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;
}
static Fastq* fastq_processNextSequence (int freeMemory, int truncateName)
{
char *line;
static Fastq* currFQ = NULL;
int count;
Seq* currSeq = NULL;
if (ls_isEof (lsFastq)) {
if (freeMemory) {
fastq_freeFastq (currFQ);
}
return NULL;
}
count = 0;
while ( (line=ls_nextLine (lsFastq)) && (count<4) ) {
if (line[0] == '\0') {
continue;
}
if (line[0] == '@') {
if (freeMemory) {
fastq_freeFastq (currFQ);
}
count++;
AllocVar (currFQ);
AllocVar (currFQ->seq);
currSeq = currFQ->seq;
currSeq->name = hlr_strdup (line + 1);
if (truncateName) {
currSeq->name = firstWordInLine (skipLeadingSpaces (currSeq->name));
}
line = ls_nextLine (lsFastq); // reading sequence
currSeq->sequence = hlr_strdup ( line );
currSeq->size = strlen (currSeq->sequence);
count++;
line = ls_nextLine (lsFastq); // reading quality ID
if( line[0] != '+' )
die("Expected quality ID: '+' or '+%s'", currSeq->name );
count++;
line = ls_nextLine (lsFastq); // reading quality
currFQ->quality = hlr_strdup( line );
count++;
}
}
ls_back (lsFastq,1);
return currFQ;
}
void pr3p_run (LineStream ls) {
/**
Parse the primer3 output and call the registered functions.<br>
Postcondition: the functions registered have been called
@param[in] ls - input line stream with prime output
*/
int goOn = 1;
char *line;
while ((line = ls_nextLine (ls)) != NULL) {
if (strStartsWithC (line,"# EPRIMER3 RESULTS FOR ")) {
char name[100];
sscanf (line+23,"%99s",name);
if (goOn && sequence_hook != NULL)
goOn = (*sequence_hook) (name);
}
else if (strStartsWithC (line,"# CLONINGPRIMERS RESULTS FOR ")) {
char name[100];
sscanf (line+29,"%99s",name);
if (goOn && sequence_hook != NULL)
goOn = (*sequence_hook) (name);
}
else if (line[0] == '#' || line[0] == '\0')
continue;
else if (strstr (line,"PRODUCT SIZE:")) {
int num,len;
if (sscanf (line,"%d %*s %*s %d",&num,&len) != 2)
die ("primer3parser: format error on line %s",line);
if (goOn && product_hook != NULL)
goOn = (*product_hook) (num,len);
}
else if (strstr (line,"FORWARD PRIMER") || strstr (line,"REVERSE PRIMER") ||
strstr (line,"INTERNAL OLIGO")) {
int start,len;
float tm,gc;
char seq[101];
if (sscanf (line,"%*s %*s %d %d %f %f %100s",
&start,&len,&tm,&gc,seq) != 5)
die ("primer3parser: format error on line %s",line);
if (strstr (line,"FORWARD PRIMER")) {
if (goOn && forward_hook != NULL)
goOn = (*forward_hook) (start,start+len-1,seq,gc,tm);
}
else if (strstr (line,"REVERSE PRIMER")) {
if (goOn && reverse_hook != NULL)
goOn = (*reverse_hook) (start,start+len-1,seq,gc,tm);
}
else if (strstr (line,"INTERNAL OLIGO")) {
if (goOn && internal_hook != NULL)
goOn = (*internal_hook) (start,start+len-1,seq,gc,tm);
}
}
}
}
/**
* Returns a pointer to next ElandQuery.
* @pre The module has been initialized using elandParser_init().
* Parse entries of the following format:
\verbatim
>FC30H5TAA_100308:2:1:1647:1161 GCTTACATTTTTCCTCTCTACATTATC U0 1 0 0 chr17.fa 8466296 F ..
>FC30H5TAA_100308:2:1:1588:122 GAGTTAGCCTTGGGACCCCTACTTCTT U0 1 0 0 chr3.fa 61525628 F ..
>FC30H5TAA_100308:2:1:1642:123 GGTGAGAGCCGCGACGGGCTTTAGGCG NM 0 0 0
>FC30H5TAA_100308:2:1:1630:119 CCGCCATTGCCAGCCCCCAGCTGACGG R2 0 0 2
>FC30H5TAA_100308:2:1:1603:120 GCAAGATGAAGTGAAAGGTAAAGAATC U1 0 1 1 chrM.fa 15277 R .. 26A
\endverbatim
*/
ElandQuery* elandParser_nextQuery (void)
{
WordIter w;
char *line,*token,*pos;
static ElandQuery *currElandQuery = NULL;
while (line = ls_nextLine (ls)) {
if (line[0] == '\0') {
continue;
}
elandParser_freeQuery (currElandQuery);
currElandQuery = NULL;
AllocVar (currElandQuery);
w = wordIterCreate (line,"\t",0);
currElandQuery->sequenceName = hlr_strdup (wordNext (w) + 1); // remove the '>' character at beginning of the line
currElandQuery->sequence = hlr_strdup (wordNext (w));
currElandQuery->matchCode = hlr_strdup (wordNext (w));
if (strEqual (currElandQuery->matchCode,"QC")) {
wordIterDestroy (w);
return currElandQuery;
}
currElandQuery->exactMatches = atoi (wordNext (w));
currElandQuery->oneErrorMatches = atoi (wordNext (w));
currElandQuery->twoErrorMatches = atoi (wordNext (w));
token = wordNext (w);
if (token == NULL) {
wordIterDestroy (w);
return currElandQuery;
}
if (!(pos = strchr (token,'.'))) {
die ("Expected '.' in chromosome name: %s",token);
}
*pos = '\0';
currElandQuery->chromosome = hlr_strdup (pos + 1);
currElandQuery->position = atoi (wordNext (w));
token = wordNext (w);
if (token[0] == 'F') {
currElandQuery->strand = '+';
}
else if (token[0] == 'R') {
currElandQuery->strand = '-';
}
wordIterDestroy (w);
return currElandQuery;
}
elandParser_freeQuery (currElandQuery);
currElandQuery = NULL;
return currElandQuery;
}
/**
* Get the next BlastQuery.
* @pre The module has been initialized using blastParser_init().
*/
BlastQuery* blastParser_nextQuery (void)
{
char *line,*pos;
static char *queryName = NULL;
static char *prevBlastQueryName = NULL;
static BlastQuery *currBlastQuery = NULL;
int first;
if (!ls_isEof (ls)) {
blastParser_freeQuery (currBlastQuery);
currBlastQuery = NULL;
AllocVar (currBlastQuery);
currBlastQuery->entries = arrayCreate (5,BlastEntry);
first = 1;
while (line = ls_nextLine (ls)) {
if (line[0] == '\0') {
continue;
}
pos = strchr (line,'\t');
*pos = '\0';
strReplace (&queryName,line);
if (first == 1 || strEqual (prevBlastQueryName,queryName)) {
blastParser_processLine (pos + 1,currBlastQuery);
}
else {
ls_back (ls,1);
return currBlastQuery;
}
if (first == 1) {
currBlastQuery->qName = hlr_strdup (queryName);
first = 0;
}
strReplace(&prevBlastQueryName,queryName);
}
if (first == 1) {
return NULL;
}
else {
return currBlastQuery;
}
}
blastParser_freeQuery (currBlastQuery);
currBlastQuery = NULL;
return NULL;
}
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;
}
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)
{
LineStream ls;
Texta tokens = NULL;
char *line;
int hasQual = 0;
int hasSeqs = 0;
int start=1;
ls = ls_createFromFile ("-");
while (line = ls_nextLine (ls)) {
// Put all the lines of the SAM header in comments
if (line[0] == '@') {
printf ("# %s\n", line);
continue;
}
// Parse each SAM entry and store into array
tokens = textFieldtokP (line, "\t");
if (arrayMax (tokens) < 11) {
textDestroy( tokens );
ls_destroy (ls);
die ("Invalid SAM entry: %s", line);
}
SamEntry *currSamE = NULL;
SamEntry *mateSamE = NULL;
AllocVar(currSamE );
int ret = generateSamEntry( tokens, currSamE, &hasSeqs, &hasQual );
textDestroy( tokens );
if ( ret==0 ) {
if ( isPaired ( currSamE ) )
ls_nextLine( ls ); // discarding next entry too (the mate)
destroySamEntry( currSamE );
freeMem( currSamE );
continue;
}
if ( isPaired( currSamE ) ) {
int hasQual2, hasSeq2;
AllocVar( mateSamE );
Texta secondEnd = NULL;
secondEnd = textFieldtok (ls_nextLine( ls ) , "\t");
ret = generateSamEntry( secondEnd, mateSamE, &hasSeq2, &hasQual2 );
textDestroy( secondEnd );
if( ret == 0 ) {
destroySamEntry( currSamE );
destroySamEntry( mateSamE );
freeMem( currSamE );
freeMem( mateSamE );
continue;
}
if (strcmp (currSamE->qname, mateSamE->qname) != 0) {
die ("Please note that for paired-end data, sam2mrf requires the mate pairs to be on subsequent lines. You may want to sort the SAM file first.\nEx: sort -r file.sam | sam2mrf > file.mrf\n");
}
}
// Print MRF headers
if( start ) {
printf ("%s", MRF_COLUMN_NAME_BLOCKS);
if (hasSeqs) printf("\t%s", MRF_COLUMN_NAME_SEQUENCE);
if (hasQual) printf("\t%s", MRF_COLUMN_NAME_QUALITY_SCORES);
printf ("\t%s\n", MRF_COLUMN_NAME_QUERY_ID);
start=0;
}
// Print AlignmentBlocks
printMrfAlignBlocks (currSamE, R_FIRST);
if( isPaired ( currSamE ) ) {
printf ("|");
printMrfAlignBlocks (mateSamE, R_SECOND);
}
seq_init();
// Print Sequence
if (hasSeqs) {
if (!currSamE->seq)
die ("Entry missing sequence column\n");
if( currSamE->flags & S_QUERY_STRAND )
seq_reverseComplement( currSamE->seq, strlen(currSamE->seq));
printf ("\t%s", currSamE->seq);
if (mateSamE) {
if (!mateSamE->seq)
die ("Entry missing sequence column\n");
if( mateSamE->flags & S_MATE_STRAND )
seq_reverseComplement( mateSamE->seq, strlen(mateSamE->seq));
printf ("|%s", mateSamE->seq);
}
}
// Print quality scores
if (hasQual) {
if (!currSamE->qual)
die ("Entry missing quality scores column\n");
printf ("\t%s", currSamE->qual);
if (mateSamE) {
if (!mateSamE->qual)
die ("Entry missing quality scores column\n");
printf ("|%s", mateSamE->qual);
}
}
// Print queryID
if (mateSamE) {
printf ("\t%s|%s", currSamE->qname,"2"); // No need to print out both IDs, but need the pipe symbol for consistency
}
else {
printf ("\t%s", currSamE->qname);
}
printf("\n");
destroySamEntry( currSamE );
freeMem( currSamE );
if( isPaired( currSamE ) ) {
destroySamEntry ( mateSamE );
freeMem( mateSamE );
}
}
// clean up
ls_destroy (ls);
return EXIT_SUCCESS;
}
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);
if (argc != 2) {
usage ("%s <blackList.txt>",argv[0]);
}
fp = fopen( argv[1], "r" );
if( !fp ) die("Unable to open file: %s", argv[1]);
// reading blacklist file
LineStream ls = ls_createFromFile( argv[1] );
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);
arraySort( blackList, (ARRAYORDERF) sortBlackListByName1);
// beginFiltering
count = 0;
countRemoved = 0;
gfr_init ("-");
puts (gfr_writeHeader ());
while (currGE = gfr_nextEntry ()) { // reading the gfr
// creating a new query to the black list
currQuery.gene1 = currGE->geneSymbolTranscript1;
currQuery.gene2 = currGE->geneSymbolTranscript2;
// 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], argv[1]);
warn ("%s_numRemoved: %d",argv[0],countRemoved);
warn ("%s_numGfrEntries: %d",argv[0],count);
return 0;
}
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;
}
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);
}
