, &sqinfo, &tr, &nseq);
if (withali != NULL)
include_alignment(withali, hmm, FALSE, &rseq, &dsq, &sqinfo, &tr, &nseq);
/* Turn traces into a multiple alignment
*/
wgt = MallocOrDie(sizeof(float) * nseq);
FSet(wgt, nseq, 1.0);
P7Traces2Alignment(dsq, sqinfo, wgt, nseq, hmm->M, tr, matchonly,
&aseq, &ainfo);
/***********************************************
* Output the alignment
***********************************************/
if (outfile != NULL && (ofp = fopen(outfile, "w")) != NULL)
{
WriteSELEX(ofp, aseq, &ainfo, 50);
printf("Alignment saved in file %s\n", outfile);
fclose(ofp);
}
else
WriteSELEX(stdout, aseq, &ainfo, 50);
/***********************************************
* Cleanup and exit
***********************************************/
for (i = 0; i < nseq; i++)
{
P7FreeTrace(tr[i]);
FreeSequence(rseq[i], &(sqinfo[i]));
free(dsq[i]);
}
FreeAlignment(aseq, &ainfo);
FreePlan7(hmm);
free(sqinfo);
free(rseq);
free(dsq);
free(wgt);
free(tr);
SqdClean();
ajStrDel(&outfname);
ajStrDel(&infname);
ajStrDel(&ajseqfile);
#ifdef MEMDEBUG
current_size = malloc_inuse(&histid2);
if (current_size != orig_size) malloc_list(2, histid1, histid2);
else fprintf(stderr, "[No memory leaks.]\n");
#endif
ajSeqsetDel(&seqset);
ajFileClose(&ajwithali);
ajFileClose(&ajmapali);
embExit();
return 0;
}
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;
}
