void TopSegmentIteratorReverseTest::checkCallBack(AlignmentConstPtr alignment)
{
BottomSegmentIteratorConstPtr bi;
TopSegmentIteratorConstPtr ti, ti2;
const Genome* parent1 = alignment->openGenome("parent1");
const Genome* child1 = alignment->openGenome("child1");
ti = child1->getTopSegmentIterator();
bi = parent1->getBottomSegmentIterator();
ti2 = child1->getTopSegmentIterator();
ti2->toChild(bi, 0);
CuAssertTrue(_testCase, ti->getStartPosition() == 0);
CuAssertTrue(_testCase, ti->getLength() == 10);
CuAssertTrue(_testCase, ti->getReversed() == false);
CuAssertTrue(_testCase, ti2->getStartPosition() == 9);
CuAssertTrue(_testCase, ti2->getLength() == 10);
CuAssertTrue(_testCase, ti2->getReversed() == true);
bi->slice(1, 3);
ti2->toChild(bi, 0);
CuAssertTrue(_testCase, bi->getStartPosition() == 1);
CuAssertTrue(_testCase, bi->getLength() == 6);
CuAssertTrue(_testCase, ti2->getStartPosition() == 8);
CuAssertTrue(_testCase, ti2->getLength() == 6);
string buffer;
bi->getString(buffer);
CuAssertTrue(_testCase, buffer == "CCTACG");
ti2->getString(buffer);
CuAssertTrue(_testCase, buffer == "CACGTA");
bi = child1->getBottomSegmentIterator();
CuAssertTrue(_testCase, bi->getReversed() == false);
ti->toParseUp(bi);
CuAssertTrue(_testCase, ti->getStartPosition() == 0);
CuAssertTrue(_testCase, ti->getLength() == 5);
bi->toReverse();
ti->toParseUp(bi);
CuAssertTrue(_testCase, ti->getStartPosition() == 4);
CuAssertTrue(_testCase, ti->getLength() == 5);
bi->toReverse();
CuAssertTrue(_testCase, bi->getReversed() == false);
bi->toRight();
ti->toParseUp(bi);
CuAssertTrue(_testCase, ti->getStartPosition() == 5);
CuAssertTrue(_testCase, ti->getLength() == 5);
bi->toReverse();
ti->toParseUp(bi);
CuAssertTrue(_testCase, ti->getStartPosition() == 9);
CuAssertTrue(_testCase, ti->getLength() == 5);
}
void DefaultGappedBottomSegmentIterator::toRightNextUngapped(
BottomSegmentIteratorConstPtr bs) const
{
while (bs->hasChild(_childIndex) == false &&
bs->getLength() <= _gapThreshold)
{
if ((!bs->getReversed() && bs->getBottomSegment()->isLast()) ||
(bs->getReversed() && bs->getBottomSegment()->isFirst()))
{
break;
}
bs->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());
}
}
}
// quickly count subsitutions without loading rearrangement machinery.
// used for benchmarks for basic file scanning... and not much else since
// the interface is still a bit wonky.
void SummarizeMutations::substitutionAnalysis(const Genome* genome,
MutationsStats& stats)
{
assert(stats._subs == 0);
if (genome->getNumChildren() == 0 || genome->getNumBottomSegments() == 0 ||
(_targetSet && _targetSet->find(genome->getName()) == _targetSet->end()))
{
return;
}
const Genome* parent = genome->getParent();
string pname = parent != NULL ? parent->getName() : string();
StrPair branchName(genome->getName(), pname);
BottomSegmentIteratorConstPtr bottom = genome->getBottomSegmentIterator();
TopSegmentIteratorConstPtr top = genome->getChild(0)->getTopSegmentIterator();
string gString, cString;
hal_size_t n = genome->getNumBottomSegments();
vector<hal_size_t> children;
hal_size_t m = genome->getNumChildren();
for (hal_size_t i = 0; i < m; ++i)
{
string cName = genome->getChild(i)->getName();
if (!_targetSet ||
(_targetSet && _targetSet->find(cName) != _targetSet->end()))
{
children.push_back(i);
}
}
if (children.empty())
{
return;
}
for (hal_size_t i = 0; i < n; ++i)
{
bool readString = false;
for (size_t j = 0; j < children.size(); ++j)
{
if (bottom->hasChild(children[j]))
{
if (readString == false)
{
bottom->getString(gString);
readString = true;
}
top->toChild(bottom, children[j]);
top->getString(cString);
assert(gString.length() == cString.length());
for (hal_size_t k = 0; k < gString.length(); ++k)
{
if (isSubstitution(gString[k], cString[k]))
{
++stats._subs;
}
}
}
}
bottom->toRight();
}
}
