C++ BottomSegmentIteratorPtr::getArrayIndex方法代码示例

// Set bottom segments to be equal width and so that segment 1, 2, 3,
// etc. corresponds to child segment 1, 2, 3, etc.
void setBottomSegments(Genome *genome, hal_size_t width) {
    hal_size_t numChildren = genome->getNumChildren();
    BottomSegmentIteratorPtr bottomIt = genome->getBottomSegmentIterator();
    hal_size_t n = genome->getNumBottomSegments();
    hal_index_t startPos = 0;
    for (; bottomIt->getArrayIndex() < n; bottomIt->toRight(), startPos += width) {
        for (hal_size_t i = 0; i < numChildren; i++) {
            bottomIt->setCoordinates(startPos, width);
            bottomIt->bseg()->setChildIndex(i, bottomIt->getArrayIndex());
            bottomIt->bseg()->setChildReversed(i, false);
            bottomIt->bseg()->setTopParseIndex(NULL_INDEX);
        }
    }
}

void checkBottomSegments(Genome *genome, hal_size_t width, CuTest *testCase) {
    hal_size_t numChildren = genome->getNumChildren();
    BottomSegmentIteratorPtr bottomIt = genome->getBottomSegmentIterator();
    hal_size_t n = genome->getNumBottomSegments();
    hal_index_t startPos = 0;
    for (; bottomIt->getArrayIndex() < n; bottomIt->toRight(), startPos += width) {
        CuAssertTrue(testCase, bottomIt->getStartPosition() == startPos);

        for (hal_size_t i = 0; i < numChildren; i++) {
            if (startPos < 170) {
                CuAssertStrEquals(testCase, "Sequence2", genome->getSequenceBySite(startPos)->getName().c_str());
                CuAssertTrue(testCase, bottomIt->bseg()->getChildIndex(i) == 13 + bottomIt->getArrayIndex());
            } else {
                CuAssertStrEquals(testCase, "Sequence1", genome->getSequenceBySite(startPos)->getName().c_str());
                CuAssertTrue(testCase, bottomIt->bseg()->getChildIndex(i) == bottomIt->getArrayIndex() - 17);
            }
        }
    }
}

int MappedSegment::boundComp(const SegmentIteratorPtr &s1, const SegmentIteratorPtr &s2) {
    int res = 0;
    bool flip = s2->getReversed();
    if (flip) {
        s2->toReverse();
    }

    if (s1->isTop() && !s2->isTop()) {
        BottomSegmentIteratorPtr bot = std::dynamic_pointer_cast<BottomSegmentIterator>(s2);
        hal_index_t lb = bot->bseg()->getTopParseIndex();
        hal_index_t ub = lb;
        if ((hal_size_t)bot->getArrayIndex() < bot->getGenome()->getNumBottomSegments() - 1) {
            bot = bot->clone();
            bot->slice(0, 0);
            bot->toRight();
            ub = bot->bseg()->getTopParseIndex();
        }
        if (s1->getArrayIndex() < lb) {
            res = -1;
        } else if (s1->getArrayIndex() > ub) {
            res = 1;
        }
    } else if (!s1->isTop() && s2->isTop()) {
        TopSegmentIteratorPtr top = std::dynamic_pointer_cast<TopSegmentIterator>(s2);
        hal_index_t lb = top->tseg()->getBottomParseIndex();
        hal_index_t ub = lb;
        if ((hal_size_t)top->getArrayIndex() < top->getGenome()->getNumTopSegments() - 1) {
            top = top->clone();
            top->slice(0, 0);
            top->toRight();
            ub = top->tseg()->getBottomParseIndex();
        }
        if (s1->getArrayIndex() < lb) {
            res = -1;
        } else if (s1->getArrayIndex() > ub) {
            res = 1;
        }
    }

    if (flip) {
        s2->toReverse();
    }

    return res;
}

void GenomeCopyTest::checkCallBack(const Alignment *alignment) {
    // FIXME: halAlignment->open() fails miserably but
    // openHalAlignmentReadOnly works? Probably some state isn't cleared
    // on close.
    AlignmentPtr tmp(getTestAlignmentInstances(alignment->getStorageFormat(), _path, WRITE_ACCESS));
    _secondAlignment = tmp;
    const Genome *ancGenome = alignment->openGenome("AncGenome");
    CuAssertTrue(_testCase, ancGenome->getName() == "AncGenome");
    CuAssertTrue(_testCase, ancGenome->getSequenceLength() == 1000000);
    CuAssertTrue(_testCase, ancGenome->getNumTopSegments() == 0);
    CuAssertTrue(_testCase, ancGenome->getNumBottomSegments() == 700000);
    const MetaData *ancMeta = ancGenome->getMetaData();
    CuAssertTrue(_testCase, ancMeta->get("Young") == "Jeezy");
    const Genome *leafGenome = alignment->openGenome("LeafGenome1");
    string ancSeq = "CAT";
    hal_index_t n = ancGenome->getSequenceLength();
    DnaIteratorPtr dnaIt = ancGenome->getDnaIterator();
    for (; dnaIt->getArrayIndex() < n; dnaIt->toRight()) {
        size_t i = dnaIt->getArrayIndex() % ancSeq.size();
        CuAssertTrue(_testCase, dnaIt->getBase() == ancSeq[i]);
    }
    TopSegmentIteratorPtr topIt = leafGenome->getTopSegmentIterator();
    n = leafGenome->getNumTopSegments();
    for (; topIt->getArrayIndex() < n; topIt->toRight()) {
        CuAssertTrue(_testCase, topIt->getStartPosition() == topIt->getArrayIndex());
        CuAssertTrue(_testCase, topIt->getLength() == 1);
        CuAssertTrue(_testCase, topIt->tseg()->getParentIndex() == 3);
        CuAssertTrue(_testCase, topIt->tseg()->getParentReversed() == true);
        CuAssertTrue(_testCase, topIt->tseg()->getBottomParseIndex() == 5);
        if (topIt->getArrayIndex() != 6) {
            CuAssertTrue(_testCase, topIt->tseg()->getNextParalogyIndex() == 6);
        } else {
            CuAssertTrue(_testCase, topIt->tseg()->getNextParalogyIndex() == 7);
        }
    }
    BottomSegmentIteratorPtr botIt = ancGenome->getBottomSegmentIterator();
    n = ancGenome->getNumBottomSegments();
    for (; botIt->getArrayIndex() < n; botIt->toRight()) {
        CuAssertTrue(_testCase, botIt->getStartPosition() == botIt->getArrayIndex());
        CuAssertTrue(_testCase, botIt->getLength() == 1);
        CuAssertTrue(_testCase, botIt->bseg()->getChildIndex(0) == 3);
        CuAssertTrue(_testCase, botIt->bseg()->getChildReversed(0) == true);
        CuAssertTrue(_testCase, botIt->bseg()->getTopParseIndex() == 5);
    }

    const Genome *copyRootGenome = _secondAlignment->openGenome("copyRootGenome");
    const Genome *copyLeafGenome = _secondAlignment->openGenome("LeafGenome1");
    CuAssertTrue(_testCase, copyRootGenome->getName() == "copyRootGenome");
    CuAssertTrue(_testCase, copyRootGenome->getSequenceLength() == 1000000);
    CuAssertTrue(_testCase, copyRootGenome->getNumTopSegments() == 0);
    CuAssertTrue(_testCase, copyRootGenome->getNumBottomSegments() == 700000);
    CuAssertTrue(_testCase, copyLeafGenome->getName() == "LeafGenome1");
    CuAssertTrue(_testCase, copyLeafGenome->getSequenceLength() == 1000000);
    CuAssertTrue(_testCase, copyLeafGenome->getNumTopSegments() == 5000);
    CuAssertTrue(_testCase, copyLeafGenome->getNumBottomSegments() == 0);
    const MetaData *copyMeta = copyRootGenome->getMetaData();
    CuAssertTrue(_testCase, copyMeta->get("Young") == "Jeezy");
    n = copyRootGenome->getSequenceLength();
    dnaIt = copyRootGenome->getDnaIterator();
    for (; dnaIt->getArrayIndex() < n; dnaIt->toRight()) {
        size_t i = dnaIt->getArrayIndex() % ancSeq.size();
        CuAssertTrue(_testCase, dnaIt->getBase() == ancSeq[i]);
    }
    topIt = copyLeafGenome->getTopSegmentIterator();
    n = copyLeafGenome->getNumTopSegments();
    for (; topIt->getArrayIndex() < n; topIt->toRight()) {
        CuAssertTrue(_testCase, topIt->getStartPosition() == topIt->getArrayIndex());
        CuAssertTrue(_testCase, topIt->getLength() == 1);
        CuAssertTrue(_testCase, topIt->tseg()->getParentIndex() == 3);
        CuAssertTrue(_testCase, topIt->tseg()->getParentReversed() == true);
        CuAssertTrue(_testCase, topIt->tseg()->getBottomParseIndex() == 5);
        if (topIt->getArrayIndex() != 6) {
            CuAssertTrue(_testCase, topIt->tseg()->getNextParalogyIndex() == 6);
        } else {
            CuAssertTrue(_testCase, topIt->tseg()->getNextParalogyIndex() == 7);
        }
    }
    botIt = copyRootGenome->getBottomSegmentIterator();
    n = copyRootGenome->getNumBottomSegments();
    for (; botIt->getArrayIndex() < n; botIt->toRight()) {
        CuAssertTrue(_testCase, botIt->getStartPosition() == botIt->getArrayIndex());
        CuAssertTrue(_testCase, botIt->getLength() == 1);
        CuAssertTrue(_testCase, botIt->bseg()->getChildIndex(0) == 3);
        CuAssertTrue(_testCase, botIt->bseg()->getChildReversed(0) == true);
        CuAssertTrue(_testCase, botIt->bseg()->getTopParseIndex() == 5);
    }

    _secondAlignment->close();
    remove(_path.c_str());
}

void GenomeCopyTest::createCallBack(Alignment *alignment) {
    hal_size_t alignmentSize = alignment->getNumGenomes();
    CuAssertTrue(_testCase, alignmentSize == 0);

    // Hacky: Need a different alignment to test copying the bottom
    // segments correctly.  (the names of a node's children are used
    // when copying bottom segments, and two genomes can't have the same
    // name in the same alignment)
    _path = getTempFile();
    _secondAlignment =
        AlignmentPtr(getTestAlignmentInstances(alignment->getStorageFormat(), _path, WRITE_ACCESS | CREATE_ACCESS));

    Genome *ancGenome = alignment->addRootGenome("AncGenome", 0);
    Genome *leafGenome = alignment->addLeafGenome("LeafGenome1", "AncGenome", 0);
    // This genome will test copyDimensions, copyTopSegments,
    // copyBottomSegments, copySequence, copyMetadata
    Genome *copyRootGenome = _secondAlignment->addRootGenome("copyRootGenome", 0);
    Genome *copyLeafGenome = _secondAlignment->addLeafGenome("LeafGenome1", "copyRootGenome", 0);

    MetaData *ancMeta = ancGenome->getMetaData();
    ancMeta->set("Young", "Jeezy");

    vector<Sequence::Info> seqVec(1);
    seqVec[0] = Sequence::Info("Sequence", 1000000, 0, 700000);
    ancGenome->setDimensions(seqVec);
    seqVec[0] = Sequence::Info("Sequence", 1000000, 5000, 0);
    leafGenome->setDimensions(seqVec);
    string ancSeq = "CAT";
    hal_index_t n = ancGenome->getSequenceLength();
    DnaIteratorPtr dnaIt = ancGenome->getDnaIterator();
    for (; dnaIt->getArrayIndex() < n; dnaIt->toRight()) {
        size_t i = dnaIt->getArrayIndex() % ancSeq.size();
        dnaIt->setBase(ancSeq[i]);
    }
    dnaIt->flush();

    n = leafGenome->getSequenceLength();
    dnaIt = leafGenome->getDnaIterator();
    for (; dnaIt->getArrayIndex() < n; dnaIt->toRight()) {
        size_t i = dnaIt->getArrayIndex() % ancSeq.size();
        dnaIt->setBase(ancSeq[i]);
    }
    dnaIt->flush();

    TopSegmentIteratorPtr topIt = leafGenome->getTopSegmentIterator();
    n = leafGenome->getNumTopSegments();
    for (; topIt->getArrayIndex() < n; topIt->toRight()) {
        topIt->setCoordinates(topIt->getArrayIndex(), 1);
        topIt->tseg()->setParentIndex(3);
        topIt->tseg()->setParentReversed(true);
        topIt->tseg()->setBottomParseIndex(5);
        if (topIt->getArrayIndex() != 6) {
            topIt->tseg()->setNextParalogyIndex(6);
        } else {
            topIt->tseg()->setNextParalogyIndex(7);
        }
    }
    BottomSegmentIteratorPtr botIt = ancGenome->getBottomSegmentIterator();
    n = ancGenome->getNumBottomSegments();
    for (; botIt->getArrayIndex() < n; botIt->toRight()) {
        botIt->setCoordinates(botIt->getArrayIndex(), 1);
        botIt->bseg()->setChildIndex(0, 3);
        botIt->bseg()->setChildReversed(0, true);
        botIt->bseg()->setTopParseIndex(5);
    }

    seqVec[0] = Sequence::Info("Sequence", 3300, 0, 1100);
    copyRootGenome->setDimensions(seqVec);
    seqVec[0] = Sequence::Info("Sequence", 3300, 2200, 0);
    copyLeafGenome->setDimensions(seqVec);
    string copySeq = "TAG";
    dnaIt = copyRootGenome->getDnaIterator();
    n = copyRootGenome->getSequenceLength();
    for (; dnaIt->getArrayIndex() < n; dnaIt->toRight()) {
        size_t i = dnaIt->getArrayIndex() % copySeq.size();
        dnaIt->setBase(copySeq[i]);
    }
    dnaIt->flush();

    dnaIt = copyLeafGenome->getDnaIterator();
    n = copyLeafGenome->getSequenceLength();
    for (; dnaIt->getArrayIndex() < n; dnaIt->toRight()) {
        size_t i = dnaIt->getArrayIndex() % copySeq.size();
        dnaIt->setBase(copySeq[i]);
    }
    dnaIt->flush();

    topIt = copyLeafGenome->getTopSegmentIterator();
    n = copyLeafGenome->getNumTopSegments();
    for (; topIt->getArrayIndex() < n; topIt->toRight()) {
        topIt->setCoordinates(7, 8);
        topIt->tseg()->setParentIndex(9);
        topIt->tseg()->setParentReversed(false);
        topIt->tseg()->setBottomParseIndex(11);
        if (topIt->getArrayIndex() != 12) {
            topIt->tseg()->setNextParalogyIndex(12);
        } else {
            topIt->tseg()->setNextParalogyIndex(7);
        }
    }
//.........这里部分代码省略.........

void LodExtract::updateBlockEdges(const Genome* inParentGenome,
                                  SegmentMap& segMap,
                                  const LodBlock* block,
                                  BottomSegmentIteratorPtr bottom,
                                  TopSegmentIteratorPtr top)
{
  Genome* outParentGenome = bottom->getGenome();
  const LodSegment* rootSeg = NULL;
  SegmentSet* segSet;
  SegmentSet::iterator setIt;

  // Zap all segments in parent genome
  SegmentMap::iterator mapIt = segMap.find(inParentGenome);
  if (mapIt != segMap.end())
  {
    segSet = mapIt->second;
    assert(segSet != NULL);
    setIt = segSet->begin();
    for (; setIt != segSet->end(); ++setIt)
    {
      bottom->setArrayIndex(outParentGenome, (*setIt)->getArrayIndex());
      for (hal_size_t i = 0; i < bottom->getNumChildren(); ++i)
      {
        bottom->setChildIndex(i, NULL_INDEX);
        bottom->setTopParseIndex(NULL_INDEX);
      }
    }

    // Choose first segment as parent to all segments in the child genome
    setIt = segSet->begin();
    rootSeg = *(setIt);
    bottom->setArrayIndex(outParentGenome, (*setIt)->getArrayIndex());
  }

  // Do the child genomes
  const Genome* inGrandParentGenome = inParentGenome->getParent();
  SegmentSet::iterator nextIt;
  for (mapIt = segMap.begin(); mapIt != segMap.end(); ++mapIt)
  {
    if (mapIt->first != inParentGenome and mapIt->first != inGrandParentGenome)
    {
      Genome* outChildGenome = 
         _outAlignment->openGenome(mapIt->first->getName());
      hal_index_t childIndex = outParentGenome->getChildIndex(outChildGenome);
      assert(childIndex >= 0);
      segSet = mapIt->second;
      assert(segSet != NULL);
      for (setIt = segSet->begin(); setIt != segSet->end(); ++setIt)
      {
        top->setArrayIndex(outChildGenome, (*setIt)->getArrayIndex());
        top->setBottomParseIndex(NULL_INDEX);

        // Connect to parent
        if (rootSeg != NULL)
        {
          top->setParentIndex(bottom->getArrayIndex());
          bool reversed = (*setIt)->getFlipped() != rootSeg->getFlipped();
          top->setParentReversed(reversed);
          if (setIt == segSet->begin())
          {
            bottom->setChildIndex(childIndex, top->getArrayIndex());         
            bottom->setChildReversed(childIndex, reversed);      
          }
        }
        else
        {
          top->setParentIndex(NULL_INDEX);
        }

        // Connect to next paralogy
        SegmentSet::iterator setNext = setIt;
        ++setNext;
        if (setNext == segSet->end())
        {
          setNext = segSet->begin();
        }
        if (setNext == setIt)
        {
          top->setNextParalogyIndex(NULL_INDEX);
        }
        else
        {
          top->setNextParalogyIndex((*setNext)->getArrayIndex());
        }
      }
    }
  }
}

void Genome::copyBottomSegments(Genome *dest) const
{
  assert(getNumBottomSegments() == dest->getNumBottomSegments());
  hal_size_t inNc = getNumChildren();
  hal_size_t outNc = dest->getNumChildren();
  // The child indices aren't consistent across files--make sure each bottom
  // segment points to the correct children
  vector<string> inChildNames;
  vector<string> outChildNames;
  for (hal_size_t inChild = 0; inChild < inNc; ++inChild)
  {
    inChildNames.push_back(getChild(inChild)->getName());
  }
  for (hal_size_t outChild = 0; outChild < outNc; ++outChild)
  {
    outChildNames.push_back(dest->getChild(outChild)->getName());
  }
  map<hal_size_t, hal_size_t> inChildToOutChild;
  for (hal_size_t inChild = 0; inChild < inNc; inChild++)
  {
    hal_size_t outChild;
    for (outChild = 0; outChild < outNc; outChild++)
    {
      if (inChildNames[inChild] == outChildNames[outChild])
      {
        inChildToOutChild[inChild] = outChild;
        break;
      }
    }
    if (outChild == outNc)
    {
      inChildToOutChild[inChild] = outNc;
    }
  }

  // Go through each sequence in this genome, find the matching
  // sequence in the dest genome, then copy over the segments for each
  // sequence.
  SequenceIteratorConstPtr seqIt = getSequenceIterator();
  SequenceIteratorConstPtr seqEndIt = getSequenceEndIterator();

  for (; seqIt != seqEndIt; seqIt->toNext())
  {
    const Sequence *inSeq = seqIt->getSequence();
    const Sequence *outSeq = dest->getSequence(inSeq->getName());
    BottomSegmentIteratorPtr inBot = inSeq->getBottomSegmentIterator();
    BottomSegmentIteratorPtr outBot = outSeq->getBottomSegmentIterator();

    cout << "DEBUG: inSeq name: " << inSeq->getName() << ", outSeq name: " << outSeq->getName() << endl;

    if (inSeq->getName() != outSeq->getName()) {
      // This check is important enough that it can't be an assert.
      stringstream ss;
      ss << "When copying bottom segments: segment #" << inBot->getArrayIndex() << " of source genome is from sequence " << inBot->getSequence()->getName() << ", but segment #" << outBot->getArrayIndex() << " is from sequence " << outBot->getSequence()->getName();
      throw hal_exception(ss.str());
    }

    if (inSeq->getNumBottomSegments() != outSeq->getNumBottomSegments()) {
      stringstream ss;
      ss << "When copying bottom segments: sequence " << inSeq->getName() << " has " << inSeq->getNumBottomSegments() << " in genome " << getName() << ", while it has " << outSeq->getNumBottomSegments() << " in genome " << dest->getName();
      throw hal_exception(ss.str());      
    }

    hal_index_t inSegmentEnd = inSeq->getBottomSegmentArrayIndex() + inSeq->getNumBottomSegments();
    cout << "DEBUG: inSegmentStart: " << inSeq->getBottomSegmentArrayIndex() << " inSegmentEnd: " << inSegmentEnd << " num bottom segments: " << inSeq->getNumBottomSegments() << endl;
    for (; inBot->getArrayIndex() < inSegmentEnd; inBot->toRight(),
           outBot->toRight())
    {
      hal_index_t outStartPosition = inBot->getStartPosition() - inSeq->getStartPosition() + outSeq->getStartPosition();

      cout << "Decided on outStartPosition " << outStartPosition << " for seg index " << outBot->getArrayIndex() << " (src index " << inBot->getArrayIndex() << ")" << endl;

      if (dest->getSequenceBySite(outStartPosition) != outSeq) {
        stringstream ss;
        ss << "When copying bottom segments from " << getName() << " to " << dest->getName() << ": expected destination sequence " << outSeq->getName() << " for segment # " << inBot->getArrayIndex() << " but got " << dest->getSequenceBySite(outStartPosition)->getName();
        throw hal_exception(ss.str());
      }
      outBot->setCoordinates(outStartPosition, inBot->getLength());
      for(hal_size_t inChild = 0; inChild < inNc; inChild++) {
        hal_size_t outChild = inChildToOutChild[inChild];
        if (outChild != outNc) {
          outBot->setChildIndex(outChild, inBot->getChildIndex(inChild));
          cout << "genome " << getName() << ": Set child index " << inChild << " to " << inBot->getChildIndex(inChild) << endl;
          outBot->setChildReversed(outChild, inBot->getChildReversed(inChild));
        }
      }
      outBot->setTopParseIndex(inBot->getTopParseIndex());
    }
  }
}