int main (int argc, char *argv[])
{
MrfEntry *currEntry;
int mode;
if (argc != 3) {
usage ("%s <file.annotation> <include|exclude>",argv[0]);
}
intervalFind_addIntervalsToSearchSpace (argv[1],0);
if (strEqual (argv[2],"include")) {
mode = MODE_INCLUDE;
}
else if (strEqual (argv[2],"exclude")) {
mode = MODE_EXCLUDE;
}
else {
usage ("%s <file.annotation> <include|exclude>",argv[0]);
}
mrf_init ("-");
puts (mrf_writeHeader ());
while (currEntry = mrf_nextEntry ()) {
processEntry (currEntry,mode);
}
mrf_deInit ();
return 0;
}
int main (int argc, char *argv[])
{
int i,j;
Array intervals;
Interval *currInterval;
SubInterval *currSubInterval;
if (argc != 2) {
usage ("%s <trackName>",argv[0]);
}
intervalFind_addIntervalsToSearchSpace ("-",0);
intervals = intervalFind_getAllIntervals ();
puts ("browser hide all");
printf ("track name=\"%s\" visibility=2\n",argv[1]);
for (i = 0; i < arrayMax (intervals); i++) {
currInterval = arrp (intervals,i,Interval);
for (j = 0; j < arrayMax (currInterval->subIntervals); j++) {
currSubInterval = arrp (currInterval->subIntervals,j,SubInterval);
printf ("%s\tannotation\texon\t%d\t%d\t.\t%c\t.\tgroup%d\n",
currInterval->chromosome,
currSubInterval->start + 1,
currSubInterval->end,
currInterval->strand,
i);
}
}
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[])
{
int i,j;
Array intervals;
Interval *currInterval;
SubInterval *currSubInterval;
Stringa sizes =NULL;
Stringa starts=NULL;
if (argc < 2) {
usage ("%s <trackName> [simple]",argv[0]);
}
if( (argc==3) && !strEqual( argv[2],"simple") ) {
usage("%s <trackName> [simple]",argv[0]);
}
intervalFind_addIntervalsToSearchSpace ("-",0);
intervals = intervalFind_getAllIntervals ();
puts ("browser hide all");
printf ("track name=\"%s\" visibility=2\n",argv[1]);
for (i = 0; i < arrayMax (intervals); i++) {
currInterval = arrp (intervals,i,Interval);
if( argc == 3 ) {
for (j = 0; j < arrayMax (currInterval->subIntervals); j++) {
currSubInterval = arrp (currInterval->subIntervals,j,SubInterval);
printf ("%s\t%d\t%d\t%s_%d\t900\t%c\t%d\t%d\t.\t1\t%d\t0\n",
currInterval->chromosome,currSubInterval->start,currSubInterval->end,currInterval->name,j+1,currInterval->strand, currSubInterval->start, currSubInterval->end, currSubInterval->end - currSubInterval->start );
}
} else {
stringCreateClear( starts, 10);
stringCreateClear( sizes, 10);
for( j = 0; j < arrayMax (currInterval->subIntervals); j++) {
currSubInterval = arrp (currInterval->subIntervals,j,SubInterval);
stringAppendf( sizes, "%d", currSubInterval->end - currSubInterval->start );
stringAppendf( starts, "%d", currSubInterval->start - currInterval->start );
if( j<arrayMax( currInterval->subIntervals) ) {
stringAppendf( sizes, "," );
stringAppendf(starts, "," );
}
}
printf ("%s\t%d\t%d\t%s\t900\t%c\t%d\t%d\t.\t%d\t%s\t%s\n",
currInterval->chromosome,currInterval->start,currInterval->end, currInterval->name,currInterval->strand, currInterval->start, currInterval->end, currInterval->subIntervalCount, string(sizes), string(starts) );
}
}
return 0;
}
/**
* \file bgrQuantifier <annotation.interval>.
* \pre: it requires a BedGraph file from STDIN normalized by the number of mapped nucleotides
*/
int main( int argc, char* argv[] ) {
Array bgrs;
Array intervals;
Array entries;
int i, j, length;
double value;
if( argc < 2 ) {
usage("%s <annotation.interval>\n%s requires a BedGraph from STDIN", argv[0], argv[0]);
}
bgrs = arrayCreate( 1000, BedGraph );
bgrParser_initFromFile ( "-" );
bgrs = bgrParser_getAllEntries ();
bgrParser_deInit();
arraySort( bgrs, (ARRAYORDERF) bgrParser_sort );
intervalFind_addIntervalsToSearchSpace ( argv[1], 0 );
intervals = intervalFind_getAllIntervals ();
for( i=0; i<arrayMax(intervals); i++ ) {
Interval *currInterval = arrp( intervals, i, Interval );
length = currInterval->end - currInterval->start;
entries = bgrParser_getValuesForRegion( bgrs, currInterval->chromosome, currInterval->start, currInterval->end);
value = 0.0;
for( j=0; j<arrayMax( entries ); j++)
value += arru( entries, j, double );
printf("%s\t%s:%d-%d\t%f\n", currInterval->name,
currInterval->chromosome,
currInterval->start+1,
currInterval->end,
value /= length / 1000.0 );
arrayDestroy( entries );
}
arrayDestroy( intervals );
return 0;
}
int main (int argc, char *argv[])
{
Array intervals;
Interval *currInterval;
SubInterval *currSubInterval;
int refLength,altLength,offset;
int h,i,j;
Stringa buffer;
Array geneTranscriptEntries;
Texta geneTranscriptIds;
Array alterations;
Alteration *currAlteration,*nextAlteration;
int numTranscripts;
Stringa transcripts;
VcfEntry *currVcfEntry;
int position;
Texta alternateAlleles;
int flag1,flag2;
VcfGenotype *currVcfGenotype;
if (argc != 3) {
usage ("%s <annotation.interval> <nameFeature>",argv[0]);
}
intervalFind_addIntervalsToSearchSpace (argv[1],0);
geneTranscriptEntries = util_getGeneTranscriptEntries (intervalFind_getAllIntervals ());
buffer = stringCreate (100);
transcripts = stringCreate (100);
alterations = arrayCreate (100,Alteration);
vcf_init ("-");
stringPrintf (buffer,"##INFO=<ID=VA,Number=.,Type=String,Description=\"Variant Annotation, %s, %s\">",argv[1],argv[2]);
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));
offset = MAX (refLength,altLength) - 1;
util_clearAlterations (alterations);
intervals = intervalFind_getOverlappingIntervals (currVcfEntry->chromosome,position,position + offset);
for (i = 0; i < arrayMax (intervals); i++) {
currInterval = arru (intervals,i,Interval*);
j = 0;
while (j < arrayMax (currInterval->subIntervals)) {
currSubInterval = arrp (currInterval->subIntervals,j,SubInterval);
if (currSubInterval->start <= position && (position + offset) < currSubInterval->end) {
break;
}
j++;
}
if (j == arrayMax (currInterval->subIntervals)) {
continue;
}
util_addAlteration (arrayp (alterations,arrayMax (alterations),Alteration),currInterval->name,argv[2],currInterval,position,0);
}
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 (transcripts);
stringAppendf (transcripts,"%s:%s:%d_%d",currAlteration->transcriptName,currAlteration->transcriptId,currAlteration->transcriptLength,currAlteration->relativePosition);
numTranscripts = 1;
j = i + 1;
while (j < arrayMax (alterations)) {
nextAlteration = arrp (alterations,j,Alteration);
if (strEqual (currAlteration->geneId,nextAlteration->geneId) &&
strEqual (currAlteration->type,nextAlteration->type)) {
stringAppendf (transcripts,":%s:%s:%d_%d",nextAlteration->transcriptName,nextAlteration->transcriptId,nextAlteration->transcriptLength,nextAlteration->relativePosition);
numTranscripts++;
}
else {
break;
}
j++;
}
i = j;
geneTranscriptIds = util_getTranscriptIdsForGeneId (geneTranscriptEntries,currAlteration->geneId);
stringAppendf (buffer,":%d/%d:%s",numTranscripts,arrayMax (geneTranscriptIds),string (transcripts));
}
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;
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[])
{
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;
}
