void block_check(Block *block) {
//Check is connected to flower properly
cactusCheck(flower_getBlock(block_getFlower(block), block_getName(block)) == block_getPositiveOrientation(block));
//Check we have actually set built blocks for the flower..
cactusCheck(flower_builtBlocks(block_getFlower(block)));
//Checks the two ends are block ends.
End *_5End = block_get5End(block);
End *_3End = block_get3End(block);
cactusCheck(end_isBlockEnd(_5End));
cactusCheck(end_isBlockEnd(_3End));
cactusCheck(end_getOrientation(_5End) == block_getOrientation(block));
cactusCheck(end_getOrientation(_3End) == block_getOrientation(block));
cactusCheck(end_getBlock(_5End) == block);
cactusCheck(end_getBlock(_3End) == block);
cactusCheck(end_getSide(_5End)); //Check the sides of the ends are consistent.
cactusCheck(!end_getSide(_3End));
cactusCheck(block_getLength(block) > 0); //check block has non-zero length
//Check reverse
Block *rBlock = block_getReverse(block);
cactusCheck(rBlock != NULL);
cactusCheck(block_getReverse(block) == rBlock);
cactusCheck(block_getOrientation(block) == !block_getOrientation(rBlock));
cactusCheck(block_getLength(block) == block_getLength(rBlock));
cactusCheck(block_get5End(block) == end_getReverse(block_get3End(rBlock)));
cactusCheck(block_get3End(block) == end_getReverse(block_get5End(rBlock)));
cactusCheck(block_getInstanceNumber(block) == block_getInstanceNumber(rBlock));
if(block_getInstanceNumber(block) > 0) {
cactusCheck(block_getFirst(block) == segment_getReverse(block_getFirst(rBlock)));
if(block_getRootInstance(block) == NULL) {
cactusCheck(block_getRootInstance(rBlock) == NULL);
}
else {
cactusCheck(block_getRootInstance(block) == segment_getReverse(block_getRootInstance(rBlock)));
}
}
//For each segment calls segment_check.
Block_InstanceIterator *iterator = block_getInstanceIterator(block);
Segment *segment;
while((segment = block_getNext(iterator)) != NULL) {
segment_check(segment);
}
block_destructInstanceIterator(iterator);
}
void testBlock_getRightEnd(CuTest* testCase) {
cactusBlockTestSetup();
End *rightEnd = block_get3End(block);
CuAssertTrue(testCase, rightEnd != NULL);
CuAssertTrue(testCase, end_getBlock(rightEnd) == block);
CuAssertTrue(testCase, block_getOrientation(block) == end_getOrientation(rightEnd));
cactusBlockTestTeardown();
}
void testBlock_getLeftEnd(CuTest* testCase) {
cactusBlockTestSetup();
End *leftEnd = block_get5End(block);
CuAssertTrue(testCase, leftEnd != NULL);
CuAssertTrue(testCase, end_getBlock(leftEnd) == block);
CuAssertTrue(testCase, block_getOrientation(block) == end_getOrientation(leftEnd));
cactusBlockTestTeardown();
}
void mapBlockToExon(Cap *cap, int level, FILE *fileHandle){
fprintf(fileHandle, "\t\t\t<block>\n");
Block *block = end_getBlock(cap_getEnd(cap));
Chain *chain = block_getChain(block);
int start = cap_getCoordinate(cap);
int end = cap_getCoordinate(cap_getOtherSegmentCap(cap)) +1;
fprintf(fileHandle, "\t\t\t\t<blockName>%s</blockName>\n", cactusMisc_nameToString(block_getName(block)));
if(chain != NULL){
fprintf(fileHandle, "\t\t\t\t<chainName>%s</chainName>\n", cactusMisc_nameToString(chain_getName(chain)));
}else{
fprintf(fileHandle, "\t\t\t\t<chainName>NA</chainName>\n");
}
fprintf(fileHandle, "\t\t\t\t<level>%d</level>\n", level);
fprintf(fileHandle, "\t\t\t\t<start>%d</start>\n", start);
fprintf(fileHandle, "\t\t\t\t<end>%d</end>\n", end);
fprintf(fileHandle, "\t\t\t</block>\n");
st_logInfo("mapBlockToExon: start: %d, end: %d\n", start, end);
}
void testEnd_getBlock(CuTest* testCase) {
cactusEndTestSetup();
Block *block = block_construct(10, flower);
End *leftEnd = block_get5End(block);
End *rightEnd = block_get3End(block);
CuAssertTrue(testCase, end_getBlock(end) == NULL);
CuAssertTrue(testCase, end_getBlock(end_getReverse(end)) == NULL);
CuAssertTrue(testCase, end_getBlock(leftEnd) == block);
CuAssertTrue(testCase, end_getBlock(end_getReverse(leftEnd)) == block_getReverse(block));
CuAssertTrue(testCase, block_getOrientation(block) == end_getOrientation(leftEnd));
CuAssertTrue(testCase, end_getBlock(rightEnd) == block);
CuAssertTrue(testCase, end_getBlock(end_getReverse(rightEnd)) == block_getReverse(block));
CuAssertTrue(testCase, block_getOrientation(block) == end_getOrientation(rightEnd));
cactusEndTestTeardown();
}
int mapGene(Cap *cap, int level, int exon, struct bed *gene, FILE *fileHandle){
/*
*Following cactus adjacencies, starting from 'cap', find regions that overlap with
*exons of input gene. Report chain relations of these regions with the exons.
*cap: current cap. Level = chain level. exon = exon number. gene = bed record of gene
*/
int64_t exonStart, exonEnd;
if(isStubCap(cap)){
Group *group = end_getGroup(cap_getEnd(cap));
Flower *nestedFlower = group_getNestedFlower(group);
if(nestedFlower != NULL){//recursive call
Cap *childCap = flower_getCap(nestedFlower, cap_getName(cap));
assert(childCap != NULL);
exon = mapGene(childCap, level + 1, exon, gene, fileHandle);
exonStart = gene->chromStarts->list[exon] + gene->chromStart;
exonEnd = exonStart + gene->blockSizes->list[exon];
}
}
cap = cap_getAdjacency(cap);
Cap *nextcap;
int64_t capCoor;
exonStart = gene->chromStarts->list[exon] + gene->chromStart;
exonEnd = exonStart + gene->blockSizes->list[exon];
Block *block = end_getBlock(cap_getEnd(cap));
if(block == NULL){
moveCapToNextBlock(&cap);
}
while(!isStubCap(cap) && exon < gene->blockCount){
End *cend = cap_getEnd(cap);
capCoor = cap_getCoordinate(cap);//Cap coordinate is always the coordinate on + strand
nextcap = cap_getAdjacency(cap_getOtherSegmentCap(cap));
st_logInfo("capCoor: %d, nextCap: %d, eStart: %d, eEnd: %d. Exon: %d\n",
capCoor, cap_getCoordinate(nextcap), exonStart, exonEnd, exon);
//keep moving if nextBlock Start is still upstream of current exon
if(cap_getCoordinate(nextcap) <= exonStart){
moveCapToNextBlock(&cap);
st_logInfo("Still upstream, nextcap <= exonStart. Move to next chainBlock\n");
}else if(capCoor >= exonEnd){//Done with current exon, move to next
st_logInfo("Done with current exon, move to next one\n\n");
fprintf(fileHandle, "\t\t</exon>\n");//end previous exon
exon++;
if(exon < gene->blockCount){
exonStart = gene->chromStarts->list[exon] + gene->chromStart;
exonEnd = exonStart + gene->blockSizes->list[exon];
fprintf(fileHandle, "\t\t<exon id=\"%d\" start=\"%" PRIi64 "\" end=\"%" PRIi64 "\">\n", exon, exonStart, exonEnd);
}
}else{//current exon overlaps with current block Or with lower level flower
Cap *oppcap = cap_getOtherSegmentCap(cap);
st_logInfo("Current exon overlaps with current block or with lower flower\n");
if(cap_getCoordinate(oppcap) >= exonStart && exonEnd > capCoor){
mapBlockToExon(cap, level, fileHandle);
if(exonEnd <= cap_getCoordinate(oppcap) + 1){
st_logInfo("Done with current exon, move to next one\n\n");
fprintf(fileHandle, "\t\t</exon>\n");//end previous exon
exon++;
if(exon < gene->blockCount){
exonStart = gene->chromStarts->list[exon] + gene->chromStart;
exonEnd = exonStart + gene->blockSizes->list[exon];
fprintf(fileHandle, "\t\t<exon id=\"%d\" start=\"%" PRIi64 "\" end=\"%" PRIi64 "\">\n", exon, exonStart, exonEnd);
}
continue;
}
}
//Traverse lower level flowers if exists
Group *group = end_getGroup(end_getOtherBlockEnd(cend));
Flower *nestedFlower = group_getNestedFlower(group);
if(nestedFlower != NULL){//recursive call
Cap *childCap = flower_getCap(nestedFlower, cap_getName(cap_getOtherSegmentCap(cap)));
assert(childCap != NULL);
exon = mapGene(childCap, level + 1, exon, gene, fileHandle);
exonStart = gene->chromStarts->list[exon] + gene->chromStart;
exonEnd = exonStart + gene->blockSizes->list[exon];
}
moveCapToNextBlock(&cap);
}
}
return exon;
}
