int main (int argc, char *argv[])
{
GfrEntry *currGE;
int count;
int countRemoved;
if (argc != 2) {
usage ("%s <string>",argv[0]);
}
count = 0;
countRemoved = 0;
gfr_init ("-");
puts (gfr_writeHeader ());
while (currGE = gfr_nextEntry ()) {
if (currGE->descriptionTranscript1 == NULL ||
currGE->descriptionTranscript2 == NULL) {
die ("Transcript description is missing");
}
if (strCaseStr (currGE->descriptionTranscript1,argv[1]) ||
strCaseStr (currGE->descriptionTranscript2,argv[1])) {
countRemoved++;
continue;
}
puts (gfr_writeGfrEntry (currGE));
count++;
}
gfr_deInit ();
warn ("%s_string: %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;
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;
}
int main (int argc, char *argv[])
{
GfrEntry *currGE;
int count;
int countRemoved;
int i, j;
int foundEST;
if (argc != 2) {
usage ("%s <EST.interval>",argv[0]);
}
intervalFind_addIntervalsToSearchSpace( argv[1], 0);
// beginFiltering
count = 0;
countRemoved = 0;
gfr_init ("-");
puts (gfr_writeHeader ());
while (currGE = gfr_nextEntry ()) { // reading the gfr
foundEST = 0;
if( strEqual( currGE->fusionType, "cis" ) ) {
if( ! strEqual( currGE->chromosomeTranscript1, currGE->chromosomeTranscript2 ) )
die("The two genes are not on the same chromosomes: %s - %s", currGE->chromosomeTranscript1, currGE->chromosomeTranscript2 );
int start1, end1, start2, end2;
findCoordinates( currGE, &start1, &end1, &start2, &end2 );
Array intervals1 = arrayCopy( intervalFind_getOverlappingIntervals( currGE->chromosomeTranscript1, start1, end1 ) );
Array intervals2 = intervalFind_getOverlappingIntervals( currGE->chromosomeTranscript2, start2, end2 );
for( i=0; i<arrayMax( intervals1 ); i++ ) {
Interval* currInterval1 = arru( intervals1, i, Interval* );
for( j=0; j<arrayMax ( intervals2 ); j++ ) {
Interval* currInterval2 = arru( intervals2, j, Interval* );
if( currInterval1==currInterval2 ) {
foundEST = 1;
i = arrayMax( intervals1 );
j = arrayMax( intervals2 );
}
}
}
arrayDestroy( intervals1 );
}
if( foundEST )
countRemoved++;
else {
puts (gfr_writeGfrEntry (currGE));
count++;
}
}
gfr_deInit ();
warn ("%s_EST_data: %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;
int count;
gfr_init ("-");
gfr_addNewColumnType (GFR_COLUMN_NAME_PAIR_COUNT);
puts (gfr_writeHeader ());
count = 0;
while (currGE = gfr_nextEntry ()){
obtainPairCounts (currGE);
puts (gfr_writeGfrEntry (currGE)); fflush (stdout);
count++;
}
gfr_deInit ();
warn ("%s_numGfrEntries: %d",argv[0],count);
return 0;
}
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;
}
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;
}
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);
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[])
{
FILE* ftmp = NULL;
if ((Conf = confp_open(getenv("FUSIONSEQ_CONFPATH"))) == NULL)
return EXIT_FAILURE;
cgiInit();
cgiHeader("text/html");
if (argc == 3) {
GfrEntry *currGE;
Stringa buffer;
GfrPairCount *currGEPC;
GfrInterRead *currGIR;
int i;
puts ("<html>");
puts ("<head>");
html_printGenericStyleSheet (12);
puts ("<title>geneFusions Details</title>\n");
puts ("</head>");
puts ("<body>");
buffer = stringCreate (100);
stringPrintf (buffer, "%s/%s.gfr", confp_get(Conf, "WEB_DATA_DIR"),argv[1]);
gfr_init (string (buffer));
while (currGE = gfr_nextEntry ()){
fflush( stdout );
if (!strEqual (currGE->id,argv[2])) {
continue;
}
printf ("<h1>Detailed summary for potential gene fusion candidate</h1><br>");
puts ("<table border=0 cellpadding=10>");
puts ("<tr align=left valign=top>");
puts ("<td width=400>");
puts ("<h2>Summary information</h2><br>");
printf ("<b>Identifier</b>: %s<br><br>\n",currGE->id);
printf ("<b>Number of inter paired-end reads</b>: %d<br><br>\n",currGE->numInter);
printf ("<b>Type</b>: %s<br><br>\n",currGE->fusionType);
stringPrintf(buffer, "%s/GFF/%s.gff", confp_get(Conf, "WEB_DATA_DIR"),currGE->id);
ftmp = fopen( string(buffer), "r" ); // displaying this only if data are present
if (ftmp) {
printf("<b>Connected Reads</b>: <a href=%s&hgt.customText=%s/GFF/%s.gff target=blank>UCSC connectivity graph</a><br>\n",
htmlLinker_generateLinkToGenomeBrowserAtUCSC ("hg18","vertebrate","human",
currGE->chromosomeTranscript1,
currGE->startTranscript1 - atoi(confp_get(Conf, "UCSC_GENOME_BROWSER_FLANKING_REGION")),
currGE->endTranscript2 + atoi(confp_get(Conf, "UCSC_GENOME_BROWSER_FLANKING_REGION"))),
confp_get(Conf, "WEB_DATA_LINK"),currGE->id);
fclose( ftmp );
}
puts ("</td>");
puts ("<td>");
puts ("<h2>Transcript connectivity graph</h2>");
printf ("<img src=%s/IMAGES/%s.jpg alt=geneFusionImage>\n", confp_get(Conf, "WEB_DATA_LINK"), currGE->id);
puts ("</td>");
puts ("<td>");
puts ("<h2>Transcript connectivity table</h2><br>");
puts ("<table border=0>");
puts ("<tr align=left>");
puts ("<th width=200>Pair Type</th>");
puts ("<th width=200>Entry transcript 1</th>");
puts ("<th width=200>Entry transcript 2</th>");
puts ("<th width=200>Counts</th>");
puts ("</tr>");
fflush( stdout );
for (i = 0; i < arrayMax (currGE->pairCounts); i++) {
currGEPC = arrp (currGE->pairCounts,i,GfrPairCount);
printf ("<tr><td>%s</td><td>%s</td><td>%s</td><td>%.2f</td></tr>\n",
getPairTypeName(currGEPC->pairType),
getEntryNumber(currGEPC->number1, currGEPC->pairType, 1),
getEntryNumber(currGEPC->number2, currGEPC->pairType, 2),
currGEPC->count);
}
puts ("</table>");
puts ("</td>");
puts ("</tr>");
puts ("</table>");
puts ("<br>");
puts ("<h2>Transcript information</h2><br>");
puts ("<table border=1 cellpadding=10 width=\"80%\">");
puts ("<tr align=left>");
puts ("<th width=\"20%\"></th>");
puts ("<th><font color='blue'>Transcript 1</font></th>");
puts ("<th><font color='orange'>Transcript 2</font></th>");
puts ("</tr>");
puts ("<tr align=left>");
puts ("<td width=\"20%\"><b>Gene symbol(s)</b></td>");
printf ("<td width=\"30%%\"><font color='blue'>%s</font></td>\n",processString (currGE->geneSymbolTranscript1));
printf ("<td width=\"30%%\"><font color='orange'>%s</font></td>\n",processString (currGE->geneSymbolTranscript2));
puts ("</tr>");
puts ("<tr align=left>");
puts ("<td width=\"20%\"><b>Coordinates</b></td>");
printf ("<td width=\"30%%\">%s:%d-%d</td>\n",currGE->chromosomeTranscript1,currGE->startTranscript1,currGE->endTranscript1);
printf ("<td width=\"30%%\">%s:%d-%d</td>\n",currGE->chromosomeTranscript2,currGE->startTranscript2,currGE->endTranscript2);
puts ("</tr>");
puts ("<tr align=left>");
puts ("<td width=\"20%\"><b>Strand</b></td>");
printf ("<td width=\"30%%\">%c</td>\n",currGE->strandTranscript1);
printf ("<td width=\"30%%\">%c</td>\n",currGE->strandTranscript2);
puts ("</tr>");
puts ("<tr align=left>");
puts ("<td width=\"20%\"><b>Gene description(s)</b></td>");
printf ("<td width=\"30%%\">%s</td>\n",processString (currGE->descriptionTranscript1));
printf ("<td width=\"30%%\">%s</td>\n",processString (currGE->descriptionTranscript2));
puts ("</tr>");
puts ("<tr align=left>");
puts ("<td width=\"20%\"><b>Number of exons</b></td>");
printf ("<td width=\"30%%\">%d</td>\n",currGE->numExonsTranscript1);
printf ("<td width=\"30%%\">%d</td>\n",currGE->numExonsTranscript2);
puts ("</tr>");
puts ("<tr align=left>");
puts ("<td width=\"20%\"><b>Number of intra paired-end reads</b></td>");
printf ("<td width=\"30%%\">%d</td>\n",currGE->numIntra1);
printf ("<td width=\"30%%\">%d</td>\n",currGE->numIntra2);
puts ("</tr>");
puts ("<tr align=left>");
puts ("<td width=\"20%\"><b>Links</b></td>");
printf ("<td width=\"30%%\">[<a href=%s&hgt.customText=%s/BED/%s_1.bed target=blank>UCSC genome browser</a>] [<a href=%s/FASTA/%s_1.fasta>FASTA file</a>]<br></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"))),
confp_get(Conf, "WEB_DATA_LINK"),
currGE->id,
confp_get(Conf, "WEB_DATA_LINK"),
currGE->id);
printf ("<td width=\"30%%\">[<a href=%s&hgt.customText=%s/BED/%s_2.bed target=blank>UCSC genome browser</a>] [<a href=%s/FASTA/%s_2.fasta>FASTA file</a>]<br></td></tr>\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"))),
confp_get(Conf, "WEB_DATA_LINK"),
currGE->id,
confp_get(Conf, "WEB_DATA_LINK"),
currGE->id);
puts ("<tr align=left>");
puts ("<td width=\"20%\"><b>Expression</b></td>");
stringPrintf(buffer, "%s/BGRS/%s_%s.bgr.gz",
confp_get(Conf, "WEB_DATA_DIR"),
argv[1],
currGE->chromosomeTranscript1);
ftmp = fopen( string(buffer), "r" ); // displaying this only if data are present
puts("<td width=\"30%\">");
if( ftmp ) {
printf ("[<a href=%s&hgt.customText=%s/BGRS/%s_%s.bgr.gz target=blank>Expression %s</a>]",
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"))),
confp_get(Conf, "WEB_DATA_LINK"),
argv[1],
currGE->chromosomeTranscript1,
currGE->chromosomeTranscript1);
fclose(ftmp);
}
puts("</td>");
stringPrintf(buffer, "%s/BGRS/%s_%s.bgr.gz", confp_get(Conf, "WEB_DATA_DIR"),argv[1],currGE->chromosomeTranscript2);
ftmp = fopen( string(buffer), "r" ); // displaying this only if data are present
puts("<td width=\"30%\">");
if( ftmp ) {
printf ("[<a href=%s&hgt.customText=%s/BGRS/%s_%s.bgr.gz target=blank>Expression %s</a>]",
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"))),
confp_get(Conf, "WEB_DATA_LINK"),
argv[1],
currGE->chromosomeTranscript2,
currGE->chromosomeTranscript2);
fclose(ftmp);
}
puts("</td>");
puts("</tr>");
puts ("</table><br><br>");
puts ("<h2>Breakpoint analysis</h2><br>");
puts ("<table border=1 width=\"80%\" cellpadding=10><thead><tr><th>Orientation</th><th>Alignments</th><th colspan=2>Breakpoints</th></tr></thead><tbody>");
puts ("<tr><td>Orientation AB</td>");
if (currGE->strandTranscript1=='+') {
currGE->strandTranscript2=='+' ? stringPrintf(buffer, "AB_trans1F_trans2F") : stringPrintf(buffer, "AB_trans1F_trans2R");
} else if( currGE->strandTranscript1 == '-') {
currGE->strandTranscript2=='+' ? stringPrintf(buffer, "AB_trans1R_trans2F") : stringPrintf(buffer, "AB_trans1R_trans2R");
} else {
die("Strand informatation is not correct (transcript 1): %c", currGE->strandTranscript1);
}
printf ("<td align=center><a href=%s/ALIGNMENTS/%s_AB_breakPointAlignments.txt><img src=%s/IMAGES/%s.png></img> AB</a></td>",
confp_get(Conf, "WEB_DATA_LINK"),
currGE->id,
confp_get(Conf, "WEB_DATA_LINK"),
string(buffer));
printf ("<td align=center><a href=%s&hgt.customText=%s/WIGS/%s_AB_breakPointsTranscript1.wig target=blank>Breakpoints transcript 1 UCSC Genome Browser</a></td>",
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"))),
confp_get(Conf, "WEB_DATA_LINK"),
currGE->id);
printf ("<td align=center><a href=%s&hgt.customText=%s/WIGS/%s_AB_breakPointsTranscript2.wig target=blank>Breakpoints transcript 2 UCSC Genome Browser</a></td></tr>",
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"))),
confp_get(Conf, "WEB_DATA_LINK"),
currGE->id);
fflush(stdout);
puts ("<tr><td>Orientation BA</td>");
if (currGE->strandTranscript1 == '+') {
currGE->strandTranscript2=='+' ? stringPrintf(buffer, "BA_trans1F_trans2F") : stringPrintf(buffer, "BA_trans1F_trans2R");
} else if( currGE->strandTranscript1 == '-') {
currGE->strandTranscript2=='+' ? stringPrintf(buffer, "BA_trans1R_trans2F") : stringPrintf(buffer, "BA_trans1R_trans2R");
} else {
die("Strand informatation is not correct (transcript2): %c", currGE->strandTranscript2);
}
printf ("<td align=center><a href=%s/ALIGNMENTS/%s_BA_breakPointAlignments.txt><img src=%s/IMAGES/%s.png></img> BA</a></td>",
confp_get(Conf, "WEB_DATA_LINK"),
currGE->id,
confp_get(Conf, "WEB_DATA_LINK"),
string(buffer));
printf ("<td align=center><a href=%s&hgt.customText=%s/WIGS/%s_BA_breakPointsTranscript2.wig target=blank>Breakpoints transcript 2 UCSC Genome Browser</a></td>",
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"))),
confp_get(Conf, "WEB_DATA_LINK"),
currGE->id);
printf ("<td align=center><a href=%s&hgt.customText=%s/WIGS/%s_BA_breakPointsTranscript1.wig target=blank>Breakpoints transcript 1 UCSC Genome Browser</a></td></tr>",
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"))),
confp_get(Conf, "WEB_DATA_LINK"),
currGE->id);
puts ("</tbody></table>");
puts ("<br><br><br>");
fflush(stdout);
puts ("<h2>Read coordinates</h2><br>");
puts ("<table border=0>");
puts ("<tr align=left>");
puts ("<th width=\"10%\">Pair Type</th>");
puts ("<th width=\"10%\">Entry Transcript 1</th>");
puts ("<th width=\"10%\">Read start transcript 1</th>");
puts ("<th width=\"10%\">Read end transcript 1</th>");
puts ("<th width=\"10%\">Entry Transcript 2</th>");
puts ("<th width=\"10%\">Read start transcript 2</th>");
puts ("<th width=\"10%\">Read end transcript 2</th>");
puts ("</tr>");
for (i = 0; i < arrayMax (currGE->interReads); i++) {
currGIR = arrp (currGE->interReads,i,GfrInterRead);
printf ("<tr><td>%s</td><td>%s</td><td>%d</td><td>%d</td><td>%s</td><td>%d</td><td>%d</td></tr>\n",
getPairTypeName(currGIR->pairType),
getEntryNumber(currGIR->number1, currGIR->pairType, 1),
currGIR->readStart1,currGIR->readEnd1,
getEntryNumber(currGIR->number2,currGIR->pairType, 2),
currGIR->readStart2,
currGIR->readEnd2);
}
puts ("</table><br><br><br>");
puts ("</body>");
puts ("</html>");
fflush (stdout);
}
}
confp_close(Conf);
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);
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;
}
