SEXP r_cgiParameters() {
int i;
char *name;
Stringa value=stringCreate(16);
Texta keys=textCreate(8);
Texta values=textCreate(8);
SEXP r_keys, r_values;
cgiGetInit();
while(name = cgiGetNext(value)) {
textAdd(keys, name);
textAdd(values, string(value));
}
int n=arrayMax(keys);
PROTECT(r_keys=allocVector(STRSXP, n));
PROTECT(r_values=allocVector(STRSXP, n));
for(i=0; i<n; ++i) {
SET_STRING_ELT(r_keys, i, mkChar(textItem(keys,i)));
SET_STRING_ELT(r_values, i, mkChar(textItem(values,i)));
}
setNames(r_values, r_keys);
stringDestroy(value);
textDestroy(keys);
textDestroy(values);
UNPROTECT(2);
return(r_values);
}
SEXP r_cgiParam(SEXP r_param, SEXP ignore_case, SEXP r_default) {
if(r_param == R_NilValue) return(R_NilValue);
char* name;
Stringa value=stringCreate(16);
char *param=cStr(r_param);
char *str=NULL;
SEXP res;
int (*fPtr)(char*, char*);
fPtr=cBool(ignore_case) ? &myStrCaseEqual : &myStrEqual;
cgiGetInit();
while(name = cgiGetNext(value)) {
if((*fPtr)(name, param)) {
str=hlr_strdup(string(value));
break;
}
}
stringDestroy(value);
if(str) {
return mkString(str);
} else {
return r_default;
}
}
int main (int argc, char *argv[])
{
LineStream ls;
char *line;
char *pos;
Stringa buffer;
if (argc != 2) {
usage ("%s <file.intraOffsets>");
}
TH1 *his = new TH1D ("","Intra-read distribution",1000,0,1000);
TCanvas *canv = new TCanvas("","canvas",1200,400);
ls = ls_createFromFile (argv[1]);
while (line = ls_nextLine (ls)) {
his->Fill (atoi (line));
}
ls_destroy (ls);
his->Draw();
his->GetXaxis()->SetLabelSize (0.04);
his->GetYaxis()->SetLabelSize (0.04);
buffer = stringCreate (100);
pos = strchr (argv[1],'.');
if (pos == NULL) {
die ("Expected <file.intraOffsets>: %s",argv[1]);
}
*pos = '\0';
stringPrintf (buffer,"%s_intraDistribution.jpg",argv[1]);
canv->Print (string (buffer),"jpg");
stringDestroy (buffer);
return 0;
}
int main (int argc, char *argv[])
{
GfrEntry *currGE;
int count;
int countRemoved;
int i;
if (argc != 3) {
usage ("%s <offsetCutoff> <minNumUniqueReads>",argv[0]);
}
count = 0;
countRemoved = 0;
int offsetCutOff = atoi (argv[1]);
int minNumUniqueReads = atoi (argv[2]);
gfr_init ("-");
puts (gfr_writeHeader ());
while (currGE = gfr_nextEntry ()) {
Array starts = arrayCreate( 100, int);
for (i = 0; i < arrayMax( currGE->interReads ); i++) {
int currStart = arrp(currGE->interReads, i, GfrInterRead)->readStart1 + arrp(currGE->interReads, i, GfrInterRead)->readStart2;
array(starts, arrayMax(starts), int) = currStart;
}
arraySort( starts, (ARRAYORDERF) arrayIntcmp );
arrayUniq( starts, NULL, (ARRAYORDERF) arrayIntcmp ) ;
int numUniqeOffsets = arrayMax( starts );
arrayDestroy( starts );
if (arrayMax( currGE->readsTranscript1 ) != arrayMax( currGE->readsTranscript2 ) )
die( "The two ends have a different number of reads");
Texta reads = textCreate(arrayMax(currGE->readsTranscript1));
for (i = 0; i < arrayMax(currGE->readsTranscript1); i++) {
Stringa strA = stringCreate( strlen(textItem( currGE->readsTranscript1, i) ) * 2 + 1);
stringAppendf( strA, textItem( currGE->readsTranscript1,i));
stringAppendf( strA, textItem( currGE->readsTranscript2,i));
textAdd( reads, string(strA));
stringDestroy( strA );
}
textUniqKeepOrder( reads );
int numRemaining = arrayMax( reads );
textDestroy ( reads );
if (numRemaining <= minNumUniqueReads || numUniqeOffsets <= offsetCutOff) {
countRemoved++;
continue;
}
puts (gfr_writeGfrEntry (currGE));
count++;
}
gfr_deInit ();
warn("%s_PCRFilter: offset=%d minNumUniqueReads=%d",
argv[0],offsetCutOff, minNumUniqueReads);
warn("%s_numRemoved: %d",argv[0],countRemoved);
warn("%s_numGfrEntries: %d",argv[0],count);
return 0;
}
static char* getBreakPointSequence (char *tileCoordinate1, char *tileCoordinate2)
{
Stringa buffer;
Stringa targetsFile;
FILE *fp;
Array targetSeqs;
int i;
Seq *currSeq;
static Stringa sequence = NULL;
buffer = stringCreate (100);
targetsFile = stringCreate (100);
stringPrintf (targetsFile,"targets_%d.txt",getpid ());
if (!(fp = fopen (string (targetsFile),"w")) ){
die ("Unable to open target file: %s",string (targetsFile));
}
fprintf (fp,"%s\n%s",tileCoordinate1,tileCoordinate2);
fclose (fp);
stringPrintf (buffer,"%s %s/%s stdout -noMask -seqList=%s",
confp_get(Conf, "BLAT_TWO_BIT_TO_FA"),
confp_get(Conf, "BLAT_DATA_DIR"),
confp_get(Conf, "BLAT_TWO_BIT_DATA_FILENAME"),
string (targetsFile));
fasta_initFromPipe (string (buffer));
targetSeqs = fasta_readAllSequences (0);
fasta_deInit ();
if (arrayMax (targetSeqs) != 2) {
die ("Expected only two target sequences");
}
stringCreateClear (sequence,100);
for (i = 0; i < arrayMax (targetSeqs); i++) {
currSeq = arrp (targetSeqs,i,Seq);
stringAppendf (sequence,"%s",currSeq->sequence);
hlr_free (currSeq->name);
hlr_free (currSeq->sequence);
}
arrayDestroy (targetSeqs);
stringPrintf (buffer,"rm -rf %s",string (targetsFile));
hlr_system (string (buffer),0);
stringDestroy (targetsFile);
stringDestroy (buffer);
return string (sequence);
}
/**
* writes circos configuration file
*/
int write_circosConf (char* prefix,
Locus locus,
Array regions,
Chrdata_t *chromosomes,
SVCfg_t *settings)
{
float rpos = 0.99;
FILE *fp;
Stringa buffer = stringCreate(50);
int scale = getScale (regions);
stringPrintf (buffer, "%s/test/circos_%s_%s_%d_%d.conf",
confp_get(Conf, "WEB_CIRCOS_DIR"),
prefix,
locus.chromosome,
locus.start,
locus.end);
if (!(fp = fopen (string (buffer), "w"))) {
die ("Unable to open target file");
return -1;
};
printf ("<h2>%i</h2>", scale);
// write conf file
conf_printHeader (fp,
confp_get(Conf, "WEB_CIRCOS_DIR"),
confp_get(Conf, "WEB_DATA_DIR"),
confp_get(Conf, "WEB_SDATA_DIR"),
chromosomes,
prefix,
locus,
scale);
conf_printUnits (fp, regions, chromosomes, scale);
if (scale <= 10000) {
conf_printDataTracks (fp,
prefix,
locus,
confp_get(Conf, "WEB_SDATA_DIR"),
confp_get(Conf, "WEB_DATA_DIR"),
&rpos,
regions,
chromosomes,
settings);
}
conf_printLinks (fp, confp_get(Conf, "WEB_DATA_DIR"), &rpos, prefix, locus, settings->readlim);
conf_printFooter (fp);
fclose (fp);
stringDestroy (buffer);
return 0;
}
void stDestroy(symtable *st)
{
if (listSize(st->blockList) != 1)
fprintf(stderr, "Symbol table block leak!\n");
block *b = (block *) listRemoveBack(st->blockList);
stringDestroy(b->name);
bstDestroy(b->symbols, bstDelSymbol); //Cleanup memory for symbols in bst
free(b);
listDestroy(st->blockList);
free(st);
st = NULL;
}
void rdbu_initLoginInfo (char *filename) {
/**
Looks in the following places for user/password\@database
info needed for logging into database in this order
1. a file named .login_db in the user's home directory
2. a file named .login_db in the current directory
3. a file named 'filename' (optional input parameter)
4. arguments 'dbuser', 'dbpassword', 'dbname' present the command line
Syntax of the files:
- only the first line is read
- this line must have the form user/password\@dbname
- each of these fields can be '-'; in this case the value
of this field is not changed. E.g. if $HOME/.login_db contains
-/-\@testdb
and 'filename' contains
scott/-@-
and on the command line there is an argument
-dbpassword tiger
then username will be scott, password will be tiger and
database will be testdb
optional: if arg_init() was called before then the command line will be
considered; else the command line is ignored
Postcondition: rdbu_user() etc can be called
*/
Stringa fn;
char *home = getenv ("HOME");
if (user != NULL)
die ("rdbu_getLoginInfo() twice");
user = stringCreate(10);
password = stringCreate(10);
database = stringCreate(10);
if (home != NULL) {
fn = stringCreate (100);
stringPrintf (fn,"%s/.login_db",home);
readUserInfoFile (string (fn));
stringDestroy (fn);
}
readUserInfoFile (".login_db");
if (filename != NULL && *filename != '\0')
readUserInfoFile (filename);
if (arg_isInit ()) {
if (arg_present ("dbuser"))
stringCpy (user,arg_get ("dbuser"));
if (arg_present ("dbpassword"))
stringCpy (password,arg_get ("dbpassword"));
if (arg_present ("dbname"))
stringCpy (database,arg_get ("dbname"));
}
}
int main (int argc, char *argv[])
{
int i,j,groupNumber;
MrfEntry *currEntry;
GffEntry *currGffEntry,*nextGffEntry;
Array gffEntries;
FILE *fp;
Stringa buffer;
short int paired;
if (argc != 2) {
usage ("%s <prefix>",argv[0]);
}
buffer = stringCreate (1000);
groupNumber = 0;
mrf_init ("-");
gffEntries = arrayCreate (100000,GffEntry);
while (currEntry = mrf_nextEntry ()) {
processRead (gffEntries, currEntry, &groupNumber);
}
mrf_deInit ();
arraySort (gffEntries,(ARRAYORDERF)sortGffEntriesByTargetNameAndGroupNumber);
i = 0;
while (i < arrayMax (gffEntries)) {
currGffEntry = arrp (gffEntries,i,GffEntry);
stringPrintf (buffer,"%s_%s.gff",argv[1],currGffEntry->targetName);
fp = fopen (string (buffer),"w");
if (fp == NULL) {
die ("Unable to open file: %s",string (buffer));
}
fprintf (fp,"browser hide all\n");
fprintf (fp,"track name=\"%s_%s\" visibility=2\n",argv[1],currGffEntry->targetName);
fprintf (fp,"%s\n",currGffEntry->line);
j = i + 1;
while (j < arrayMax (gffEntries)) {
nextGffEntry = arrp (gffEntries,j,GffEntry);
if (!strEqual (currGffEntry->targetName,nextGffEntry->targetName)) {
break;
}
fprintf (fp,"%s\n",nextGffEntry->line);
j++;
}
i = j;
fclose (fp);
}
stringDestroy (buffer);
return 0;
}
int generateSamEntry ( Texta tokens, SamEntry *currSamE,
int* hasSeqs,
int* hasQual)
{
int j;
currSamE->qname = strdup (textItem (tokens, 0));
currSamE->flags = atoi (textItem (tokens, 1));
currSamE->rname = strdup (textItem (tokens, 2));
currSamE->pos = atoi (textItem (tokens, 3));
currSamE->mapq = atoi (textItem (tokens, 4));
currSamE->cigar = strdup (textItem (tokens, 5));
currSamE->mrnm = strdup (textItem (tokens, 6));
currSamE->mpos = atoi (textItem (tokens, 7));
currSamE->isize = atoi (textItem (tokens, 8));
currSamE->seq = NULL;
currSamE->qual = NULL;
currSamE->tags = NULL;
// Skip if unmapped or fails platform/vendor checks
if (currSamE->flags & S_QUERY_UNMAPPED ||
currSamE->flags & S_MATE_UNMAPPED ||
currSamE->flags & S_FAILS_CHECKS)
return 0;
// Get tokens
if (arrayMax (tokens) > 11) {
Stringa tags = stringCreate (10);
for (j = 11; j < arrayMax (tokens); j++) {
if (j > 11)
stringAppendf (tags, "\t");
stringAppendf (tags, "%s", textItem (tokens, j));
}
currSamE->tags = strdup (string(tags));
stringDestroy (tags);
}
if (strcmp (textItem (tokens, 9), "*") != 0) {
*hasSeqs = 1;
currSamE->seq = strdup (textItem (tokens, 9));
}
if (strcmp (textItem (tokens, 10), "*") != 0) {
*hasQual = 1;
currSamE->qual = strdup (textItem (tokens, 10));
}
return 1;
}
int main (int argc, char *argv[])
{
GfrEntry *currGE;
Array kgTreeFams;
Stringa buffer;
int count;
int countRemoved;
config *conf;
if ((conf = confp_open(getenv("FUSIONSEQ_CONFPATH"))) == NULL)
return EXIT_FAILURE;
buffer = stringCreate (100);
stringPrintf (buffer,"%s/%s",
confp_get(conf, "ANNOTATION_DIR"),
confp_get(conf, "KNOWN_GENE_TREE_FAM_FILENAME"));
kgTreeFams = util_readKnownGeneTreeFams (string (buffer));
arraySort (kgTreeFams,(ARRAYORDERF)sortKgTreeFamsByTranscriptName);
stringDestroy (buffer);
count = 0;
countRemoved = 0;
gfr_init ("-");
puts (gfr_writeHeader ());
while (currGE = gfr_nextEntry ()){
if (isHomologous (kgTreeFams,currGE->nameTranscript1,currGE->nameTranscript2)) {
countRemoved++;
continue;
}
puts (gfr_writeGfrEntry (currGE));
count++;
}
gfr_deInit ();
warn ("%s_numRemoved: %d",argv[0],countRemoved);
warn ("%s_numGfrEntries: %d",argv[0],count);
confp_close(conf);
return EXIT_SUCCESS;
}
int main (int argc, char *argv[])
{
GfrEntry *currGE;
GfrInterRead *currGIR;
int i;
Stringa buffer;
FILE *fp1,*fp2;
int count;
count = 0;
buffer = stringCreate (100);
gfr_init ("-");
puts (gfr_writeHeader ());
while (currGE = gfr_nextEntry ()) {
stringPrintf (buffer,"%s_1.bed",currGE->id);
fp1 = fopen (string (buffer),"w");
stringPrintf (buffer,"%s_2.bed",currGE->id);
fp2 = fopen (string (buffer),"w");
if (fp1 == NULL || fp2 == NULL) {
die ("Unable to open BED files");
}
fprintf (fp1,"browser full knownGene\n");
fprintf (fp1,"track name=\"Inter paird-ends: %s_1\" visibility=2\n",currGE->id);
fprintf (fp2,"browser full knownGene\n");
fprintf (fp2,"track name=\"Inter paird-ends: %s_2\" visibility=2\n",currGE->id);
for (i = 0; i < arrayMax (currGE->interReads); i++) {
currGIR = arrp (currGE->interReads,i,GfrInterRead);
fprintf (fp1,"%s\t%d\t%d\n",currGE->chromosomeTranscript1,currGIR->readStart1,currGIR->readEnd1);
fprintf (fp2,"%s\t%d\t%d\n",currGE->chromosomeTranscript2,currGIR->readStart2,currGIR->readEnd2);
}
fclose (fp1);
fclose (fp2);
puts (gfr_writeGfrEntry (currGE));
count++;
}
gfr_deInit ();
stringDestroy (buffer);
warn ("%s_numGfrEntries: %d",argv[0],count);
return 0;
}
void conf_printUnits (FILE *fp, Array regions, Chrdata_t *chromosomes, int scale)
{
int i;
SRegion_t *tmp;
char *chrname;
int sflag = 1;
Stringa chrshow = stringCreate (50);
tmp = arrayp (regions, 0, SRegion_t);
if (tmp->chromosome == 0) {
fprintf (fp, "chromosomes_display_default = yes\n");
fprintf (fp, "chromosomes_units = 1000000\n");
}
else {
stringPrintf (chrshow, "chromosomes = ");
fprintf (fp, "chromosomes_units = %i\n", scale);
fprintf (fp, "chromosomes_display_default = no\n");
for (i = 0; i < arrayMax (regions); i++) {
tmp = arrayp (regions, i, SRegion_t);
if (chromosomes[tmp->chromosome].show && tmp->show) {
chrname = getHchrname (tmp->chromosome);
if (sflag == 0) {
stringAppendf (chrshow, ";");
}
stringAppendf (chrshow, "%s:%i-%i", chrname, tmp->start/scale, tmp->end/scale);
sflag = 0;
free (chrname);
}
}
fprintf (fp, "%s\n", string (chrshow));
}
stringDestroy (chrshow);
}
symtable *stCreate()
{
symtable *st = (symtable *) malloc(sizeof(symtable));
st->blockList = listCreate();
//Add block0
block *b = (block *) malloc(sizeof(block));
b->name = stringCreate();
stringAppendCharArray(b->name, "block0", 6*sizeof(char));
b->symbols = bstCreate(bstCompareSymbol);
b->nextLoc = 0;
listAddBack(st->blockList, b);
//Initialize block0 with PJ's builtin procedures and input/output files
for (unsigned int i = 0; i < builtin_num; i++)
{
const char *builtinName = pjbuiltinString(i);
string *name = stringCreate();
stringAppendCharArray(name, builtinName, strlen(builtinName));
symbol *sym = symbolCreate(name);
stringDestroy(name);
symbolSetBuiltin(sym, i);
stAddSymbol(st, sym);
}
return st;
}
SEXP c_read_biokit_exprs (SEXP filename) {
LineStream ls;
char* line;
const int MAND_NCOL=7; // the first column is the row name, and column 2-7 are mandatory
int add_ncol=0;
Texta it;
Texta rnames=textCreate(128);
Array mrpkms=arrayCreate(128, double);
Array mreads=arrayCreate(128, int);
Array srpkms=arrayCreate(128, double);
Array sreads=arrayCreate(128, int);
Array mprop=arrayCreate(128, double);
Array allmap = arrayCreate(128, int);
Array annos=arrayCreate(128, Texta);
Texta anno=NULL; // must have a NULL assigned; otherwise textCreateClear leads to memory error
Stringa str=stringCreate(8);
SEXP R_rnames, R_mrpkms, R_mreads, R_srpkms, R_sreads, R_mprop, R_allmap, R_res;
SEXP R_colnames, R_class;
int nprot=0;
int i=0;
int j=0;
int nrow=0;
const char* fn=CHAR(STRING_ELT(filename, 0));
ls = ls_createFromFile(strdup(fn));
ls_nextLine(ls); // skip the first header line
while(line = ls_nextLine(ls)) {
it = textFieldtokP(line, "\t");
if(arrayMax(it)<MAND_NCOL)
error("Input file must contain no less than %d columns", MAND_NCOL);
textAdd(rnames, textItem(it, 0));
array(mrpkms, arrayMax(mrpkms), double)=atof(textItem(it, 1));
array(mreads, arrayMax(mreads), int)=atoi(textItem(it, 2));
array(srpkms, arrayMax(srpkms), double)=atof(textItem(it, 3));
array(sreads, arrayMax(sreads), int)=atoi(textItem(it, 4));
array(mprop, arrayMax(mprop), double)=atof(textItem(it, 5));
array(allmap, arrayMax(allmap), int)=atoi(textItem(it, 6));
add_ncol = max(arrayMax(it)-MAND_NCOL, add_ncol);
textCreateClear(anno, arrayMax(it)-MAND_NCOL);
for(i=MAND_NCOL; i<arrayMax(it); ++i) {
textAdd(anno, textItem(it, i));
}
array(annos, arrayMax(annos), Texta)=textClone(anno);
nrow++;
}
R_rnames=PROTECT(allocVector(STRSXP, nrow)); nprot++;
R_mrpkms=PROTECT(allocVector(REALSXP, nrow)); nprot++;
R_mreads=PROTECT(allocVector(INTSXP, nrow)); nprot++;
R_srpkms=PROTECT(allocVector(REALSXP, nrow)); nprot++;
R_sreads=PROTECT(allocVector(INTSXP, nrow)); nprot++;
R_mprop=PROTECT(allocVector(REALSXP, nrow)); nprot++;
R_allmap=PROTECT(allocVector(INTSXP, nrow)); nprot++;
for(i=0; i<nrow; ++i) {
SET_STRING_ELT(R_rnames, i, mkChar(textItem(rnames, i)));
REAL(R_mrpkms)[i]=arru(mrpkms, i, double);
INTEGER(R_mreads)[i]=arru(mreads, i, int);
REAL(R_srpkms)[i]=arru(srpkms, i, double);
INTEGER(R_sreads)[i]=arru(sreads, i, int);
REAL(R_mprop)[i]=arru(mprop, i, double);
INTEGER(R_allmap)[i]=arru(allmap, i, int);
}
R_res=PROTECT(allocVector(VECSXP, MAND_NCOL+add_ncol-1)); nprot++;
SET_VECTOR_ELT(R_res, 0, R_mrpkms);
SET_VECTOR_ELT(R_res, 1, R_mreads);
SET_VECTOR_ELT(R_res, 2, R_srpkms);
SET_VECTOR_ELT(R_res, 3, R_sreads);
SET_VECTOR_ELT(R_res, 4, R_mprop);
SET_VECTOR_ELT(R_res, 5, R_allmap);
for(i=0; i<add_ncol; ++i) {
SEXP R_anno=NULL;
R_anno=PROTECT(allocVector(STRSXP, nrow));
for(j=0; j<nrow; ++j) {
anno=array(annos, j, Texta);
if(arrayMax(anno)>i) {
SET_STRING_ELT(R_anno, j, mkChar(textItem(anno, i)));
} else {
SET_STRING_ELT(R_anno, j, R_NaString);
}
}
SET_VECTOR_ELT(R_res, i+MAND_NCOL-1, R_anno); // -1 because the first column is row name
UNPROTECT(1);
}
PROTECT(R_colnames=allocVector(STRSXP, MAND_NCOL+add_ncol-1)); nprot++;
PROTECT(R_class=allocVector(STRSXP, 1)); nprot++;
SET_STRING_ELT(R_colnames, 0, mkChar("RPKM_MultiMap"));
SET_STRING_ELT(R_colnames, 1, mkChar("ReadCount_MultiMap"));
SET_STRING_ELT(R_colnames, 2, mkChar("RPKM_UniqMap"));
SET_STRING_ELT(R_colnames, 3, mkChar("ReadCount_UniqMap"));
SET_STRING_ELT(R_colnames, 4, mkChar("MultiProp"));
SET_STRING_ELT(R_colnames, 5, mkChar("AllMappingReads"));
for(i=0; i<add_ncol; ++i) {
stringPrintf(str, "Annotation%d", i+1);
SET_STRING_ELT(R_colnames, i+MAND_NCOL-1,
mkChar(string(str)));
}
SET_STRING_ELT(R_class, 0, mkChar("data.frame"));
setAttrib(R_res, install("names"), R_colnames);
setAttrib(R_res, install("row.names"), R_rnames);
setAttrib(R_res, install("class"), R_class);
for(i=0; i<nrow; ++i) {
textDestroy(array(annos, i, Texta));
}
arrayDestroy(annos);
arrayDestroy(rnames);
arrayDestroy(mrpkms);
arrayDestroy(mreads);
arrayDestroy(srpkms);
arrayDestroy(sreads);
arrayDestroy(mprop);
arrayDestroy(allmap);
stringDestroy(str);
ls_destroy(ls);
UNPROTECT(nprot);
return(R_res);
}
int main (int argc, char *argv[])
{
Array kgXrefs;
Stringa buffer;
LineStream ls;
int count=0;
char* geneSymbolTranscript;
char* descriptionTranscript;
char* line;
char* exonID = NULL;
config *conf;
if ((conf = confp_open(getenv("FUSIONSEQ_CONFPATH"))) == NULL)
return EXIT_FAILURE;
buffer = stringCreate (100);
stringPrintf (buffer,"%s/%s",
confp_get(conf, "ANNOTATION_DIR"),
confp_get(conf, "KNOWN_GENE_XREF_FILENAME"));
kgXrefs = util_readKnownGeneXrefs (string (buffer));
arraySort (kgXrefs,(ARRAYORDERF)sortKgXrefsByTranscriptName);
stringDestroy (buffer);
// gfr_init ("-");
ls = ls_createFromFile("-");
while (line = ls_nextLine(ls)) {
char *lineP = hlr_strdup(line);
WordIter w = wordIterCreate( line, "\t", 0);
char *nameTranscript = wordNext( w );
char *p = rindex(nameTranscript, '_');
if (p) {
exonID = hlr_strdup( p+1 );
*p='\0';
}
transcript2geneSymbolAndGeneDescription(kgXrefs,
nameTranscript,
&geneSymbolTranscript,
&descriptionTranscript);
if (exonID) {
printf("%s_%s\t%s\t%s\t%s",
nameTranscript,
exonID,
geneSymbolTranscript,
exonID,
descriptionTranscript);
hlr_free(exonID);
} else {
printf("%s\t%s\t1\t%s",
nameTranscript,
geneSymbolTranscript,
descriptionTranscript);
}
printf("%s\n", lineP+strlen(nameTranscript));
count++;
hlr_free(lineP);
wordIterDestroy(w);
}
ls_destroy (ls);
warn ("%s_numGfrEntries: %d",argv[0],count);
confp_close(conf);
return EXIT_SUCCESS;
}
void incl_getExonHlightFile (FILE *fp, Array regions, char *sdata_dir)
{
LineStream src;
FILE *out;
char *line;
Texta entry;
int i, astart, aend;
Stringa buffer = stringCreate (50);
stringPrintf (buffer, "%s/tmp/exons.hlight_s.txt", sdata_dir);
if (!(out = fopen (string (buffer), "w"))) {
fprintf (stderr, "Cannot open exons.hlight_s.txt\n");
return;
}
SRegion_t *tmp;
tmp = arrayp (regions, 0, SRegion_t);
if (tmp->chromosome == 0) {
fprintf (fp, "file = %s/exons.hlight.txt\n", sdata_dir);
}
else {
for (i = 0; i < arrayMax (regions); i++) {
tmp = arrayp (regions, i, SRegion_t);
if (tmp->chromosome == 23) {
stringPrintf (buffer, "%s/X/exons.hlight.txt", sdata_dir);
}
else if (tmp->chromosome == 24) {
stringPrintf (buffer, "%s/Y/exons.hlight.txt", sdata_dir);
}
else {
stringPrintf (buffer, "%s/%i/exons.hlight.txt", sdata_dir, tmp->chromosome);
}
if ((src = ls_createFromFile (string (buffer))) == NULL) {
fprintf (stderr, "Cannot open exons.hlight.txt\n");
return;
}
while ((line = ls_nextLine (src)) != NULL) {
entry = textFieldtokP (line, " ");
astart = atoi (textItem (entry, 1));
aend = atoi (textItem (entry, 2));
if ((astart >= tmp->start && astart <= tmp->end) ||
(aend >= tmp->start && aend <= tmp->end)) {
fprintf (out, "%s\n", line);
}
textDestroy (entry);
}
}
fprintf (fp, "file = %s/tmp/exons.hlight_s.txt\n", sdata_dir);
}
stringDestroy (buffer);
ls_destroy (src);
fclose (out);
}
int main (int argc, char *argv[])
{
GfrEntry *currGE;
int count;
int countRemoved;
int mitochondrialCount;
unsigned int minReadSize;
int i;
Stringa cmd;
BlatQuery *blQ=NULL;
config *conf = NULL; /**< Pointer to configuration file .fusionseqrc */
if ((conf = confp_open(getenv("FUSIONSEQ_CONFPATH"))) == NULL) {
die("%s:\tCannot find .fusionseqrc: %s", argv[0], getenv("FUSIONSEQ_CONFPATH"));
return EXIT_FAILURE;
}
if( confp_get( conf,"MAX_OVERLAP_ALLOWED")==NULL ) {
die("%s:\tCannot find MAX_OVERLAP_ALLOWED in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf,"MAX_FRACTION_HOMOLOGOUS")==NULL ) {
die("%s:\tCannot find MAX_FRACTION_HOMOLOGOUS in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "MITOCHONDRIAL_DIR")==NULL ) {
die("%s:\tCannot find MITOCHONDRIAL_DIR in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "MITOCHONDRIAL_FILENAME")==NULL ) {
die("%s:\tCannot find MITOCHONDRIAL_FILENAME in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "TMP_DIR")==NULL ) {
die("%s:\tCannot find TMP_DIR in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "BLAT_GFSERVER")==NULL ) {
die("%s:\tCannot find BLAT_GFSERVER in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "BLAT_GFCLIENT")==NULL ) {
die("%s:\tCannot find BLAT_GFCLIENT in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "BLAT_GFSERVER_HOST")==NULL ) {
die("%s:\tCannot find BLAT_GFSERVER_HOST in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}if( confp_get( conf, "BLAT_GFSERVER_PORT")==NULL ) {
die("%s:\tCannot find BLAT_GFSERVER_PORT in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
count = 0;
countRemoved = 0;
cmd = stringCreate (100);
// initializing the gfServers
stringPrintf( cmd, "%s status %s %d &> /dev/null", confp_get( conf, "BLAT_GFSERVER"), confp_get( conf, "BLAT_GFSERVER_HOST"), atoi(confp_get( conf, "BLAT_GFSERVER_PORT")) + 2);
int ret = hlr_system( string(cmd), 1 );
if( ret != 0 ) { // not initialized
stringPrintf( cmd , "%s -repMatch=100000 -tileSize=12 -canStop -log=%s/gfServer_mitochondrial.log start %s %d %s/%s &", confp_get( conf, "BLAT_GFSERVER"), confp_get( conf, "TMP_DIR"), confp_get( conf, "BLAT_GFSERVER_HOST"), atoi(confp_get( conf, "BLAT_GFSERVER_PORT")) + 2, confp_get( conf, "MITOCHONDRIAL_DIR"), confp_get( conf,"MITOCHONDRIAL_FILENAME"));
hlr_system( string( cmd ), 0 );
long int startTime = time(0);
stringPrintf( cmd , "%s status %s %d &> /dev/null", confp_get( conf, "BLAT_GFSERVER"), confp_get( conf, "BLAT_GFSERVER_HOST"), atoi(confp_get( conf, "BLAT_GFSERVER_PORT")) + 2);
while( hlr_system( string(cmd), 1) && (time(0)-startTime)<600 ) ;
if( hlr_system( string(cmd), 1 ) != 0 ) {
die("gfServer for %s/%s not initialized: %s %s %s", confp_get( conf, "MITOCHONDRIAL_DIR"), confp_get( conf, "MITOCHONDRIAL_FILENAME"), confp_get( conf, "BLAT_GFSERVER"), confp_get( conf, "BLAT_GFSERVER_HOST"), confp_get( conf, "BLAT_GFSERVER_PORT"));
return EXIT_FAILURE;
}
}
gfr_init ("-");
puts (gfr_writeHeader ());
while (currGE = gfr_nextEntry ()) {
if (strEqual(currGE->chromosomeTranscript1, "chrM") ||
strEqual(currGE->chromosomeTranscript2, "chrM")) {
countRemoved++;
continue;
} else {
mitochondrialCount = 0;
minReadSize=1000;
writeFasta( currGE, &minReadSize, confp_get( conf, "TMP_DIR") ); // in util.c
stringPrintf(cmd, "cd %s;%s %s %d / -t=dna -q=dna -minScore=%d -out=psl %s_reads.fa %s.mito.psl &>/dev/null", confp_get( conf, "TMP_DIR"), confp_get( conf, "BLAT_GFCLIENT"), confp_get( conf, "BLAT_GFSERVER_HOST"), atoi(confp_get( conf, "BLAT_GFSERVER_PORT")) + 2, minReadSize - 5 > 20 ? minReadSize - 5 : 20 , currGE->id, currGE->id);
int attempts=0;
ret = hlr_system( string(cmd), 1 );
while( hlr_system( string(cmd), 1 ) && attempts<5000 ) attempts++;
if( attempts == 5000 ) {
die("Cannot map the reads %s", string( cmd ));
return EXIT_FAILURE;
}
// reading the results of blast from File
stringPrintf(cmd, "%s/%s.mito.psl", confp_get( conf, "TMP_DIR"), currGE->id);
blatParser_initFromFile( string(cmd) );
while( blQ = blatParser_nextQuery() ) {
//warn("iter %d\tquery %s", iter, blQ->qName );iter++;
int nucleotideOverlap = getNucleotideOverlap ( blQ );
if (nucleotideOverlap > (((double) minReadSize) * strtod(confp_get( conf, "MAX_OVERLAP_ALLOWED"), NULL))) {
char* value = strchr( blQ->qName,'/' );
if( value ) *value = '\0'; else die("Not a valid index in the blat query name:\t%s", blQ->qName );
int indexOfInter = atoi( blQ->qName ); // the following three lines should removed the read if writing the GFR entry
GfrInterRead *currGIR = arrp( currGE->interReads, indexOfInter, GfrInterRead );
currGIR->flag = 1;
mitochondrialCount++;
}
}
blatParser_deInit();
if ( ( (double) mitochondrialCount / (double) ( arrayMax(currGE->readsTranscript1) + arrayMax(currGE->readsTranscript2) ) ) <= strtod(confp_get( conf, "MAX_FRACTION_HOMOLOGOUS"), NULL)) {
if( mitochondrialCount > 0 ) updateStats( currGE );
// writing the gfrEntry
puts (gfr_writeGfrEntry (currGE));
count++;
} else {
countRemoved++;
}
// removing temporary files
stringPrintf (cmd,"rm -rf %s/%s_reads.fa %s/%s.mito.psl", confp_get( conf, "TMP_DIR"), currGE->id, confp_get( conf, "TMP_DIR"), currGE->id );
hlr_system( string(cmd) , 1);
}
}
gfr_deInit ();
stringDestroy( cmd );
warn ("%s_numRemoved: %d",argv[0],countRemoved);
warn ("%s_numGfrEntries: %d",argv[0],count);
confp_close(conf);
return 0;
}
int main (int argc, char *argv[])
{
GfrEntry *currGE;
int i,j,k,l, h,index;
Stringa buffer,cmd,fnSequencesToAlign;
FILE *fp;
FILE *fp1;
FILE *fp2;
FILE *freads1;
FILE *freads2;
Array gfrEntries;
BowtieQuery *currBQ,testBQ;
BowtieEntry *currBE;
Texta seqNames;
int readSize1, readSize2, minReadSize;
Array bowtieQueries;
char transcriptNumber;
int isHomologous,homologousCount;
int count;
int countRemoved;
unsigned short int tooMany;
BlatQuery *blQ;
config *conf;
if ((conf = confp_open(getenv("FUSIONSEQ_CONFPATH"))) == NULL) {
die("%s:\tCannot find .fusionseqrc", argv[0]);
return EXIT_FAILURE;
}
if ( (confp_get( conf, "BLAT_TWO_BIT_TO_FA")) == NULL) {
die("%s:\tCannot find BLAT_TWO_BIT_TO_FA in the configuration file: %s", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if ( (confp_get( conf,"BLAT_DATA_DIR")) == NULL) {
die("%s:\tCannot find BLAT_DATA_DIR in the configuration file: %sc", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "TMP_DIR")==NULL ) {
die("%s:\tCannot find TMP_DIR in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "BLAT_GFSERVER")==NULL ) {
die("%s:\tCannot find BLAT_GFSERVER in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "BLAT_GFCLIENT")==NULL ) {
die("%s:\tCannot find BLAT_GFCLIENT in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "BLAT_GFSERVER_HOST")==NULL ) {
die("%s:\tCannot find BLAT_GFSERVER_HOST in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}if( confp_get( conf, "BLAT_GFSERVER_PORT")==NULL ) {
die("%s:\tCannot find BLAT_GFSERVER_PORT in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "PSEUDOGENE_DIR")==NULL ) {
die("%s:\tCannot find PSEUDOGENE_DIR in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( conf, "PSEUDOGENE_FILENAME")==NULL ) {
die("%s:\tCannot find PSEUDOGENE_FILENAME in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
cmd = stringCreate (100);
// initializing the gfServers
stringPrintf( cmd, "%s status %s %s &> /dev/null", confp_get( conf, "BLAT_GFSERVER"), confp_get( conf, "BLAT_GFSERVER_HOST"), confp_get( conf, "BLAT_GFSERVER_PORT") );
int ret = hlr_system( string(cmd), 1 );
if( ret != 0 ) { // not initialized
stringPrintf( cmd , "%s -repMatch=100000 -tileSize=12 -canStop -log=%s/gfServer_genome.log start %s %s %s/%s &", confp_get( conf, "BLAT_GFSERVER"), confp_get(conf, "TMP_DIR"),confp_get( conf, "BLAT_GFSERVER_HOST"), confp_get( conf, "BLAT_GFSERVER_PORT"), confp_get(conf, "BLAT_DATA_DIR"), confp_get(conf, "BLAT_TWO_BIT_DATA_FILENAME"));
hlr_system( string( cmd ), 0 );
long int startTime = time(0);
stringPrintf( cmd , "%s status %s %s &2> /dev/null", confp_get( conf, "BLAT_GFSERVER"), confp_get( conf, "BLAT_GFSERVER_HOST"), confp_get( conf, "BLAT_GFSERVER_PORT"));
while( hlr_system( string(cmd), 1) && (time(0)-startTime)<600 ) ;
if( hlr_system( string(cmd), 1 ) != 0 ) {
die("gfServer for %s/%s not initialized: %s %s %s", confp_get(conf, "BLAT_DATA_DIR"), confp_get(conf, "BLAT_TWO_BIT_DATA_FILENAME"), confp_get( conf, "BLAT_GFSERVER"), confp_get( conf, "BLAT_GFSERVER_HOST"), confp_get( conf, "BLAT_GFSERVER_PORT"));
return EXIT_FAILURE;
}
}
// end initialization
gfr_init ("-");
gfrEntries = gfr_parse ();
if (arrayMax (gfrEntries) == 0){
puts (gfr_writeHeader ());
gfr_deInit ();
return 0;
}
seqNames = textCreate (10000);
buffer = stringCreate (100);
fnSequencesToAlign = stringCreate (100);
count = 0;
countRemoved = 0;
stringPrintf( buffer, "%s/%s", confp_get( conf, "PSEUDOGENE_DIR"), confp_get( conf, "PSEUDOGENE_FILENAME") );
intervalFind_addIntervalsToSearchSpace (string(buffer),0);
puts (gfr_writeHeader ());
for (i = 0; i < arrayMax (gfrEntries); i++) {
currGE = arrp (gfrEntries,i,GfrEntry);
homologousCount = 0;
minReadSize=10000;
// creating two fasta files with the two genes
stringPrintf( cmd, "%s %s/%s -seq=%s -start=%d -end=%d %s/%s_transcript1.fa", confp_get(conf, "BLAT_TWO_BIT_TO_FA") , confp_get(conf, "BLAT_DATA_DIR"), confp_get(conf, "BLAT_TWO_BIT_DATA_FILENAME"), currGE->chromosomeTranscript1, currGE->startTranscript1, currGE->endTranscript1, confp_get(conf, "TMP_DIR"), currGE->id);
hlr_system( string(cmd) , 0);
stringPrintf( cmd, "%s %s/%s -seq=%s -start=%d -end=%d %s/%s_transcript2.fa", confp_get(conf, "BLAT_TWO_BIT_TO_FA"), confp_get(conf, "BLAT_DATA_DIR"), confp_get(conf, "BLAT_TWO_BIT_DATA_FILENAME"), currGE->chromosomeTranscript2, currGE->startTranscript2, currGE->endTranscript2, confp_get(conf, "TMP_DIR"), currGE->id);
hlr_system( string(cmd) , 0);
Stringa fa1 = stringCreate( 100 );
Stringa fa2 = stringCreate( 100 );
// creating the two fasta files with the reads
stringPrintf( fa1, "%s/%s_reads1.fa", confp_get(conf, "TMP_DIR"), currGE->id);
if (!(freads1 = fopen ( string(fa1) ,"w"))) {
die ("Unable to open file: %s",string (fa1));
}
// writing the reads of the first end into file
for (l = 0; l < arrayMax (currGE->readsTranscript1); l++) {
char* currRead1 = hlr_strdup( textItem (currGE->readsTranscript1,l)); // read1
readSize1 = strlen( currRead1 );
if( readSize1 == 0 ) die("Read size cannot be zero: read1[ %s ]", currRead1);
if( readSize1 < minReadSize ) minReadSize = readSize1;
fprintf( freads1, ">%d\n%s\n", l, currRead1 );
hlr_free( currRead1 );
}
fclose( freads1 );
stringPrintf( fa2, "%s/%s_reads2.fa", confp_get(conf, "TMP_DIR"), currGE->id);
if (!(freads2 = fopen ( string(fa2) ,"w"))) {
die ("Unable to open file: %s",string (fa2));
}
// writing the reads of the second end into file
for (l = 0; l < arrayMax (currGE->readsTranscript2); l++) {
char* currRead2 = hlr_strdup( textItem (currGE->readsTranscript2,l)); // read2
readSize2 = strlen( currRead2 );
if( readSize2 == 0 ) die("Read size cannot be zero: read2[ %s ]", currRead2);
if( readSize2 < minReadSize ) minReadSize = readSize2;
fprintf( freads2, ">%d\n%s\n", l, currRead2 );
hlr_free( currRead2 );
}
fclose( freads2 );
// collapse the reads 2 ## requires the FASTX package
stringPrintf( cmd, "%s -i %s/%s_reads2.fa -o %s/%s_reads2.collapsed.fa", confp_get(conf, "FASTX_COLLAPSER"), confp_get(conf, "TMP_DIR"), currGE->id, confp_get(conf, "TMP_DIR"), currGE->id );
hlr_system (string (cmd),0);
//blat of reads2 against the first transcript
stringPrintf( cmd, "%s -t=dna -out=psl -fine -tileSize=15 %s/%s_transcript1.fa %s/%s_reads2.collapsed.fa stdout",confp_get(conf, "BLAT_BLAT"), confp_get(conf, "TMP_DIR"), currGE->id, confp_get(conf, "TMP_DIR"), currGE->id );
// reading the results of blast from Pipe
blatParser_initFromPipe( string(cmd) );
while( blQ = blatParser_nextQuery() ) {
int nucleotideOverlap = getNucleotideOverlap ( blQ );
if ( nucleotideOverlap > ( ((double)readSize2)* atof(confp_get(conf,"MAX_OVERLAP_ALLOWED"))) ) {
char* value = strchr(blQ->qName,'-');
homologousCount+=atoi(value+1);
}
}
blatParser_deInit();
// collapse the reads 1 ## requires the FASTX package on the path
stringPrintf( cmd, "%s -i %s/%s_reads1.fa -o %s/%s_reads1.collapsed.fa", confp_get(conf, "FASTX_COLLAPSER"), confp_get(conf, "TMP_DIR"), currGE->id, confp_get(conf, "TMP_DIR"), currGE->id );
hlr_system (string (cmd),0);
//blat of reads1 against the second transcript
stringPrintf( cmd, "%s -t=dna -out=psl -fine -tileSize=15 %s/%s_transcript2.fa %s/%s_reads1.collapsed.fa stdout",confp_get(conf, "BLAT_BLAT"), confp_get(conf, "TMP_DIR"), currGE->id, confp_get(conf, "TMP_DIR"), currGE->id );
blatParser_initFromPipe( string(cmd) );
while( blQ = blatParser_nextQuery() ) {
int nucleotideOverlap = getNucleotideOverlap ( blQ );
if ( nucleotideOverlap > ( ((double)readSize1)* atof(confp_get(conf,"MAX_OVERLAP_ALLOWED"))) ) {
char* value = strchr(blQ->qName,'-');
homologousCount+=atoi(value+1);
}
}
blatParser_deInit();
stringPrintf (cmd,"cd %s;rm -rf %s_reads?.fa %s_reads?.collapsed.fa %s_transcript?.fa", confp_get(conf, "TMP_DIR"), currGE->id,currGE->id,currGE->id);
hlr_system( string(cmd) , 0);
if (((double)homologousCount / (double)arrayMax(currGE->readsTranscript1)) <= atof(confp_get(conf, "MAX_FRACTION_HOMOLOGOUS")) ) {
homologousCount = 0;
// there is no homology between the two genes, but what about the rest of the genome
writeFasta( currGE, &minReadSize, confp_get(conf, "TMP_DIR") );
stringPrintf(cmd, "cd %s; %s %s %s / -t=dna -q=dna -minScore=%d -out=psl %s_reads.fa %s.smallhomology.psl &>/dev/null", confp_get(conf, "TMP_DIR"), confp_get( conf, "BLAT_GFCLIENT"), confp_get( conf, "BLAT_GFSERVER_HOST"), confp_get( conf, "BLAT_GFSERVER_PORT"), minReadSize - (int)(0.1 * minReadSize) > 20 ? minReadSize - (int) (0.1 * minReadSize) : 20 , currGE->id, currGE->id);
int attempts=0;
ret = hlr_system( string(cmd), 1 );
while( hlr_system( string(cmd), 1 ) && attempts<5000 ) attempts++;
if( attempts == 5000 ) {
die("Cannot map the reads %s", string( cmd ));
return EXIT_FAILURE;
}
// reading the results of blast from File
stringPrintf(cmd, "%s/%s.smallhomology.psl", confp_get( conf, "TMP_DIR"), currGE->id);
blatParser_initFromFile( string(cmd) );
tooMany = 1;
while( blQ = blatParser_nextQuery() ) {
tooMany = 0;
checkPseudogeneOverlap( blQ );
if( arrayMax( blQ->entries ) > 1 ) {
homologousCount+= arrayMax( blQ->entries ) - 1;
char* value = strchr( blQ->qName,'/' );
if( value ) *value = '\0'; else die("Not a valid index in the blat query name:\t%s", blQ->qName );
int indexOfInter = atoi( blQ->qName ); // the following three lines should removed the read if writing the GFR entry
GfrInterRead *currGIR = arrp( currGE->interReads, indexOfInter, GfrInterRead );
currGIR->flag = 1;
}
}
blatParser_deInit();
if ( tooMany == 1 || ( ( (double) homologousCount / (double) ( arrayMax(currGE->readsTranscript1) + arrayMax(currGE->readsTranscript2) ) ) > atof(confp_get(conf, "MAX_FRACTION_HOMOLOGOUS")) ) ) {
countRemoved++;
stringPrintf (cmd,"cd %s; rm -rf %s_reads*.fa %s_reads?.collapsed.fa %s_transcript?.fa %s.smallhomology.psl", confp_get(conf, "TMP_DIR"), currGE->id,currGE->id,currGE->id,currGE->id);
hlr_system( string(cmd), 1 );
continue;
}
// writing the gfrEntry, if everthing else didn't stop
if( homologousCount > 0 ) updateStats( currGE );
puts (gfr_writeGfrEntry (currGE));
count++;
// removing temporary files
stringPrintf (cmd,"cd %s;rm -rf %s_reads*.fa %s_reads?.collapsed.fa %s_transcript?.fa %s.smallhomology.psl", confp_get(conf, "TMP_DIR"), currGE->id,currGE->id,currGE->id,currGE->id);
hlr_system( string(cmd) , 1);
} else {
countRemoved++;
}
}
gfr_deInit ();
stringDestroy (fnSequencesToAlign);
stringDestroy (cmd);
stringDestroy (buffer);
warn ("%s_numRemoved: %d",argv[0],countRemoved);
warn ("%s_numGfrEntries: %d",argv[0],count);
confp_close(conf);
return EXIT_SUCCESS;
}
static void generateOutput (char* prefix, char* typeSelected, int minNum)
{
GfrEntry *currGE;
Stringa buffer;
char *pos;
puts ("<html>");
puts ("<head>");
puts ("<title>Results - Gene Fusions</title>");
html_printGenericStyleSheet (12);
puts ("</head>");
puts ("<body>");
if (prefix[0] == '\0') {
die ("Invalid prefix");
}
printf ("<h1>Results - %s</h1><br><br><br>",prefix);
buffer = stringCreate(50);
//Chromosome expression, if present
LineStream ls;
char* chrSignal=NULL;
stringPrintf(buffer, "ls -1 %s/BGRS/%s_chr*.bgr.gz 2> /dev/null",
confp_get(Conf, "WEB_DATA_DIR"),
prefix);
ls = ls_createFromPipe(string(buffer));
int countCol = 0;
puts ("Expression signal: ");
fflush(stdout);
while( chrSignal = ls_nextLine(ls)) {
char* chrTmp = stringBetween( prefix, ".bgr.gz", chrSignal );
chrTmp++;
printf ("[<a href=%s&hgt.customText=%s/BGRS/%s_%s.bgr.gz target='blank'>%s</a>] ",
htmlLinker_generateLinkToGenomeBrowserAtUCSC("hg18","vertebrate","human", chrTmp,
confp_get(Conf, "UCSC_GENOME_BROWSER_FLANKING_REGION"),
50000000 + confp_get(Conf, "UCSC_GENOME_BROWSER_FLANKING_REGION")),
confp_get(Conf, "WEB_DATA_LINK"),
prefix,
chrTmp,
chrTmp);
if (countCol > 10) {
puts( "<BR>" );
countCol=0;
}
countCol++;
}
if( countCol==0) puts( "No data available yet" );
ls_destroy(ls);
puts ("<br><br>");
puts ("For a definition of SPER, DASPER and RESPER see <a href=http://rnaseq.gersteinlab.org/fusionseq/>FusionSeq</a>");
puts ("<br><br>");
puts ("<br><table border=0 width=100% align=center cellpadding=10>");
puts ("<tr align=left>");
puts ("<th>SPER</th>");
puts ("<th>DASPER</th>");
puts ("<th>RESPER</th>");
puts ("<th>Number of inter paired-end reads</th>");
puts ("<th>Type</th>");
puts ("<th>Genomic coordinates</th>");
puts ("<th>Gene symbol</th>");
puts ("<th>Description</th>");
puts ("<th>Genomic coordinates</th>");
puts ("<th>Gene symbol</th>");
puts ("<th>Description</th>");
puts ("<th></th>");
puts ("</tr>");
fflush(stdout);
stringPrintf (buffer,"%s/%s.gfr", confp_get(Conf, "WEB_DATA_DIR"), prefix);
gfr_init (string (buffer));
int countElements = 0;
while (currGE = gfr_nextEntry ()) {
if (currGE->numInter < minNum) {
continue;
}
if (strEqual (typeSelected,"all") || strEqual (currGE->fusionType,typeSelected) ||
( strEqual(currGE->fusionType,"cis") && strEqual( typeSelected,"same") ) ||
( strEqual(currGE->fusionType,"read-through") && strEqual( typeSelected,"same") ) ) {
if (pos = strchr (currGE->descriptionTranscript1,'|')) {
*pos = '\0';
}
if (pos = strchr (currGE->descriptionTranscript2,'|')) {
*pos = '\0';
}
puts ("<tr>");
printf ("<td align=left>%1.3f</td>\n",currGE->SPER);
printf ("<td align=left>%1.3f</td>\n",currGE->DASPER);
printf ("<td align=left>%1.3f</td>\n",currGE->RESPER);
printf ("<td align=left>%d</td>\n",currGE->numInter);
printf ("<td align=left>%s</td>\n",currGE->fusionType);
printf ("<td align=left><a href=%s target=blank>%s:%d-%d</a></td>\n",
htmlLinker_generateLinkToGenomeBrowserAtUCSC ("hg18","vertebrate","human",
currGE->chromosomeTranscript1,
currGE->startTranscript1 - atoi(confp_get(Conf, "UCSC_GENOME_BROWSER_FLANKING_REGION")),
currGE->endTranscript1 + atoi(confp_get(Conf, "UCSC_GENOME_BROWSER_FLANKING_REGION"))),
currGE->chromosomeTranscript1,currGE->startTranscript1,currGE->endTranscript1);
printf ("<td align=left>%s</td>\n",processString (currGE->geneSymbolTranscript1));
printf ("<td align=left>%s</td>\n",currGE->descriptionTranscript1);
printf ("<td align=left><a href=%s target=blank>%s:%d-%d</a></td>\n",
htmlLinker_generateLinkToGenomeBrowserAtUCSC ("hg18","vertebrate","human",
currGE->chromosomeTranscript2,
currGE->startTranscript2 - atoi(confp_get(Conf, "UCSC_GENOME_BROWSER_FLANKING_REGION")),
currGE->endTranscript2 + atoi(confp_get(Conf, "UCSC_GENOME_BROWSER_FLANKING_REGION"))),
currGE->chromosomeTranscript2,currGE->startTranscript2,currGE->endTranscript2);
printf ("<td align=left>%s</td>\n",processString (currGE->geneSymbolTranscript2));
printf ("<td align=left>%s</td>\n",currGE->descriptionTranscript2);
printf ("<td align=left><a href=%s/showDetails_cgi?%s+%s>Details</a></td>\n", confp_get(Conf, "WEB_URL_CGI"), prefix,currGE->id);
puts ("</tr>");
countElements++;
}
}
gfr_deInit ();
stringDestroy (buffer);
puts ("</table><br><br>");
if( countElements == 0) puts("No fusion candidates can be found satisfying all specified criteria.");
puts ("</body>");
puts ("</html>");
fflush (stdout);
}
int main (int argc, char *argv[])
{
GfrEntry *currGE;
BLEntry *currBLE;
BLEntry currQuery;
FILE *fp;
char *line;
int count;
int countRemoved;
int index;
WordIter w;
Array blackList = arrayCreate(20, BLEntry);
config *Conf;
if ((Conf = confp_open(getenv("FUSIONSEQ_CONFPATH"))) == NULL) {
die("%s:\tCannot find .fusionseqrc: %s", argv[0], getenv("FUSIONSEQ_CONFPATH"));
return EXIT_FAILURE;
}
if( confp_get( Conf, "ANNOTATION_DIR")==NULL ) {
die("%s:\tCannot find ANNOTATION_DIR in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
if( confp_get( Conf, "BLACKLIST_FILENAME")==NULL ) {
die("%s:\tCannot find BLACKLIST_FILENAME in the configuration file: %s)", argv[0], getenv("FUSIONSEQ_CONFPATH") );
return EXIT_FAILURE;
}
Stringa buffer=stringCreate( 100 );
stringPrintf( buffer, "%s/%s", confp_get( Conf, "ANNOTATION_DIR"), confp_get( Conf, "BLACKLIST_FILENAME") );
/* fp = fopen( string( buffer ), "r" );
if( !fp ) die("Unable to open file: %s", string(buffer));
stringDestroy( buffer );
*/
// reading blacklist file
LineStream ls = ls_createFromFile( string(buffer) );
while( line = ls_nextLine(ls) ) {
w = wordIterCreate( line, "\t", 1);
currBLE = arrayp( blackList, arrayMax(blackList), BLEntry);
currBLE->gene1 = hlr_strdup ( wordNext(w) );
currBLE->gene2 = hlr_strdup ( wordNext(w) );
wordIterDestroy(w);
}
//fclose(fp);
ls_destroy( ls );
stringDestroy( buffer );
arraySort( blackList, (ARRAYORDERF) sortBlackListByName1);
// beginFiltering
count = 0;
countRemoved = 0;
gfr_init ("-");
puts (gfr_writeHeader ());
while (currGE = gfr_nextEntry ()) { // reading the gfr
if( currGE->geneSymbolTranscript1 == NULL ) {
die("Gene symbols are not present in the GFR file. Please run gfrAddInfo before gfrBlackListFilter.");
return EXIT_FAILURE;
}
// creating a new query to the black list
currQuery.gene1 = currGE->geneSymbolTranscript1;
currQuery.gene2 = currGE->geneSymbolTranscript2;
if( strEqual( currQuery.gene1 , currQuery.gene2 ) ) {
countRemoved++;
continue;
}
// searching against read_1/read_2
int res = arrayFind( blackList, &currQuery,
&index, (ARRAYORDERF) sortBlackListByName1);
if( !res ) { // not found, then searching against read_2/read_1
currQuery.gene1 = currGE->geneSymbolTranscript2;
currQuery.gene2 = currGE->geneSymbolTranscript1;
res = arrayFind( blackList, &currQuery,
&index, (ARRAYORDERF) sortBlackListByName1 );
if( !res ) { // not found, write the instance to stdout, update the counts
puts (gfr_writeGfrEntry (currGE));
count++;
} else { // found: read2/read1
countRemoved++;
}
} else { //found: read1/read2
countRemoved++;
}
}
gfr_deInit ();
arrayDestroy( blackList );
warn ("%s_BlackListFilter: %s",argv[0], confp_get( Conf, "BLACKLIST_FILENAME"));
warn ("%s_numRemoved: %d",argv[0],countRemoved);
warn ("%s_numGfrEntries: %d",argv[0],count);
confp_close( Conf);
return 0;
}
