const Genome::Contig *Genome::getContigForRead(unsigned location, unsigned readLength, unsigned *extraBasesClippedBefore) const
{
const Contig *contig = getContigAtLocation(location);
//
// Sometimes, a read aligns before the beginning of a chromosome (imagine prepending a few bases to the read).
// In that case, we want to handle it by soft-clipping the bases off of the beginning of the read. We detect it
// here by looking to see if the aligned location plus the read length crosses a contig boundary. It also might
// happen that it is aligned before the first contig, in which case contig will be NULL.
//
if (NULL == contig || location + readLength > contig->beginningOffset + contig->length) {
//
// We should never align over the end of a chromosome, only before the beginning. So move this into the next
// chromosome.
//
contig = getNextContigAfterLocation(location);
_ASSERT(NULL != contig);
_ASSERT(contig->beginningOffset > location && contig->beginningOffset < location + readLength);
*extraBasesClippedBefore = contig->beginningOffset - location;
} else {
*extraBasesClippedBefore = 0;
}
return contig;
}
//
// Makes a copy of a Genome, but with only one of the sex chromosomes.
//
// The fate of the mitochondrion is that of the X chromosome.
//
Genome *
Genome::copy(bool copyX, bool copyY, bool copyM) const
{
Genome *newCopy = new Genome(getCountOfBases(),getCountOfBases(), chromosomePadding);
if (NULL == newCopy) {
WriteErrorMessage("Genome::copy: failed to allocate space for copy.\n");
return NULL;
}
const Genome::Contig *currentContig = NULL;
const Genome::Contig *nextContig = getContigAtLocation(0);
unsigned offsetInReference = 0;
while (offsetInReference < getCountOfBases()) {
if (NULL != nextContig && offsetInReference >= nextContig->beginningOffset) {
//
// Start of a new contig. See if we want to skip it.
//
currentContig = nextContig;
nextContig = getNextContigAfterLocation(offsetInReference + 1);
if ((!copyX && !strcmp(currentContig->name,"chrX")) ||
(!copyY && !strcmp(currentContig->name,"chrY")) ||
(!copyM && !strcmp(currentContig->name,"chrM"))) {
//
// Yes, skip over this contig.
//
nextContig = getNextContigAfterLocation(offsetInReference + 1);
if (NULL == nextContig) {
//
// The chromosome that we're skipping was the last one, so we're done.
//
break;
} else {
offsetInReference = nextContig->beginningOffset;
continue;
}
} // If skipping this chromosome
newCopy->startContig(currentContig->name);
} // If new contig beginning
const size_t maxCopySize = 10000;
char dataBuffer[maxCopySize + 1];
unsigned amountToCopy = maxCopySize;
if (nextContig && nextContig->beginningOffset < offsetInReference + amountToCopy) {
amountToCopy = nextContig->beginningOffset - offsetInReference;
}
if (getCountOfBases() < offsetInReference + amountToCopy) {
amountToCopy = getCountOfBases() - offsetInReference;
}
memcpy(dataBuffer,getSubstring(offsetInReference,amountToCopy), amountToCopy);
dataBuffer[amountToCopy] = '\0';
newCopy->addData(dataBuffer);
offsetInReference += amountToCopy;
}
newCopy->fillInContigLengths();
newCopy->sortContigsByName();
return newCopy;
}
int
Genome::getContigNumAtLocation(GenomeLocation location) const
{
const Contig *contig = getContigAtLocation(location);
return (int)(contig - contigs);
}
