void hal::validateDuplications(const Genome* genome)
{
const Genome* parent = genome->getParent();
if (parent == NULL)
{
return;
}
TopSegmentIteratorConstPtr topIt = genome->getTopSegmentIterator();
TopSegmentIteratorConstPtr endIt = genome->getTopSegmentEndIterator();
vector<unsigned char> pcount(parent->getNumBottomSegments(), 0);
for (; topIt != endIt; topIt->toRight())
{
if (topIt->hasParent())
{
if (pcount[topIt->getTopSegment()->getParentIndex()] < 250)
{
++pcount[topIt->getTopSegment()->getParentIndex()];
}
}
}
for (topIt = genome->getTopSegmentIterator(); topIt != endIt; topIt->toRight())
{
if (topIt->hasParent())
{
size_t count = pcount[topIt->getTopSegment()->getParentIndex()];
assert(count > 0);
{
if (topIt->hasNextParalogy() == false && count > 1)
{
stringstream ss;
ss << "Top Segment " << topIt->getTopSegment()->getArrayIndex()
<< " in genome " << genome->getName() << " is not marked as a"
<< " duplication but it shares its parent "
<< topIt->getTopSegment()->getArrayIndex() << " with at least "
<< count - 1 << " other segments in the same genome";
throw hal_exception(ss.str());
}
}
}
}
}
void Genome::copyTopSegments(Genome *dest) const
{
const Genome *inParent = getParent();
const Genome *outParent = dest->getParent();
TopSegmentIteratorConstPtr inTop = getTopSegmentIterator();
TopSegmentIteratorPtr outTop = dest->getTopSegmentIterator();
hal_size_t n = dest->getNumTopSegments();
assert(n == 0 || n == getNumTopSegments());
if (n == 0) {
// Nothing to do if there are no top segments.
return;
}
BottomSegmentIteratorConstPtr inParentBottomSegIt = inParent->getBottomSegmentIterator();
BottomSegmentIteratorConstPtr outParentBottomSegIt = outParent->getBottomSegmentIterator();
for (; (hal_size_t)inTop->getArrayIndex() < n; inTop->toRight(),
outTop->toRight())
{
hal_index_t genomePos = inTop->getStartPosition();
assert(genomePos != NULL_INDEX);
string inSeqName = getSequenceBySite(genomePos)->getName();
string outSeqName = dest->getSequenceBySite(genomePos)->getName();
// if (inSeqName != outSeqName) {
// stringstream ss;
// ss << "When copying top segments from " << getName() << " to " << dest->getName() << ": sequence " << inSeqName << " != " << outSeqName << " at site " << genomePos;
// throw hal_exception(ss.str());
// }
outTop->setCoordinates(inTop->getStartPosition(), inTop->getLength());
outTop->setParentIndex(inTop->getParentIndex());
outTop->setParentReversed(inTop->getParentReversed());
outTop->setBottomParseIndex(inTop->getBottomParseIndex());
outTop->setNextParalogyIndex(inTop->getNextParalogyIndex());
// Check that the sequences from the bottom segments we point to are the same. If not, correct the indices so that they are.
if (inTop->getParentIndex() != NULL_INDEX) {
inParentBottomSegIt->toParent(inTop);
const Sequence *inParentSequence = inParentBottomSegIt->getSequence();
const Sequence *outParentSequence = outParent->getSequence(inParentSequence->getName());
hal_index_t inParentSegmentOffset = inTop->getParentIndex() - inParentSequence->getBottomSegmentArrayIndex();
hal_index_t outParentSegmentIndex = inParentSegmentOffset + outParentSequence->getBottomSegmentArrayIndex();
outTop->setParentIndex(outParentSegmentIndex);
}
}
}
void SummarizeMutations::subsAndGapInserts(
GappedTopSegmentIteratorConstPtr gappedTop, MutationsStats& stats)
{
assert(gappedTop->getReversed() == false);
hal_size_t numGaps = gappedTop->getNumGaps();
if (numGaps > 0)
{
stats._gapInsertionLength.add(gappedTop->getNumGapBases(), numGaps);
}
string parent, child;
TopSegmentIteratorConstPtr l = gappedTop->getLeft();
TopSegmentIteratorConstPtr r = gappedTop->getRight();
BottomSegmentIteratorConstPtr p =
l->getTopSegment()->getGenome()->getParent()->getBottomSegmentIterator();
for (TopSegmentIteratorConstPtr i = l->copy();
i->getTopSegment()->getArrayIndex() <=
r->getTopSegment()->getArrayIndex();
i->toRight())
{
if (i->hasParent())
{
p->toParent(i);
i->getString(child);
p->getString(parent);
assert(child.length() == parent.length());
for (size_t j = 0; j < child.length(); ++j)
{
if (isTransition(child[j], parent[j]))
{
++stats._transitions;
++stats._subs;
}
else if (isTransversion(child[j], parent[j]))
{
++stats._transversions;
++stats._subs;
}
else if (isSubstitution(child[j], parent[j]))
{
++stats._subs;
}
else if (!isMissingData(child[j]) && !isMissingData(parent[j]))
{
++stats._matches;
}
}
}
}
}
void DefaultGappedBottomSegmentIterator::toRightNextUngapped(
TopSegmentIteratorConstPtr ts) const
{
while (ts->hasParent() == false &&
ts->getLength() <= _gapThreshold)
{
if ((!ts->getReversed() && ts->getTopSegment()->isLast()) ||
(ts->getReversed() && ts->getTopSegment()->isFirst()))
{
break;
}
ts->toRight();
}
}
void hal::validateSequence(const Sequence* sequence)
{
// Verify that the DNA sequence doesn't contain funny characters
DNAIteratorConstPtr dnaIt = sequence->getDNAIterator();
hal_size_t length = sequence->getSequenceLength();
for (hal_size_t i = 0; i < length; ++i)
{
char c = dnaIt->getChar();
if (isNucleotide(c) == false)
{
stringstream ss;
ss << "Non-nucleotide character discoverd at position "
<< i << " of sequence " << sequence->getName() << ": " << c;
throw hal_exception(ss.str());
}
}
// Check the top segments
if (sequence->getGenome()->getParent() != NULL)
{
hal_size_t totalTopLength = 0;
TopSegmentIteratorConstPtr topIt = sequence->getTopSegmentIterator();
hal_size_t numTopSegments = sequence->getNumTopSegments();
for (hal_size_t i = 0; i < numTopSegments; ++i)
{
const TopSegment* topSegment = topIt->getTopSegment();
validateTopSegment(topSegment);
totalTopLength += topSegment->getLength();
topIt->toRight();
}
if (totalTopLength != length)
{
stringstream ss;
ss << "Sequence " << sequence->getName() << " has length " << length
<< " but its top segments add up to " << totalTopLength;
throw hal_exception(ss.str());
}
}
// Check the bottom segments
if (sequence->getGenome()->getNumChildren() > 0)
{
hal_size_t totalBottomLength = 0;
BottomSegmentIteratorConstPtr bottomIt =
sequence->getBottomSegmentIterator();
hal_size_t numBottomSegments = sequence->getNumBottomSegments();
for (hal_size_t i = 0; i < numBottomSegments; ++i)
{
const BottomSegment* bottomSegment = bottomIt->getBottomSegment();
validateBottomSegment(bottomSegment);
totalBottomLength += bottomSegment->getLength();
bottomIt->toRight();
}
if (totalBottomLength != length)
{
stringstream ss;
ss << "Sequence " << sequence->getName() << " has length " << length
<< " but its bottom segments add up to " << totalBottomLength;
throw hal_exception(ss.str());
}
}
}
void TopSegmentSimpleIteratorTest::checkCallBack(AlignmentConstPtr alignment)
{
const Genome* ancGenome = alignment->openGenome("Anc0");
CuAssertTrue(_testCase,
ancGenome->getNumTopSegments() == _topSegments.size());
TopSegmentIteratorConstPtr tsIt = ancGenome->getTopSegmentIterator(0);
for (size_t i = 0; i < ancGenome->getNumTopSegments(); ++i)
{
CuAssertTrue(_testCase,
(size_t)tsIt->getTopSegment()->getArrayIndex() == i);
_topSegments[i].compareTo(tsIt, _testCase);
tsIt->toRight();
}
tsIt = ancGenome->getTopSegmentIterator(
ancGenome->getNumTopSegments() - 1);
for (hal_index_t i = ancGenome->getNumTopSegments() - 1; i >= 0; --i)
{
CuAssertTrue(_testCase, tsIt->getTopSegment()->getArrayIndex() == i);
_topSegments[i].compareTo(tsIt, _testCase);
tsIt->toLeft();
}
tsIt = ancGenome->getTopSegmentIterator(0);
tsIt->slice(0, tsIt->getLength() - 1);
for (hal_index_t i = 0; i < (hal_index_t)ancGenome->getSequenceLength(); ++i)
{
CuAssertTrue(_testCase, tsIt->getLength() == 1);
CuAssertTrue(_testCase, tsIt->getStartPosition() == i);
tsIt->toRight(tsIt->getStartPosition() + 1);
}
tsIt = ancGenome->getTopSegmentIterator(
ancGenome->getNumTopSegments() - 1);
tsIt->slice(tsIt->getLength() - 1, 0);
for (hal_index_t i = ancGenome->getSequenceLength() - 1; i >= 0; --i)
{
CuAssertTrue(_testCase, tsIt->getLength() == 1);
CuAssertTrue(_testCase, tsIt->getStartPosition() == i);
tsIt->toLeft(tsIt->getStartPosition() - 1);
}
tsIt = ancGenome->getTopSegmentIterator(0);
tsIt->toReverse();
CuAssertTrue(_testCase, tsIt->getReversed() == true);
tsIt->slice(tsIt->getLength() - 1, 0);
for (hal_index_t i = 0; i < (hal_index_t)ancGenome->getSequenceLength(); ++i)
{
CuAssertTrue(_testCase, tsIt->getLength() == 1);
CuAssertTrue(_testCase, tsIt->getStartPosition() == i);
tsIt->toLeft(tsIt->getStartPosition() + 1);
}
tsIt = ancGenome->getTopSegmentIterator(
ancGenome->getNumTopSegments() - 1);
tsIt->toReverse();
tsIt->slice(0, tsIt->getLength() - 1);
for (hal_index_t i = ancGenome->getSequenceLength() - 1; i >= 0; --i)
{
CuAssertTrue(_testCase, tsIt->getLength() == 1);
CuAssertTrue(_testCase, tsIt->getStartPosition() == i);
tsIt->toRight(tsIt->getStartPosition() - 1);
}
}
