/**
* 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;
}
static void addSubstitution (Alteration *currAlteration, char* proteinSequenceBeforeIndel, char *proteinSequenceAfterIndel, int indelOffset)
{
int lengthBefore,lengthAfter;
static Stringa buffer = NULL;
int i;
int diff;
int index;
stringCreateClear (buffer,100);
index = ((currAlteration->relativePosition - 1) / 3);
lengthBefore = strlen (proteinSequenceBeforeIndel);
lengthAfter = strlen (proteinSequenceAfterIndel);
diff = abs (lengthBefore - lengthAfter);
if (lengthBefore < lengthAfter) {
stringPrintf (buffer,"%d_%c->",index + 1,proteinSequenceBeforeIndel[index]);
for (i = 0; i <= diff; i++) {
stringAppendf (buffer,"%c",proteinSequenceAfterIndel[index + i]);
}
}
else if (lengthBefore > lengthAfter) {
stringPrintf (buffer,"%d_",index + 1);
for (i = 0; i <= diff; i++) {
stringAppendf (buffer,"%c",proteinSequenceBeforeIndel[index + i]);
}
stringAppendf (buffer,"->%c",proteinSequenceAfterIndel[index]);
}
else {
stringPrintf (buffer,"%d_%s->",index,subString (proteinSequenceBeforeIndel,index - 1,index + (int)ceil ((double)indelOffset / 3)));
stringAppendf (buffer,"%s",subString (proteinSequenceAfterIndel,index - 1,index + (int)ceil ((double)indelOffset / 3)));
}
currAlteration->substitution = hlr_strdup (string (buffer));
}
int main (int argc, char *argv[])
{
MrfEntry *currEntry;
char *targetName;
int targetStart,targetEnd;
WordIter w;
int count=0;
int currCount=0;
if (argc != 2) {
usage ("%s <targetName:targetStart-targetEnd>",argv[0]);
}
w = wordIterCreate (argv[1],":- ",0);
targetName = hlr_strdup (wordNext (w));
targetStart = atoi (wordNext (w));
targetEnd = atoi (wordNext (w));
wordIterDestroy (w);
mrf_init ("-");
while (currEntry = mrf_nextEntry ()) {
currCount = processEntry (currEntry,targetName,targetStart,targetEnd);
count = currCount+count;
}
printf("Count for %s:%d-%d = %d\n", targetName, targetStart, targetEnd, count);
mrf_deInit ();
hlr_free (targetName);
return 0;
}
int main (int argc, char *argv[])
{
MrfEntry *currEntry;
char *targetName;
int targetStart,targetEnd;
WordIter w;
if (argc != 2) {
usage ("%s <targetName:targetStart-targetEnd>",argv[0]);
}
w = wordIterCreate (argv[1],":- ",0);
targetName = hlr_strdup (wordNext (w));
targetStart = atoi (wordNext (w));
targetEnd = atoi (wordNext (w));
wordIterDestroy (w);
mrf_init ("-");
puts (mrf_writeHeader ());
while (currEntry = mrf_nextEntry ()) {
processEntry (currEntry,targetName,targetStart,targetEnd);
}
mrf_deInit ();
hlr_free (targetName);
return 0;
}
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;
}
}
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;
}
static void createGffEntry( Array gffEntries, MrfRead *currRead, int groupNumber ) {
int i;
MrfBlock *currBlock;
GffEntry *currGffEntry;
static Stringa buffer = NULL;
stringCreateClear (buffer,100);
for (i = 0; i < arrayMax (currRead->blocks); i++) {
currBlock = arrp (currRead->blocks,i,MrfBlock);
currGffEntry = arrayp (gffEntries,arrayMax (gffEntries),GffEntry);
stringPrintf (buffer,"%s\tMRF\texon\t%d\t%d\t.\t.\t.\tTG%d",
currBlock->targetName,
currBlock->targetStart,
currBlock->targetEnd,
//currBlock->strand,
groupNumber);
currGffEntry->targetName = hlr_strdup (currBlock->targetName);
currGffEntry->line = hlr_strdup (string (buffer));
}
}
static char* lookUpTreeFam (Array kgTreeFams, char *transcript)
{
KgTreeFam testKGTF;
int index;
int foundIt;
foundIt = 0;
testKGTF.transcriptName = hlr_strdup (transcript);
foundIt = arrayFind (kgTreeFams,&testKGTF,&index,(ARRAYORDERF)sortKgTreeFamsByTranscriptName);
hlr_free (testKGTF.transcriptName);
if (foundIt) {
return arrp (kgTreeFams,index,KgTreeFam)->treeFamId;
}
return NULL;
}
/**
* 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;
}
static void blastParser_processLine (char* line, BlastQuery* currBlastQuery)
{
WordIter w;
BlastEntry *currEntry;
currEntry = arrayp (currBlastQuery->entries,arrayMax (currBlastQuery->entries),BlastEntry);
w = wordIterCreate (line,"\t",0);
currEntry->tName = hlr_strdup (wordNext (w));
currEntry->percentIdentity = atof (wordNext (w));
currEntry->alignmentLength = atoi (wordNext (w));
currEntry->misMatches = atoi (wordNext (w));
currEntry->gapOpenings = atoi (wordNext (w));
currEntry->qStart = atoi (wordNext (w));
currEntry->qEnd = atoi (wordNext (w));
currEntry->tStart = atoi (wordNext (w));
currEntry->tEnd = atoi (wordNext (w));
currEntry->evalue = atof (wordNext (w));
currEntry->bitScore = atof (wordNext (w));
wordIterDestroy (w);
}
void performSegmentation (Array tars, Array wigs, char* targetName, double threshold, int maxGap, int minRun)
{
Tar *currTar;
Wig *currWig,*nextWig;
int i,j,endPosition;
int countBelowThreshold;
i = 0;
while (i < arrayMax (wigs)) {
currWig = arrp (wigs,i,Wig);
if (currWig->value < threshold) {
i++;
continue;
}
j = i + 1;
endPosition = j;
countBelowThreshold = 0;
while (j < arrayMax (wigs)) {
nextWig = arrp (wigs,j,Wig);
if (nextWig->value < threshold) {
countBelowThreshold++;
if (countBelowThreshold >= maxGap) {
break;
}
}
else {
countBelowThreshold = 0;
endPosition = j;
}
j++;
}
if ((endPosition - 1 - currWig->position + 1) >= minRun) {
currTar = arrayp (tars,arrayMax (tars),Tar);
currTar->start = currWig->position;
currTar->end = endPosition + 1;
currTar->targetName = hlr_strdup (targetName);
}
i = j;
}
}
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;
}
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;
}
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;
}
int main (int argc, char *argv[])
{
char *queryString;
if ((Conf = confp_open(getenv("FUSIONSEQ_CONFPATH"))) == NULL)
return EXIT_FAILURE;
cgiInit();
cgiHeader("text/html");
queryString = cgiGet2Post();
if (queryString[0] == '\0') {
puts ("<html>");
puts ("<head>");
html_printGenericStyleSheet (12);
puts ("<title>geneFusions</title>\n");
puts ("</head>");
puts ("<body>");
puts ("<h1>Identification of potential gene fusions using paired-end reads</h1><br><br>");
printf ("<form action=%s/geneFusions_cgi method=get>", confp_get(Conf, "WEB_URL_CGI"));
puts ("<b>Data prefix</b>: ");
puts ("<input type=text name=prefix>");
puts ("<br><br><br>");
puts ("<b>Minimum number of paired-end reads connecting two genes</b>: ");
puts ("<select name=minNum>");
puts ("<option value=2>2");
puts ("<option value=3>3");
puts ("<option value=5 selected>5");
puts ("<option value=10>10");
puts ("</select>");
puts ("<br><br><br>");
puts ("<b>Type of gene fusion</b>: ");
puts ("<select name=type>");
puts ("<option value=read-through>Read-through events");
puts ("<option value=cis>Cis events");
puts ("<option value=intra>Intra-chromosomal events");
puts ("<option value=same>Genes on the same chromosome");
puts ("<option value=inter>Genes on different chromosomes");
puts ("<option value=all selected>All potential gene fusions");
puts ("</select>");
puts ("<br><br><br>");
puts ("<input type=submit value=Submit>");
puts ("<input type=reset value=Reset>");
puts ("</form>");
puts ("</body>");
puts ("</html>");
fflush (stdout);
}
else {
int first;
Stringa item = stringCreate (20);
Stringa value = stringCreate (20);
char *iPtr,*vPtr,*prefix,*type;
int minNum;
first = 1;
cgiGetInit ();
while (cgiGetNextPair (&first,item,value)) {
iPtr = string (item);
vPtr = string (value);
if (strEqual (iPtr,"prefix")) {
prefix = hlr_strdup (vPtr);
}
if (strEqual (iPtr,"type")) {
type = hlr_strdup (vPtr);
}
if (strEqual (iPtr,"minNum")) {
minNum = atoi (vPtr);
}
}
generateOutput (prefix,type,minNum);
}
confp_close(Conf);
return EXIT_SUCCESS;
}
int main (int argc, char *argv[])
{
Array intervals;
Interval *currInterval;
SubInterval *currSubInterval;
int h,i,j;
Array seqs;
Seq *currSeq,testSeq;
int index;
Stringa buffer;
Array geneTranscriptEntries;
Texta geneTranscriptIds;
Array alterations;
Alteration *currAlteration,*nextAlteration;
char *proteinSequenceBeforeIndel;
char *proteinSequenceAfterIndel;
int numDisabledTranscripts;
Stringa disabledTranscripts;
int seqLength,refLength,altLength;
char *sequenceBeforeIndel = NULL;
int overlapMode;
int numOverlaps;
int sizeIndel,indelOffset;
int overlap;
Array coordinates;
VcfEntry *currVcfEntry;
VcfGenotype *currVcfGenotype;
int position;
Texta alternateAlleles;
int flag1,flag2;
if (argc != 3) {
usage ("%s <annotation.interval> <annotation.fa>",argv[0]);
}
intervalFind_addIntervalsToSearchSpace (argv[1],0);
geneTranscriptEntries = util_getGeneTranscriptEntries (intervalFind_getAllIntervals ());
seq_init ();
fasta_initFromFile (argv[2]);
seqs = fasta_readAllSequences (0);
fasta_deInit ();
arraySort (seqs,(ARRAYORDERF)util_sortSequencesByName);
buffer = stringCreate (100);
disabledTranscripts = stringCreate (100);
alterations = arrayCreate (100,Alteration);
vcf_init ("-");
stringPrintf (buffer,"##INFO=<ID=VA,Number=.,Type=String,Description=\"Variant Annotation, %s\">",argv[1]);
vcf_addComment (string (buffer));
puts (vcf_writeMetaData ());
puts (vcf_writeColumnHeaders ());
while (currVcfEntry = vcf_nextEntry ()) {
if (vcf_isInvalidEntry (currVcfEntry)) {
continue;
}
flag1 = 0;
flag2 = 0;
position = currVcfEntry->position - 1; // make zero-based
alternateAlleles = vcf_getAlternateAlleles (currVcfEntry);
for (h = 0; h < arrayMax (alternateAlleles); h++) {
refLength = strlen (currVcfEntry->referenceAllele);
altLength = strlen (textItem (alternateAlleles,h));
sizeIndel = abs (refLength - altLength);
indelOffset = MAX (refLength,altLength) - 1;
util_clearAlterations (alterations);
intervals = intervalFind_getOverlappingIntervals (currVcfEntry->chromosome,position,position + indelOffset);
for (i = 0; i < arrayMax (intervals); i++) {
currInterval = arru (intervals,i,Interval*);
overlapMode = OVERLAP_NONE;
numOverlaps = 0;
for (j = 0; j < arrayMax (currInterval->subIntervals); j++) {
currSubInterval = arrp (currInterval->subIntervals,j,SubInterval);
overlap = rangeIntersection (position,position + indelOffset,currSubInterval->start,currSubInterval->end);
if (currSubInterval->start <= position && (position + indelOffset) < currSubInterval->end) {
overlapMode = OVERLAP_FULLY_CONTAINED;
numOverlaps++;
}
else if (j == 0 && overlap > 0 && position < currSubInterval->start) {
overlapMode = OVERLAP_START;
numOverlaps++;
}
else if (j == (arrayMax (currInterval->subIntervals) - 1) && overlap > 0 && (position + indelOffset) >= currSubInterval->end) {
overlapMode = OVERLAP_END;
numOverlaps++;
}
else if (overlap > 0 && overlap <= indelOffset) {
overlapMode = OVERLAP_SPLICE;
numOverlaps++;
}
}
if (overlapMode == OVERLAP_NONE) {
continue;
}
currAlteration = arrayp (alterations,arrayMax (alterations),Alteration);
if (numOverlaps > 1) {
util_addAlteration (currAlteration,currInterval->name,"multiExonHit",currInterval,position,0);
continue;
}
else if (numOverlaps == 1 && overlapMode == OVERLAP_SPLICE) {
util_addAlteration (currAlteration,currInterval->name,"spliceOverlap",currInterval,position,0);
continue;
}
else if (numOverlaps == 1 && overlapMode == OVERLAP_START) {
util_addAlteration (currAlteration,currInterval->name,"startOverlap",currInterval,position,0);
continue;
}
else if (numOverlaps == 1 && overlapMode == OVERLAP_END) {
util_addAlteration (currAlteration,currInterval->name,"endOverlap",currInterval,position,0);
continue;
}
else if (numOverlaps == 1 && overlapMode == OVERLAP_FULLY_CONTAINED && altLength > refLength) {
if ((sizeIndel % 3) == 0) {
util_addAlteration (currAlteration,currInterval->name,"insertionNFS",currInterval,position,0);
}
else {
util_addAlteration (currAlteration,currInterval->name,"insertionFS",currInterval,position,0);
}
}
else if (numOverlaps == 1 && overlapMode == OVERLAP_FULLY_CONTAINED && altLength < refLength) {
if ((sizeIndel % 3) == 0) {
util_addAlteration (currAlteration,currInterval->name,"deletionNFS",currInterval,position,0);
}
else {
util_addAlteration (currAlteration,currInterval->name,"deletionFS",currInterval,position,0);
}
}
else if (numOverlaps == 1 && overlapMode == OVERLAP_FULLY_CONTAINED && altLength == refLength) {
util_addAlteration (currAlteration,currInterval->name,"substitution",currInterval,position,0);
}
else {
die ("Unexpected type: %d %s %s %s",
currVcfEntry->position,currVcfEntry->chromosome,
currVcfEntry->referenceAllele,currVcfEntry->alternateAllele);
}
if ((sizeIndel % 3) != 0 && altLength != refLength) {
continue;
}
// Only run the remaining block of code if the indel is fully contained (insertion or deletion) AND does not cause a frameshift OR
// if it is a substitution that is fully contained in the coding sequence
stringPrintf (buffer,"%s|%s|%c|",currInterval->name,currInterval->chromosome,currInterval->strand);
for (j = 0; j < arrayMax (currInterval->subIntervals); j++) {
currSubInterval = arrp (currInterval->subIntervals,j,SubInterval);
stringAppendf (buffer,"%d|%d%s",currSubInterval->start,currSubInterval->end,j < arrayMax (currInterval->subIntervals) - 1 ? "|" : "");
}
testSeq.name = hlr_strdup (string (buffer));
if (!arrayFind (seqs,&testSeq,&index,(ARRAYORDERF)util_sortSequencesByName)) {
die ("Expected to find %s in seqs",string (buffer));
}
hlr_free (testSeq.name);
currSeq = arrp (seqs,index,Seq);
strReplace (&sequenceBeforeIndel,currSeq->sequence);
seqLength = strlen (sequenceBeforeIndel);
coordinates = util_getCoordinates (currInterval);
// arraySort (coordinates,(ARRAYORDERF)util_sortCoordinatesByChromosomeAndTranscriptPosition); Array is already sorted by definition
j = 0;
stringClear (buffer);
while (j < seqLength) {
if (util_getGenomicCoordinate (coordinates,j,currVcfEntry->chromosome) == position) {
if (altLength > refLength) {
stringCat (buffer,textItem (alternateAlleles,h));
j++;
continue;
}
else if (altLength < refLength) {
stringCatChar (buffer,sequenceBeforeIndel[j]);
j = j + refLength - altLength + 1;
continue;
}
else {
stringCat (buffer,textItem (alternateAlleles,h));
j = j + altLength;
continue;
}
}
stringCatChar (buffer,sequenceBeforeIndel[j]);
j++;
}
util_destroyCoordinates (coordinates);
proteinSequenceBeforeIndel = hlr_strdup (util_translate (currInterval,sequenceBeforeIndel));
proteinSequenceAfterIndel = hlr_strdup (util_translate (currInterval,string (buffer)));
addSubstitution (currAlteration,proteinSequenceBeforeIndel,proteinSequenceAfterIndel,indelOffset);
hlr_free (proteinSequenceBeforeIndel);
hlr_free (proteinSequenceAfterIndel);
}
if (arrayMax (alterations) == 0) {
continue;
}
arraySort (alterations,(ARRAYORDERF)util_sortAlterationsByGeneIdAndType);
stringClear (buffer);
i = 0;
while (i < arrayMax (alterations)) {
currAlteration = arrp (alterations,i,Alteration);
stringAppendf (buffer,"%s%d:%s:%s:%c:%s",stringLen (buffer) == 0 ? "" : ",",h + 1,currAlteration->geneName,currAlteration->geneId,currAlteration->strand,currAlteration->type);
stringClear (disabledTranscripts);
if (currAlteration->substitution[0] != '\0') {
stringAppendf (disabledTranscripts,"%s:%s:%d_%d_%s",currAlteration->transcriptName,currAlteration->transcriptId,currAlteration->transcriptLength,currAlteration->relativePosition,currAlteration->substitution);
}
else if (strEqual (currAlteration->type,"multiExonHit") || strEqual (currAlteration->type,"spliceOverlap") ||
strEqual (currAlteration->type,"startOverlap") || strEqual (currAlteration->type,"endOverlap")) {
stringAppendf (disabledTranscripts,"%s:%s:%d",currAlteration->transcriptName,currAlteration->transcriptId,currAlteration->transcriptLength);
}
else {
stringAppendf (disabledTranscripts,"%s:%s:%d_%d",currAlteration->transcriptName,currAlteration->transcriptId,currAlteration->transcriptLength,currAlteration->relativePosition);
}
numDisabledTranscripts = 1;
j = i + 1;
while (j < arrayMax (alterations)) {
nextAlteration = arrp (alterations,j,Alteration);
if (strEqual (currAlteration->geneId,nextAlteration->geneId) &&
strEqual (currAlteration->type,nextAlteration->type)) {
if (nextAlteration->substitution[0] != '\0') {
stringAppendf (disabledTranscripts,":%s:%s:%d_%d_%s",nextAlteration->transcriptName,nextAlteration->transcriptId,nextAlteration->transcriptLength,nextAlteration->relativePosition,nextAlteration->substitution);
}
else if (strEqual (nextAlteration->type,"multiExonHit") || strEqual (nextAlteration->type,"spliceOverlap") ||
strEqual (nextAlteration->type,"startOverlap") || strEqual (nextAlteration->type,"endOverlap")) {
stringAppendf (disabledTranscripts,":%s:%s:%d",nextAlteration->transcriptName,nextAlteration->transcriptId,nextAlteration->transcriptLength);
}
else {
stringAppendf (disabledTranscripts,":%s:%s:%d_%d",nextAlteration->transcriptName,nextAlteration->transcriptId,nextAlteration->transcriptLength,nextAlteration->relativePosition);
}
numDisabledTranscripts++;
}
else {
break;
}
j++;
}
i = j;
geneTranscriptIds = util_getTranscriptIdsForGeneId (geneTranscriptEntries,currAlteration->geneId);
stringAppendf (buffer,":%d/%d:%s",numDisabledTranscripts,arrayMax (geneTranscriptIds),string (disabledTranscripts));
}
if (flag1 == 0) {
printf ("%s\t%d\t%s\t%s\t%s\t%s\t%s\t%s;VA=",
currVcfEntry->chromosome,currVcfEntry->position,currVcfEntry->id,
currVcfEntry->referenceAllele,currVcfEntry->alternateAllele,
currVcfEntry->quality,currVcfEntry->filter,currVcfEntry->info);
flag1 = 1;
}
printf ("%s%s",flag2 == 1 ? "," : "",string (buffer));
flag2 = 1;
}
if (flag1 == 1) {
for (i = 0; i < arrayMax (currVcfEntry->genotypes); i++) {
currVcfGenotype = arrp (currVcfEntry->genotypes,i,VcfGenotype);
if (i == 0) {
printf ("\t%s\t",currVcfEntry->genotypeFormat);
}
printf ("%s%s%s%s",currVcfGenotype->genotype,
currVcfGenotype->details[0] != '\0' ? ":" : "",
currVcfGenotype->details[0] != '\0' ? currVcfGenotype->details : "",
i < arrayMax (currVcfEntry->genotypes) - 1 ? "\t" : "");
}
puts ("");
}
}
vcf_deInit ();
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);
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;
}
