Mapper *PredictionTranscript_getcDNACoordMapper(PredictionTranscript *trans) {
Mapper *mapper;
int start = 1;
int i;
Vector *translateable;
if (trans->exonCoordMapper) {
return trans->exonCoordMapper;
}
//
// the mapper is loaded with OBJECTS in place of the IDs !!!!
// the objects are the contigs in the exons
//
// NIY: What should coordsystems be?
mapper = Mapper_new( "cdna", "genomic", NULL, NULL );
translateable = PredictionTranscript_getAllTranslateableExons(trans);
for (i=0; i<Vector_getNumElement(translateable); i++) {
PredictionExon *exon = Vector_getElementAt(translateable,i);
PredictionExon_loadGenomicMapper((Exon*)exon, mapper, (IDType)trans, start);
start += PredictionExon_getLength(exon);
}
trans->exonCoordMapper = mapper;
Vector_free(translateable);
return mapper;
}
AssemblyMapper *AssemblyMapper_new(AssemblyMapperAdaptor *adaptor, Vector *coordSystems) {
AssemblyMapper *am;
if ((am = (AssemblyMapper *)calloc(1, sizeof(AssemblyMapper))) == NULL) {
fprintf(stderr, "ERROR: Failed allocating space for AssemblyMapper\n");
return NULL;
}
am->objectType = CLASS_ASSEMBLYMAPPER;
am->funcs = &assemblyMapperFuncs;
Object_incRefCount(am);
AssemblyMapper_setAdaptor(am, adaptor);
AssemblyMapperAdaptor_cacheSeqIdsWithMultAssemblies(adaptor);
if ( Vector_getNumElement(coordSystems) != 2 ) {
fprintf(stderr, "Can only map between two coordinate systems %d were provided\n", Vector_getNumElement(coordSystems));
exit(1);
}
// Set the component and assembled coordinate systems
AssemblyMapper_setAssembledCoordSystem(am, Vector_getElementAt(coordSystems, 0));
AssemblyMapper_setComponentCoordSystem(am, Vector_getElementAt(coordSystems, 1));
AssemblyMapper_setAssembledRegister(am, IDHash_new(IDHASH_MEDIUM));
AssemblyMapper_setComponentRegister(am, IDHash_new(IDHASH_MEDIUM));
// We load the mapper calling the 'ASSEMBLED' the 'from' coord system
// and the 'COMPONENT' the 'to' coord system.
AssemblyMapper_setMapper(am, Mapper_new("assembled", "component", AssemblyMapper_getAssembledCoordSystem(am),
AssemblyMapper_getComponentCoordSystem(am)));
AssemblyMapper_setMaxPairCount(am, AM_DEFAULT_MAX_PAIR_COUNT);
return am;
}
int main(int argc, char *argv[]) {
initEnsC(argc, argv);
Mapper *mapper = Mapper_new( "rawcontig", "virtualcontig", NULL, NULL );
int nToLoad = loadSGPDump(mapper, 0 );
// loading done successfully
ok(1, nToLoad == 100);
{
// transform a segment entirely within the first rawcontig
int testOutput[][4] = {1, 2, 5, -1};
testTransform (mapper, 627012, 2, 5, -1, "rawcontig", testOutput, NumOutput(testOutput));
}
{
// now a split coord
int testOutput[][4] = {
{314696, 31917, 31937, -1},
{341, 126, 59773, -1},
{315843, 5332, 5963, +1}
};
testTransform (mapper, 1, 383700, 444000, +1, "virtualcontig",testOutput, NumOutput(testOutput));
}
{
// now a simple gap
int testOutput[][4] = {
{ 627011, 7447, 7507, +1 },
{ 1, 273762, 273781, 0 }
};
testTransform (mapper, 1, 273701, 273781, +1, "virtualcontig", testOutput, NumOutput(testOutput));
}
//
// check if the mapper can do merging
//
mapper = Mapper_new( "asm1", "asm2", NULL, NULL );
Mapper_addMapCoordinates(mapper, 1, 1, 10, 1, 1, 101, 110 );
Mapper_addMapCoordinates(mapper, 1, 21, 30, 1, 1, 121, 130 );
Mapper_addMapCoordinates(mapper, 1, 11, 20, 1, 1, 111, 120 );
{
int testOutput[][4] = {{ 1, 105, 125, 1 }};
testTransform(mapper, 1, 5, 25, 1, "asm1", testOutput, NumOutput(testOutput));
}
//
// Slightly differnt merge case
//
mapper = Mapper_new( "asm1", "asm2", NULL, NULL );
Mapper_addMapCoordinates(mapper, 1, 1, 10, 1, 1, 101, 110 );
Mapper_addMapCoordinates(mapper, 1, 21, 30, 1, 1, 121, 130 );
Mapper_addMapCoordinates(mapper, 1, 12, 20, 1, 1, 112, 120 );
{
int testOutput[][4] = {
{ 1, 105, 110, 1 },
{ 1, 11, 11, 0 },
{ 1, 112, 125, 1 }
};
testTransform( mapper, 1, 5, 25, 1, "asm1" , testOutput, NumOutput(testOutput));
}
//
// dont merge on wrong orientation
//
mapper = Mapper_new( "asm1", "asm2", NULL, NULL );
Mapper_addMapCoordinates(mapper, 1, 1, 10, 1, 1, 101, 110 );
Mapper_addMapCoordinates(mapper, 1, 21, 30, 1, 1, 121, 130 );
Mapper_addMapCoordinates(mapper, 1, 11, 20, -1, 1, 111, 120 );
{
int testOutput[][4] = {
{ 1, 105, 110, 1 },
{ 1, 111, 120, -1 },
{ 1, 121, 125, 1 }
};
testTransform( mapper, 1, 5, 25, 1, "asm1" , testOutput, NumOutput(testOutput));
}
//
// can reverse strands merge?
//
mapper = Mapper_new( "asm1", "asm2", NULL, NULL );
Mapper_addMapCoordinates(mapper, 1, 1, 10, -1, 1, 121, 130 );
Mapper_addMapCoordinates(mapper, 1, 21, 30, -1, 1, 101, 110 );
Mapper_addMapCoordinates(mapper, 1, 11, 20, -1, 1, 111, 120 );
{
int testOutput[][4] = {{ 1, 106, 126, -1 } };
testTransform( mapper, 1, 5, 25, 1, "asm1", testOutput, NumOutput(testOutput));
}
//
// normal merge, not three
//
mapper = Mapper_new( "asm1", "asm2", NULL, NULL );
Mapper_addMapCoordinates(mapper, 1, 1, 10, 1, 1, 101, 110 );
Mapper_addMapCoordinates(mapper, 1, 11, 20, 1, 1, 111, 120 );
Mapper_addMapCoordinates(mapper, 1, 22, 30, 1, 1, 132, 140 );
Mapper_addMapCoordinates(mapper, 1, 51, 70, 1, 1, 161, 180 );
Mapper_addMapCoordinates(mapper, 1, 31, 35, 1, 1, 141, 145 );
{
int testOutput[][4] = {
{ 1, 105, 120, 1 },
{ 1, 21, 21, 0 },
{ 1, 132, 145, 1 },
{ 1, 36, 45, 0 }
};
testTransform( mapper, 1, 5, 45, 1, "asm1" , testOutput, NumOutput(testOutput));
}
//
// test tranformation of 'insertion' coordinates where end = start -1
//
mapper = Mapper_new("asm1", "asm2", NULL, NULL);
Mapper_addMapCoordinates(mapper,1, 1, 10, 1, 2, 101, 110);
Mapper_addMapCoordinates(mapper,1, 11, 20, -1, 3, 1, 10);
{
// boundary insert, expect 2 edge inserts back
int testOutput[][4] = {
{2, 111, 110, 1},
{3, 11, 10, -1}
};
testTransform(mapper, 1, 11, 10, 1, "asm1", testOutput, NumOutput(testOutput));
}
{
// edge insert, negative strand, expect edge insert negative strand
int testOutput[][4] = {{2, 101, 100, -1}};
testTransform(mapper, 1, 1, 0, -1, "asm1", testOutput, NumOutput(testOutput));
}
{
// normal case, expect single insert in middle
int testOutput[][4] = {{2, 102, 101, 1}};
testTransform(mapper, 1, 2, 1, 1, "asm1", testOutput, NumOutput(testOutput));
}
{
// expect a gap
int testOutput[][4] = {{1, 100, 200, 0}};
testTransform(mapper, 1, 100, 200, 1, "asm1", testOutput, NumOutput(testOutput));
}
return 0;
}