static void debugScaffoldPathsP(Cap *cap, stList *haplotypePath,
stHash *haplotypePathToScaffoldPathHash, stHash *haplotypeToMaximalHaplotypeLengthHash,
stHash *segmentToMaximalHaplotypePathHash, stList *haplotypeEventStrings, stList *contaminationEventStrings, CapCodeParameters *capCodeParameters, bool capDir) {
int64_t insertLength;
int64_t deleteLength;
Cap *otherCap;
enum CapCode capCode = getCapCode(cap, &otherCap, haplotypeEventStrings, contaminationEventStrings, &insertLength, &deleteLength, capCodeParameters);
if (capCode == SCAFFOLD_GAP || capCode == AMBIGUITY_GAP) {
Segment *adjacentSegment = getAdjacentCapsSegment(cap);
assert(adjacentSegment != NULL);
while (!hasCapInEvents(cap_getEnd(capDir ? segment_get5Cap(adjacentSegment) : segment_get3Cap(adjacentSegment)), haplotypeEventStrings)) {
adjacentSegment = getAdjacentCapsSegment(capDir ? segment_get5Cap(adjacentSegment) : segment_get3Cap(adjacentSegment));
assert(adjacentSegment != NULL);
}
assert(adjacentSegment != NULL);
assert(hasCapInEvents(cap_getEnd(segment_get5Cap(adjacentSegment)), haplotypeEventStrings)); //isHaplotypeEnd(cap_getEnd(segment_get5Cap(adjacentSegment))));
stIntTuple *j = stHash_search(haplotypeToMaximalHaplotypeLengthHash, haplotypePath);
(void)j;
assert(j != NULL);
stList *adjacentHaplotypePath = stHash_search(segmentToMaximalHaplotypePathHash, adjacentSegment);
if (adjacentHaplotypePath == NULL) {
adjacentHaplotypePath = stHash_search(segmentToMaximalHaplotypePathHash,
segment_getReverse(adjacentSegment));
}
assert(adjacentHaplotypePath != NULL);
assert(adjacentHaplotypePath != haplotypePath);
stIntTuple *k = stHash_search(haplotypeToMaximalHaplotypeLengthHash, adjacentHaplotypePath);
(void)k;
assert(k != NULL);
assert(stIntTuple_get(j, 0) == stIntTuple_get(k, 0));
assert(stHash_search(haplotypePathToScaffoldPathHash, haplotypePath) ==
stHash_search(haplotypePathToScaffoldPathHash, adjacentHaplotypePath));
}
}
void testCap_getSegment(CuTest* testCase) {
cactusCapTestSetup();
Block *block = block_construct(2, flower);
Segment *segment = segment_construct(block, rootEvent);
CuAssertTrue(testCase, cap_getSegment(segment_get5Cap(segment)) == segment);
CuAssertTrue(testCase, cap_getSegment(segment_get3Cap(segment)) == segment);
CuAssertTrue(testCase, cap_getOrientation(segment_get5Cap(segment)) == segment_getOrientation(segment));
CuAssertTrue(testCase, cap_getOrientation(segment_get3Cap(segment)) == segment_getOrientation(segment));
CuAssertTrue(testCase, cap_getSegment(cap_getReverse(segment_get5Cap(segment))) == segment_getReverse(segment));
CuAssertTrue(testCase, cap_getSegment(cap_getReverse(segment_get3Cap(segment))) == segment_getReverse(segment));
cactusCapTestTeardown();
}
void testCap_segmentCoordinatesReverseStrand(CuTest* testCase) {
/*
* Tests the coordinates of an segment and its 5 and 3 prime caps.
*/
cactusCapTestSetup();
Block *block = block_construct(3, flower);
Segment *segment = segment_construct2(block, 2, 0, sequence);
Cap *_5Cap = segment_get5Cap(segment);
Cap *_3Cap = segment_get3Cap(segment);
CuAssertTrue(testCase, cap_getSide(_5Cap));
CuAssertTrue(testCase, !cap_getSide(_3Cap));
CuAssertTrue(testCase, !cap_getStrand(_5Cap));
CuAssertTrue(testCase, !cap_getStrand(_3Cap));
CuAssertIntEquals(testCase, 4, cap_getCoordinate(_5Cap));
CuAssertIntEquals(testCase, 2, cap_getCoordinate(_3Cap));
CuAssertTrue(testCase, !segment_getStrand(segment));
CuAssertIntEquals(testCase, 4, segment_getStart(segment));
CuAssertIntEquals(testCase, 3, segment_getLength(segment));
CuAssertTrue(testCase, !cap_getSide(cap_getReverse(_5Cap)));
CuAssertTrue(testCase, cap_getSide(cap_getReverse(_3Cap)));
CuAssertTrue(testCase, cap_getStrand(cap_getReverse(_5Cap)));
CuAssertTrue(testCase, cap_getStrand(cap_getReverse(_3Cap)));
CuAssertIntEquals(testCase, 4, cap_getCoordinate(cap_getReverse(_5Cap)));
CuAssertIntEquals(testCase, 2, cap_getCoordinate(cap_getReverse(_3Cap)));
CuAssertTrue(testCase, segment_getStrand(segment_getReverse(segment)));
CuAssertIntEquals(testCase, 2, segment_getStart(segment_getReverse(segment)));
CuAssertIntEquals(testCase, 3, segment_getLength(segment_getReverse(segment)));
cactusCapTestTeardown();
}
void block_setRootInstance(Block *block, Segment *segment) {
block = block_getPositiveOrientation(block);
segment = segment_getPositiveOrientation(segment);
assert(block_getInstance(block, segment_getName(segment)) == segment);
end_setRootInstance(block_get5End(block), segment_get5Cap(segment));
end_setRootInstance(block_get3End(block), segment_get3Cap(segment));
}
static void getMaximalHaplotypePathsP2(Segment *segment,
stList *maximalHaplotypePath, stSortedSet *segmentSet, stList *eventStrings) {
/*
* Iterate all the way to one end of the contig then start the traversal to define the maximal
* haplotype path.
*/
Cap *_5Cap = segment_get5Cap(segment);
assert(hasCapInEvents(cap_getEnd(segment_get3Cap(segment)), eventStrings)); //isHaplotypeEnd(cap_getEnd(segment_get3Cap(segment))));
if (trueAdjacency(_5Cap, eventStrings)) { //Check that the adjacency is supported by a haplotype path
Segment *otherSegment = getAdjacentCapsSegment(_5Cap);
assert(segment != otherSegment);
assert(segment_getReverse(segment) != otherSegment);
if (otherSegment != NULL) {
assert(stSortedSet_search(segmentSet, otherSegment) == NULL);
assert(stSortedSet_search(segmentSet, segment_getReverse(
otherSegment)) == NULL);
assert(hasCapInEvents(cap_getEnd(segment_get3Cap(otherSegment)), eventStrings)); //isHaplotypeEnd(cap_getEnd(segment_get3Cap(otherSegment))));
getMaximalHaplotypePathsP2(otherSegment, maximalHaplotypePath,
segmentSet, eventStrings);
} else { //We need to start the maximal haplotype recursion
getMaximalHaplotypePathsP3(segment, maximalHaplotypePath,
segmentSet, eventStrings);
}
} else {
getMaximalHaplotypePathsP3(segment, maximalHaplotypePath, segmentSet, eventStrings);
}
}
void printPositions(stList *positions, const char *substitutionType, FILE *fileHandle) {
for (int64_t i = 0; i < stList_length(positions); i++) {
SegmentHolder *segmentHolder = stList_get(positions, i);
int64_t j = segment_getStart(segmentHolder->segment);
if (segment_getStrand(segmentHolder->segment)) {
j += segmentHolder->offset;
assert(
cap_getCoordinate(segment_get5Cap(segmentHolder->segment)) == segment_getStart(
segmentHolder->segment));
assert(
segment_getStart(segmentHolder->segment) + segment_getLength(segmentHolder->segment) - 1
== cap_getCoordinate(segment_get3Cap(segmentHolder->segment)));
} else {
j -= segmentHolder->offset;
assert(
segment_getStart(segmentHolder->segment) - segment_getLength(segmentHolder->segment) + 1
== cap_getCoordinate(segment_get3Cap(segmentHolder->segment)));
}
fprintf(fileHandle, "%s: %s_%" PRIi64 " %" PRIi64 " %c %c %c\n", substitutionType,
event_getHeader(segment_getEvent(segmentHolder->segment)),
sequence_getLength(segment_getSequence(segmentHolder->segment)), j,
segmentHolder->base1, segmentHolder->base2, segmentHolder->base3);
getMAFBlock(segment_getBlock(segmentHolder->segment), fileHandle);
}
}
static void debugScaffoldPaths(stList *haplotypePaths, stHash *haplotypePathToScaffoldPathHash,
stHash *haplotypeToMaximalHaplotypeLengthHash, stList *haplotypeEventStrings, stList *contaminationEventStrings, CapCodeParameters *capCodeParameters) {
stHash *segmentToMaximalHaplotypePathHash = buildSegmentToContigPathHash(haplotypePaths);
for (int64_t i = 0; i < stList_length(haplotypePaths); i++) {
stList *haplotypePath = stList_get(haplotypePaths, i);
assert(stList_length(haplotypePath) > 0);
//Traversing from 5' end..
Segment *_5Segment = stList_get(haplotypePath, 0);
Segment *_3Segment = stList_get(haplotypePath, stList_length(haplotypePath) - 1);
assert(segment_getStrand(_5Segment) == segment_getStrand(_3Segment));
if (!segment_getStrand(_5Segment)) {
Segment *j = _5Segment;
_5Segment = segment_getReverse(_3Segment);
_3Segment = segment_getReverse(j);
}
assert(segment_getStrand(_5Segment));
assert(segment_getStrand(_3Segment));
Cap *_5Cap = segment_get5Cap(_5Segment);
Cap *_3Cap = segment_get3Cap(_3Segment);
if (getAdjacentCapsSegment(_5Cap) != NULL) {
assert(!trueAdjacency(_5Cap, haplotypeEventStrings));
}
if (getAdjacentCapsSegment(_3Cap) != NULL) {
assert(!trueAdjacency(_3Cap, haplotypeEventStrings));
}
debugScaffoldPathsP(_5Cap, haplotypePath,
haplotypePathToScaffoldPathHash, haplotypeToMaximalHaplotypeLengthHash,
segmentToMaximalHaplotypePathHash, haplotypeEventStrings, contaminationEventStrings, capCodeParameters, 1);
debugScaffoldPathsP(_3Cap, haplotypePath,
haplotypePathToScaffoldPathHash, haplotypeToMaximalHaplotypeLengthHash,
segmentToMaximalHaplotypePathHash, haplotypeEventStrings, contaminationEventStrings, capCodeParameters, 0);
}
stHash_destruct(segmentToMaximalHaplotypePathHash);
}
static void getMaximalHaplotypePathsCheck(Flower *flower,
stSortedSet *segmentSet, const char *eventString, stList *eventStrings) {
/*
* Do debug checks that the haplotypes paths are well formed.
*/
Flower_SegmentIterator *segmentIt = flower_getSegmentIterator(flower);
Segment *segment;
while ((segment = flower_getNextSegment(segmentIt)) != NULL) {
if (strcmp(event_getHeader(segment_getEvent(segment)), eventString) == 0) {
if (hasCapInEvents(cap_getEnd(segment_get5Cap(segment)), eventStrings)) { //isHaplotypeEnd(cap_getEnd(segment_get5Cap(segment)))) {
assert(stSortedSet_search(segmentSet, segment) != NULL
|| stSortedSet_search(segmentSet, segment_getReverse(
segment)) != NULL);
}
}
}
flower_destructSegmentIterator(segmentIt);
Flower_GroupIterator *groupIt = flower_getGroupIterator(flower);
Group *group;
while ((group = flower_getNextGroup(groupIt)) != NULL) {
if (group_getNestedFlower(group) != NULL) {
getMaximalHaplotypePathsCheck(group_getNestedFlower(group),
segmentSet, eventString, eventStrings);
}
}
flower_destructGroupIterator(groupIt);
}
bool getCapGetAtEndOfPath(Cap *cap, Cap **pathEndCap,
int64_t *pathLength, int64_t *nCount, stList *haplotypeEventStrings, stList *contaminationEventStrings) {
//Account for length of adjacency
*pathLength += getTerminalAdjacencyLength(cap);
*nCount += getNumberOfNsInAdjacency(cap);
Segment *segment = getAdjacentCapsSegment(cap);
if (segment == NULL) {
*pathEndCap = cap_getAdjacency(getTerminalCap(cap));
assert(*pathEndCap != NULL);
return 0;
}
Cap *adjacentCap = cap_getSide(cap) ? segment_get3Cap(segment)
: segment_get5Cap(segment);
assert(
cap_getName(adjacentCap) == cap_getName(
cap_getAdjacency(getTerminalCap(cap))));
End *adjacentEnd = cap_getEnd(adjacentCap);
if (hasCapInEvents(adjacentEnd, contaminationEventStrings) || hasCapInEvents(adjacentEnd, haplotypeEventStrings)) { //hasCapNotInEvent(adjacentEnd, event_getHeader(cap_getEvent(cap)))) { //isContaminationEnd(adjacentEnd) || isHaplotypeEnd(adjacentEnd)) {
*pathEndCap = adjacentCap;
return 1;
}
*pathLength += segment_getLength(segment);
*nCount += getNumberOfNsInSegment(segment);
return getCapGetAtEndOfPath(cap_getOtherSegmentCap(adjacentCap),
pathEndCap, pathLength, nCount, haplotypeEventStrings, contaminationEventStrings);
}
void testBlock_getSetRootInstance(CuTest *testCase) {
cactusBlockTestSetup();
Block *block2 = block_construct(1, flower);
CuAssertTrue(testCase, block_getRootInstance(block2) == NULL);
block_destruct(block2);
//block_setRootInstance(block, segment_getReverse(rootSegment)); //set in the constructor code of the test.
CuAssertTrue(testCase, block_getRootInstance(block) == segment_getReverse(rootSegment));
CuAssertTrue(testCase, end_getRootInstance(block_get5End(block)) == segment_get5Cap(segment_getReverse(rootSegment)));
CuAssertTrue(testCase, end_getRootInstance(block_get3End(block)) == segment_get3Cap(segment_getReverse(rootSegment)));
cactusBlockTestTeardown();
}
bool linked(Segment *segmentX, Segment *segmentY, int64_t difference,
const char *eventString, bool *aligned) {
assert(segment_getStrand(segmentX));
assert(segment_getStrand(segmentY));
*aligned = 0;
if (segment_getStart(segmentX) < segment_getStart(segmentY)) {
Block *blockX = segment_getBlock(segmentX);
Block *blockY = segment_getBlock(segmentY);
Block_InstanceIterator *instanceItX = block_getInstanceIterator(blockX);
Segment *segmentX2;
while ((segmentX2 = block_getNext(instanceItX)) != NULL) {
if (strcmp(event_getHeader(segment_getEvent(segmentX2)),
eventString) == 0) {
Block_InstanceIterator *instanceItY =
block_getInstanceIterator(blockY);
Segment *segmentY2;
while ((segmentY2 = block_getNext(instanceItY)) != NULL) {
if (strcmp(event_getHeader(segment_getEvent(segmentY2)),
eventString) == 0) {
*aligned = 1;
if (sequence_getMetaSequence(
segment_getSequence(segmentX2))
== sequence_getMetaSequence(
segment_getSequence(segmentY2))) { //Have the same assembly sequence
//Now check if the two segments are connected by a path of adjacency from the 3' end of segmentX to the 5' end of segmentY.
int64_t separationDistance;
if (capsAreAdjacent(segment_get3Cap(segmentX2),
segment_get5Cap(segmentY2),
&separationDistance)) {
//if(difference < 10000 || (separationDistance <= difference * 1.5 && difference <= separationDistance * 1.5)) {
block_destructInstanceIterator(instanceItX);
block_destructInstanceIterator(instanceItY);
return 1;
//}
}
}
}
}
block_destructInstanceIterator(instanceItY);
}
}
block_destructInstanceIterator(instanceItX);
} else {
assert(segmentX == segmentY);
if(hasCapInEvent(block_get5End(segment_getBlock(segmentX)),
eventString)) {
*aligned = 1;
return 1;
}
}
return 0;
}
Segment *block_splitP(Segment *segment,
Block *leftBlock, Block *rightBlock) {
Segment *leftSegment = segment_getSequence(segment) != NULL
? segment_construct2(leftBlock, segment_getStart(segment), segment_getStrand(segment), segment_getSequence(segment))
: segment_construct(leftBlock, segment_getEvent(segment));
Segment *rightSegment = segment_getSequence(segment) != NULL
? segment_construct2(rightBlock, segment_getStart(segment) + block_getLength(leftBlock), segment_getStrand(segment), segment_getSequence(segment))
: segment_construct(rightBlock, segment_getEvent(segment));
//link together.
cap_makeAdjacent(segment_get3Cap(leftSegment), segment_get5Cap(rightSegment));
//update adjacencies.
Cap *_5Cap = segment_get5Cap(segment);
Cap *new5Cap = segment_get5Cap(leftSegment);
Cap *_3Cap = segment_get3Cap(segment);
Cap *new3Cap = segment_get3Cap(rightSegment);
if(cap_getAdjacency(_5Cap) != NULL) {
cap_makeAdjacent(cap_getAdjacency(_5Cap), new5Cap);
}
if(cap_getAdjacency(_3Cap) != NULL) {
cap_makeAdjacent(cap_getAdjacency(_3Cap), new3Cap);
}
return leftSegment;
}
void testCap_getOtherSegmentCap(CuTest *testCase) {
cactusCapTestSetup();
Block *block = block_construct(3, flower);
Segment *segment = segment_construct2(block, 2, 1, sequence);
Cap *_5Cap = segment_get5Cap(segment);
Cap *_3Cap = segment_get3Cap(segment);
CuAssertTrue(testCase, cap_getOtherSegmentCap(leaf1Cap) == NULL);
CuAssertTrue(testCase, cap_getOtherSegmentCap(cap_getReverse(leaf1Cap)) == NULL);
CuAssertTrue(testCase, cap_getOtherSegmentCap(_5Cap) == _3Cap);
CuAssertTrue(testCase, cap_getOtherSegmentCap(_3Cap) == _5Cap);
CuAssertTrue(testCase, cap_getOtherSegmentCap(cap_getReverse(_5Cap)) == cap_getReverse(_3Cap));
CuAssertTrue(testCase, cap_getOtherSegmentCap(cap_getReverse(_3Cap)) == cap_getReverse(_5Cap));
cactusCapTestTeardown();
}
static int64_t getBoundingNs(Cap *cap) {
assert(cap != NULL);
Segment *segment = getCapsSegment(cap);
if (segment == NULL) {
return 0;
}
Cap *_5TerminalCap = getTerminalCap(segment_get5Cap(segment));
Cap *_3TerminalCap = getTerminalCap(segment_get3Cap(segment));
(void)_3TerminalCap;
assert(_5TerminalCap != NULL);
assert(_3TerminalCap != NULL);
//return 0;
if (cap_getName(_5TerminalCap) == cap_getName(cap)) {
return getBoundingNsP(segment);
} else {
assert(cap_getName(_3TerminalCap) == cap_getName(cap));
return getBoundingNsP(segment_getReverse(segment));
}
}
static void getMaximalHaplotypePathsP(Flower *flower,
stList *maximalHaplotypePaths, stSortedSet *segmentSet,
const char *eventString,
stList *eventStrings) {
/*
* Iterate through the segments in this flower.
*/
Flower_SegmentIterator *segmentIt = flower_getSegmentIterator(flower);
Segment *segment;
while ((segment = flower_getNextSegment(segmentIt)) != NULL) {
if (stSortedSet_search(segmentSet, segment) == NULL
&& stSortedSet_search(segmentSet, segment_getReverse(segment))
== NULL) { //Check we haven't yet seen this segment
if (strcmp(event_getHeader(segment_getEvent(segment)), eventString)
== 0) { //Check if the segment is in the assembly
if (hasCapInEvents(cap_getEnd(segment_get5Cap(segment)), eventStrings)) { //Is a block in a haplotype segment
assert(hasCapInEvents(cap_getEnd(segment_get3Cap(segment)), eventStrings)); //isHaplotypeEnd(cap_getEnd(segment_get3Cap(segment))));
stList *maximalHaplotypePath = stList_construct();
stList_append(maximalHaplotypePaths, maximalHaplotypePath);
getMaximalHaplotypePathsP2(segment, maximalHaplotypePath,
segmentSet, eventStrings);
} else {
assert(!hasCapInEvents(cap_getEnd(segment_get3Cap(segment)), eventStrings));//assert(!isHaplotypeEnd(cap_getEnd(segment_get3Cap(segment))));
}
}
}
}
flower_destructSegmentIterator(segmentIt);
/*
* Now recurse on the contained flowers.
*/
Flower_GroupIterator *groupIt = flower_getGroupIterator(flower);
Group *group;
while ((group = flower_getNextGroup(groupIt)) != NULL) {
if (group_getNestedFlower(group) != NULL) {
getMaximalHaplotypePathsP(group_getNestedFlower(group),
maximalHaplotypePaths, segmentSet, eventString, eventStrings);
}
}
flower_destructGroupIterator(groupIt);
}
int64_t flower_getTotalBaseLength(Flower *flower) {
/*
* The implementation of this function is very like that in group_getTotalBaseLength, with a few differences. Consider merging them.
*/
Flower_EndIterator *endIterator = flower_getEndIterator(flower);
End *end;
int64_t totalLength = 0;
while ((end = flower_getNextEnd(endIterator)) != NULL) {
if (!end_isBlockEnd(end)) {
End_InstanceIterator *instanceIterator = end_getInstanceIterator(end);
Cap *cap;
while ((cap = end_getNext(instanceIterator)) != NULL) {
cap = cap_getStrand(cap) ? cap : cap_getReverse(cap);
if (!cap_getSide(cap) && cap_getSequence(cap) != NULL) {
Cap *cap2 = cap_getAdjacency(cap);
assert(cap2 != NULL);
while (end_isBlockEnd(cap_getEnd(cap2))) {
Segment *segment = cap_getSegment(cap2);
assert(segment != NULL);
assert(segment_get5Cap(segment) == cap2);
cap2 = cap_getAdjacency(segment_get3Cap(segment));
assert(cap2 != NULL);
assert(cap_getStrand(cap2));
assert(cap_getSide(cap2));
}
assert(cap_getStrand(cap2));
assert(cap_getSide(cap2));
int64_t length = cap_getCoordinate(cap2) - cap_getCoordinate(cap) - 1;
assert(length >= 0);
totalLength += length;
}
}
end_destructInstanceIterator(instanceIterator);
}
}
flower_destructEndIterator(endIterator);
return totalLength;
}
stList *getContigPaths(Flower *flower, const char *eventString, stList *eventStrings) {
stList *maximalHaplotypePaths = stList_construct3(0,
(void(*)(void *)) stList_destruct);
stSortedSet *segmentSet = stSortedSet_construct();
getMaximalHaplotypePathsP(flower, maximalHaplotypePaths, segmentSet, eventString, eventStrings);
//Do some debug checks..
st_logDebug("We have %" PRIi64 " maximal haplotype paths\n", stList_length(
maximalHaplotypePaths));
getMaximalHaplotypePathsCheck(flower, segmentSet, eventString, eventStrings);
for (int64_t i = 0; i < stList_length(maximalHaplotypePaths); i++) {
stList *maximalHaplotypePath = stList_get(maximalHaplotypePaths, i);
st_logDebug("We have a maximal haplotype path with length %" PRIi64 "\n",
stList_length(maximalHaplotypePath));
assert(stList_length(maximalHaplotypePath) > 0);
Segment *_5Segment = stList_get(maximalHaplotypePath, 0);
Segment *_3Segment = stList_get(maximalHaplotypePath, stList_length(
maximalHaplotypePath) - 1);
if (getAdjacentCapsSegment(segment_get5Cap(_5Segment)) != NULL) {
assert(!trueAdjacency(segment_get5Cap(_5Segment), eventStrings));
}
if (getAdjacentCapsSegment(segment_get3Cap(_3Segment)) != NULL) {
assert(!trueAdjacency(segment_get3Cap(_3Segment), eventStrings));
}
for (int64_t j = 0; j < stList_length(maximalHaplotypePath) - 1; j++) {
_5Segment = stList_get(maximalHaplotypePath, j);
_3Segment = stList_get(maximalHaplotypePath, j + 1);
assert(trueAdjacency(segment_get3Cap(_5Segment), eventStrings));
assert(trueAdjacency(segment_get5Cap(_3Segment), eventStrings));
assert(cap_getAdjacency(getTerminalCap(segment_get3Cap(_5Segment)))
== getTerminalCap(segment_get5Cap(_3Segment)));
assert(strcmp(event_getHeader(segment_getEvent(_5Segment)),
eventString) == 0);
assert(strcmp(event_getHeader(segment_getEvent(_3Segment)),
eventString) == 0);
assert(hasCapInEvents(cap_getEnd(segment_get5Cap(_5Segment)), eventStrings)); //isHaplotypeEnd(cap_getEnd(segment_get5Cap(_5Segment))));
assert(hasCapInEvents(cap_getEnd(segment_get5Cap(_3Segment)), eventStrings)); //isHaplotypeEnd(cap_getEnd(segment_get5Cap(_3Segment))));
}
}
stSortedSet_destruct(segmentSet);
return maximalHaplotypePaths;
}
static stHash *getScaffoldPathsP(stList *haplotypePaths, stHash *haplotypePathToScaffoldPathHash,
stList *haplotypeEventStrings, stList *contaminationEventStrings, CapCodeParameters *capCodeParameters) {
stHash *haplotypeToMaximalHaplotypeLengthHash = buildContigPathToContigPathLengthHash(haplotypePaths);
stHash *segmentToMaximalHaplotypePathHash = buildSegmentToContigPathHash(haplotypePaths);
for (int64_t i = 0; i < stList_length(haplotypePaths); i++) {
stSortedSet *bucket = stSortedSet_construct();
stHash_insert(haplotypePathToScaffoldPathHash, stList_get(haplotypePaths, i), bucket);
stSortedSet_insert(bucket, stList_get(haplotypePaths, i));
}
for (int64_t i = 0; i < stList_length(haplotypePaths); i++) {
stList *haplotypePath = stList_get(haplotypePaths, i);
assert(stList_length(haplotypePath) > 0);
Segment *_5Segment = stList_get(haplotypePath, 0);
if (!segment_getStrand(_5Segment)) {
_5Segment = segment_getReverse(stList_get(haplotypePath, stList_length(haplotypePath) - 1));
}
assert(segment_getStrand(_5Segment));
if (getAdjacentCapsSegment(segment_get5Cap(_5Segment)) != NULL) {
assert(!trueAdjacency(segment_get5Cap(_5Segment), haplotypeEventStrings));
}
int64_t insertLength;
int64_t deleteLength;
Cap *otherCap;
enum CapCode _5CapCode = getCapCode(segment_get5Cap(_5Segment), &otherCap, haplotypeEventStrings, contaminationEventStrings, &insertLength, &deleteLength, capCodeParameters);
if (_5CapCode == SCAFFOLD_GAP || _5CapCode == AMBIGUITY_GAP) {
assert(stHash_search(haplotypeToMaximalHaplotypeLengthHash, haplotypePath) != NULL);
int64_t j = stIntTuple_get(stHash_search(haplotypeToMaximalHaplotypeLengthHash, haplotypePath), 0);
Segment *adjacentSegment = getAdjacentCapsSegment(segment_get5Cap(_5Segment));
assert(adjacentSegment != NULL);
while (!hasCapInEvents(cap_getEnd(segment_get5Cap(adjacentSegment)), haplotypeEventStrings)) { //is not a haplotype end
adjacentSegment = getAdjacentCapsSegment(segment_get5Cap(adjacentSegment));
assert(adjacentSegment != NULL);
}
assert(adjacentSegment != NULL);
assert(hasCapInEvents(cap_getEnd(segment_get5Cap(adjacentSegment)), haplotypeEventStrings)); //is a haplotype end
stList *adjacentHaplotypePath = stHash_search(segmentToMaximalHaplotypePathHash, adjacentSegment);
if (adjacentHaplotypePath == NULL) {
adjacentHaplotypePath = stHash_search(segmentToMaximalHaplotypePathHash, segment_getReverse(
adjacentSegment));
}
assert(adjacentHaplotypePath != NULL);
assert(adjacentHaplotypePath != haplotypePath);
assert(stHash_search(haplotypeToMaximalHaplotypeLengthHash, adjacentHaplotypePath) != NULL);
int64_t k = stIntTuple_get(stHash_search(haplotypeToMaximalHaplotypeLengthHash, adjacentHaplotypePath), 0);
//Now merge the buckets and make new int tuples..
stSortedSet *bucket1 = stHash_search(haplotypePathToScaffoldPathHash, haplotypePath);
stSortedSet *bucket2 = stHash_search(haplotypePathToScaffoldPathHash, adjacentHaplotypePath);
assert(bucket1 != NULL);
assert(bucket2 != NULL);
assert(bucket1 != bucket2);
stSortedSet *bucket3 = stSortedSet_getUnion(bucket1, bucket2);
stSortedSetIterator *bucketIt = stSortedSet_getIterator(bucket3);
stList *l;
while ((l = stSortedSet_getNext(bucketIt)) != NULL) {
//Do the bucket first
assert(stHash_search(haplotypePathToScaffoldPathHash, l) == bucket1 || stHash_search(haplotypePathToScaffoldPathHash, l) == bucket2);
stHash_remove(haplotypePathToScaffoldPathHash, l);
stHash_insert(haplotypePathToScaffoldPathHash, l, bucket3);
//Now the length
stIntTuple *m = stHash_remove(haplotypeToMaximalHaplotypeLengthHash, l);
assert(m != NULL);
assert(stIntTuple_get(m, 0) == j || stIntTuple_get(m, 0) == k);
stHash_insert(haplotypeToMaximalHaplotypeLengthHash, l, stIntTuple_construct1( j + k));
stIntTuple_destruct(m);
}
assert(stHash_search(haplotypePathToScaffoldPathHash, haplotypePath) == bucket3);
assert(stHash_search(haplotypePathToScaffoldPathHash, adjacentHaplotypePath) == bucket3);
stSortedSet_destructIterator(bucketIt);
}
}
stHash_destruct(segmentToMaximalHaplotypePathHash);
return haplotypeToMaximalHaplotypeLengthHash;
}