void testCactusDisk_getUniqueID_Unique(CuTest* testCase) {
cactusDiskTestSetup();
stSortedSet *uniqueNames = stSortedSet_construct3(testCactusDisk_getUniqueID_UniqueP, free);
for (int64_t i = 0; i < 100000; i++) { //Gets a billion ids, checks we are good.
Name uniqueName = cactusDisk_getUniqueID(cactusDisk);
CuAssertTrue(testCase, uniqueName > 0);
CuAssertTrue(testCase, uniqueName < INT64_MAX);
CuAssertTrue(testCase, uniqueName != NULL_NAME);
char *cA = cactusMisc_nameToString(uniqueName);
CuAssertTrue(testCase, stSortedSet_search(uniqueNames, cA) == NULL);
CuAssertTrue(testCase, cactusMisc_stringToName(cA) == uniqueName);
stSortedSet_insert(uniqueNames, cA);
}
stSortedSet_destruct(uniqueNames);
cactusDiskTestTeardown();
}
void testCactusDisk_getUniqueID_UniqueIntervals(CuTest* testCase) {
cactusDiskTestSetup();
stSortedSet *uniqueNames = stSortedSet_construct3(testCactusDisk_getUniqueID_UniqueP, free);
for (int64_t i = 0; i < 10; i++) { //Gets a billion ids, checks we are good.
int64_t intervalSize = st_randomInt(0, 100000);
Name uniqueName = cactusDisk_getUniqueIDInterval(cactusDisk, intervalSize);
for(int64_t j=0; j<intervalSize; j++) {
CuAssertTrue(testCase, uniqueName > 0);
CuAssertTrue(testCase, uniqueName < INT64_MAX);
CuAssertTrue(testCase, uniqueName != NULL_NAME);
char *cA = cactusMisc_nameToString(uniqueName);
CuAssertTrue(testCase, stSortedSet_search(uniqueNames, cA) == NULL);
CuAssertTrue(testCase, cactusMisc_stringToName(cA) == uniqueName);
stSortedSet_insert(uniqueNames, cA);
uniqueName++;
}
}
stSortedSet_destruct(uniqueNames);
cactusDiskTestTeardown();
}
int main(int argc, char *argv[]) {
st_setLogLevelFromString(argv[1]);
st_logDebug("Set up logging\n");
stKVDatabaseConf *kvDatabaseConf = stKVDatabaseConf_constructFromString(argv[2]);
CactusDisk *cactusDisk = cactusDisk_construct(kvDatabaseConf, 0);
stKVDatabaseConf_destruct(kvDatabaseConf);
st_logDebug("Set up the flower disk\n");
Name flowerName = cactusMisc_stringToName(argv[3]);
Flower *flower = cactusDisk_getFlower(cactusDisk, flowerName);
int64_t totalBases = flower_getTotalBaseLength(flower);
int64_t totalEnds = flower_getEndNumber(flower);
int64_t totalFreeEnds = flower_getFreeStubEndNumber(flower);
int64_t totalAttachedEnds = flower_getAttachedStubEndNumber(flower);
int64_t totalCaps = flower_getCapNumber(flower);
int64_t totalBlocks = flower_getBlockNumber(flower);
int64_t totalGroups = flower_getGroupNumber(flower);
int64_t totalChains = flower_getChainNumber(flower);
int64_t totalLinkGroups = 0;
int64_t maxEndDegree = 0;
int64_t maxAdjacencyLength = 0;
int64_t totalEdges = 0;
Flower_EndIterator *endIt = flower_getEndIterator(flower);
End *end;
while((end = flower_getNextEnd(endIt)) != NULL) {
assert(end_getOrientation(end));
if(end_getInstanceNumber(end) > maxEndDegree) {
maxEndDegree = end_getInstanceNumber(end);
}
stSortedSet *ends = stSortedSet_construct();
End_InstanceIterator *capIt = end_getInstanceIterator(end);
Cap *cap;
while((cap = end_getNext(capIt)) != NULL) {
if(cap_getSequence(cap) != NULL) {
Cap *adjacentCap = cap_getAdjacency(cap);
assert(adjacentCap != NULL);
End *adjacentEnd = end_getPositiveOrientation(cap_getEnd(adjacentCap));
stSortedSet_insert(ends, adjacentEnd);
int64_t adjacencyLength = cap_getCoordinate(cap) - cap_getCoordinate(adjacentCap);
if(adjacencyLength < 0) {
adjacencyLength *= -1;
}
assert(adjacencyLength >= 1);
if(adjacencyLength >= maxAdjacencyLength) {
maxAdjacencyLength = adjacencyLength;
}
}
}
end_destructInstanceIterator(capIt);
totalEdges += stSortedSet_size(ends);
if(stSortedSet_search(ends, end) != NULL) { //This ensures we count self edges twice, so that the division works.
totalEdges += 1;
}
stSortedSet_destruct(ends);
}
assert(totalEdges % 2 == 0);
flower_destructEndIterator(endIt);
Flower_GroupIterator *groupIt = flower_getGroupIterator(flower);
Group *group;
while((group = flower_getNextGroup(groupIt)) != NULL) {
if(group_getLink(group) != NULL) {
totalLinkGroups++;
}
}
flower_destructGroupIterator(groupIt);
printf("flower name: %" PRIi64 " total bases: %" PRIi64 " total-ends: %" PRIi64 " total-caps: %" PRIi64 " max-end-degree: %" PRIi64 " max-adjacency-length: %" PRIi64 " total-blocks: %" PRIi64 " total-groups: %" PRIi64 " total-edges: %" PRIi64 " total-free-ends: %" PRIi64 " total-attached-ends: %" PRIi64 " total-chains: %" PRIi64 " total-link groups: %" PRIi64 "\n",
flower_getName(flower), totalBases, totalEnds, totalCaps, maxEndDegree, maxAdjacencyLength, totalBlocks, totalGroups, totalEdges/2, totalFreeEnds, totalAttachedEnds, totalChains, totalLinkGroups);
return 0;
}
void cactusDisk_write(CactusDisk *cactusDisk) {
Flower *flower;
int64_t recordSize;
stList *removeRequests = stList_construct3(0, (void (*)(void *)) stIntTuple_destruct);
st_logDebug("Starting to write the cactus to disk\n");
stSortedSetIterator *it = stSortedSet_getIterator(cactusDisk->flowers);
//Sort flowers to update.
while ((flower = stSortedSet_getNext(it)) != NULL) {
cactusDisk_addUpdateRequest(cactusDisk, flower);
}
stSortedSet_destructIterator(it);
st_logDebug("Got the flowers to update\n");
//Remove nets that are marked for deletion..
it = stSortedSet_getIterator(cactusDisk->flowerNamesMarkedForDeletion);
char *nameString;
while ((nameString = stSortedSet_getNext(it)) != NULL) {
Name name = cactusMisc_stringToName(nameString);
if (containsRecord(cactusDisk, name)) {
stList_append(cactusDisk->updateRequests, stKVDatabaseBulkRequest_constructUpdateRequest(name, &name, 0)); //We set it to null in the first atomic operation.
stList_append(removeRequests, stIntTuple_construct1(name));
}
}
stSortedSet_destructIterator(it);
st_logDebug("Avoided updating nets marked for deletion\n");
// Insert and/or update meta-sequences.
it = stSortedSet_getIterator(cactusDisk->metaSequences);
MetaSequence *metaSequence;
while ((metaSequence = stSortedSet_getNext(it)) != NULL) {
void *vA =
binaryRepresentation_makeBinaryRepresentation(metaSequence,
(void (*)(void *, void (*)(const void * ptr, size_t size, size_t count))) metaSequence_writeBinaryRepresentation,
&recordSize);
//Compression
vA = compress(vA, &recordSize);
if (!containsRecord(cactusDisk, metaSequence_getName(metaSequence))) {
stList_append(cactusDisk->updateRequests,
stKVDatabaseBulkRequest_constructInsertRequest(metaSequence_getName(metaSequence), vA, recordSize));
} else {
stList_append(cactusDisk->updateRequests,
stKVDatabaseBulkRequest_constructUpdateRequest(metaSequence_getName(metaSequence), vA, recordSize));
}
free(vA);
}
stSortedSet_destructIterator(it);
st_logDebug("Got the sequences we are going to add to the database.\n");
if (!containsRecord(cactusDisk, CACTUS_DISK_PARAMETER_KEY)) { //We only write the parameters once.
//Finally the database info.
void *cactusDiskParameters =
binaryRepresentation_makeBinaryRepresentation(cactusDisk,
(void (*)(void *, void (*)(const void * ptr, size_t size, size_t count))) cactusDisk_writeBinaryRepresentation,
&recordSize);
//Compression
cactusDiskParameters = compress(cactusDiskParameters, &recordSize);
stList_append(cactusDisk->updateRequests,
stKVDatabaseBulkRequest_constructInsertRequest(CACTUS_DISK_PARAMETER_KEY, cactusDiskParameters,
recordSize));
free(cactusDiskParameters);
}
st_logDebug("Checked if need to write the initial parameters\n");
if (stList_length(cactusDisk->updateRequests) > 0) {
st_logDebug("Going to write %" PRIi64 " updates\n", stList_length(cactusDisk->updateRequests));
stTry
{
st_logDebug("Writing %" PRIi64 " updates\n", stList_length(cactusDisk->updateRequests));
assert(stList_length(cactusDisk->updateRequests) > 0);
stKVDatabase_bulkSetRecords(cactusDisk->database, cactusDisk->updateRequests);
}
stCatch(except)
{
stThrowNewCause(except, ST_KV_DATABASE_EXCEPTION_ID,
"Failed when trying to set records in updating the cactus disk");
}stTryEnd
;
}
int main(int argc, char *argv[]) {
/*
* Script for adding a reference genome to a flower.
*/
/*
* Arguments/options
*/
char * logLevelString = NULL;
char * cactusDiskDatabaseString = NULL;
char *referenceEventString =
(char *) cactusMisc_getDefaultReferenceEventHeader();
char *outputFile = NULL;
Name flowerName = NULL_NAME;
///////////////////////////////////////////////////////////////////////////
// (0) Parse the inputs handed by genomeCactus.py / setup stuff.
///////////////////////////////////////////////////////////////////////////
while (1) {
static struct option long_options[] = { { "logLevel",
required_argument, 0, 'a' }, { "cactusDisk", required_argument,
0, 'c' }, { "flowerName", required_argument,
0, 'd' },
{ "referenceEventString", required_argument, 0, 'g' }, {
"help", no_argument, 0, 'h' }, { "outputFile",
required_argument, 0, 'k' },
{ 0, 0, 0, 0 } };
int option_index = 0;
int key = getopt_long(argc, argv, "a:c:d:g:hk:", long_options,
&option_index);
if (key == -1) {
break;
}
switch (key) {
case 'a':
logLevelString = stString_copy(optarg);
break;
case 'c':
cactusDiskDatabaseString = stString_copy(optarg);
break;
case 'd':
flowerName = cactusMisc_stringToName(optarg);
break;
case 'g':
referenceEventString = stString_copy(optarg);
break;
case 'h':
usage();
return 0;
case 'k':
outputFile = stString_copy(optarg);
break;
default:
usage();
return 1;
}
}
///////////////////////////////////////////////////////////////////////////
// (0) Check the inputs.
///////////////////////////////////////////////////////////////////////////
assert(cactusDiskDatabaseString != NULL);
//////////////////////////////////////////////
//Set up logging
//////////////////////////////////////////////
st_setLogLevelFromString(logLevelString);
//////////////////////////////////////////////
//Load the database
//////////////////////////////////////////////
stKVDatabaseConf *kvDatabaseConf = stKVDatabaseConf_constructFromString(
cactusDiskDatabaseString);
CactusDisk *cactusDisk = cactusDisk_construct(kvDatabaseConf, 0);
stKVDatabaseConf_destruct(kvDatabaseConf);
st_logInfo("Set up the flower disk\n");
///////////////////////////////////////////////////////////////////////////
// Get the set of flowers to manipulate
///////////////////////////////////////////////////////////////////////////
Flower *flower = cactusDisk_getFlower(cactusDisk, flowerName);
///////////////////////////////////////////////////////////////////////////
// Get the reference event name
///////////////////////////////////////////////////////////////////////////
Event *referenceEvent = eventTree_getEventByHeader(
flower_getEventTree(flower), referenceEventString);
assert(referenceEvent != NULL);
Name referenceEventName = event_getName(referenceEvent);
///////////////////////////////////////////////////////////////////////////
// Now process each flower in turn.
///////////////////////////////////////////////////////////////////////////
if(outputFile == NULL) {
st_errAbort("No output file specified\n");
}
FILE *fileHandle = fopen(outputFile, "w");
printFastaSequences(flower, fileHandle, referenceEventName);
if(fileHandle != NULL) {
fclose(fileHandle);
}
///////////////////////////////////////////////////////////////////////////
//Clean up memory
///////////////////////////////////////////////////////////////////////////
cactusDisk_destruct(cactusDisk);
//return 0; //Exit without clean up is quicker, enable cleanup when doing memory leak detection.
free(cactusDiskDatabaseString);
free(referenceEventString);
free(logLevelString);
st_logInfo("Cleaned stuff up and am finished\n");
//while(1);
return 0;
}
int testCactusDisk_getUniqueID_UniqueP(const void *a, const void *b) {
return cactusMisc_nameCompare(cactusMisc_stringToName(a), cactusMisc_stringToName(b));
}
int main(int argc, char *argv[]) {
char * logLevelString = NULL;
char * cactusDiskDatabaseString = NULL;
char * flowerName = NULL;
char * outputFile = NULL;
char *referenceEventString = NULL;
///////////////////////////////////////////////////////////////////////////
// (0) Parse the inputs handed by genomeCactus.py / setup stuff.
///////////////////////////////////////////////////////////////////////////
while (1) {
static struct option long_options[] = {
{ "logLevel", required_argument, 0, 'a' },
{ "referenceEventString", required_argument, 0, 'b' },
{ "cactusDisk", required_argument, 0, 'c' },
{ "flowerName", required_argument, 0, 'e' },
{ "outputFile", required_argument, 0, 'f' },
{ "help", no_argument, 0, 'h' }, { 0, 0, 0, 0 } };
int option_index = 0;
int key = getopt_long(argc, argv, "a:b:c:d:e:f:h", long_options,
&option_index);
if (key == -1) {
break;
}
switch (key) {
case 'a':
logLevelString = stString_copy(optarg);
break;
case 'b':
referenceEventString = stString_copy(optarg);
break;
case 'c':
cactusDiskDatabaseString = stString_copy(optarg);
break;
case 'e':
flowerName = stString_copy(optarg);
break;
case 'f':
outputFile = stString_copy(optarg);
break;
case 'h':
usage();
return 0;
default:
usage();
return 1;
}
}
///////////////////////////////////////////////////////////////////////////
// (0) Check the inputs.
///////////////////////////////////////////////////////////////////////////
assert(flowerName != NULL);
assert(referenceEventString != NULL);
assert(cactusDiskDatabaseString != NULL);
assert(outputFile != NULL);
//////////////////////////////////////////////
//Set up logging
//////////////////////////////////////////////
st_setLogLevelFromString(logLevelString);
//////////////////////////////////////////////
//Log (some of) the inputs
//////////////////////////////////////////////
st_logInfo("Flower name : %s\n", flowerName);
st_logInfo("Sequence name : %s\n", referenceEventString);
st_logInfo("Output file : %s\n", outputFile);
//////////////////////////////////////////////
//Load the database
//////////////////////////////////////////////
stKVDatabaseConf *kvDatabaseConf = stKVDatabaseConf_constructFromString(
cactusDiskDatabaseString);
CactusDisk *cactusDisk = cactusDisk_construct(kvDatabaseConf, false, true);
st_logInfo("Set up the flower disk\n");
///////////////////////////////////////////////////////////////////////////
// Parse the basic reconstruction problem
///////////////////////////////////////////////////////////////////////////
Flower *flower = cactusDisk_getFlower(cactusDisk, cactusMisc_stringToName(
flowerName));
st_logInfo("Parsed the top level flower of the cactus tree to check\n");
///////////////////////////////////////////////////////////////////////////
// Recursive check the flowers.
///////////////////////////////////////////////////////////////////////////
//int64_t startTime = time(NULL);
//flower = flower_addReferenceSequence(flower, cactusDisk, name);
//st_logInfo("Added the reference sequence in %" PRIi64 " seconds/\n", time(NULL) - startTime);
int64_t numSequences = flower_getSequenceNumber(flower);
//Make sure that referenceSequence has already been added:
if(getSequenceMatchesEvent(flower, referenceEventString) == NULL &&
numSequences > 0){
fprintf(stderr, "No reference sequence found in cactusDisk\n");
exit(EXIT_FAILURE);
}
FILE *fileHandle = fopen(outputFile, "w");
if (numSequences > 0) {
getReferenceSequences(fileHandle, flower, referenceEventString);
}
else {
st_logCritical("cactus_getReferenceSeq found no reference sequence in empty cactus disk %s",
cactusDiskDatabaseString);
}
fclose(fileHandle);
///////////////////////////////////////////////////////////////////////////
// Clean up.
///////////////////////////////////////////////////////////////////////////
cactusDisk_destruct(cactusDisk);
return 0; //Exit without clean up is quicker, enable cleanup when doing memory leak detection.
stKVDatabaseConf_destruct(kvDatabaseConf);
return 0;
}
int main(int argc, char *argv[]) {
Flower *flower;
FILE *fileHandle;
/*
* Arguments/options
*/
char * logLevelString = NULL;
char * cactusDiskDatabaseString = NULL;
char * flowerName = NULL;
char * outputFile = NULL;
///////////////////////////////////////////////////////////////////////////
// (0) Parse the inputs handed by genomeCactus.py / setup stuff.
///////////////////////////////////////////////////////////////////////////
while (1) {
static struct option long_options[] = { { "logLevel",
required_argument, 0, 'a' }, { "cactusDisk", required_argument, 0,
'c' }, { "flowerName", required_argument, 0, 'd' }, {
"outputFile", required_argument, 0, 'e' }, { "scaleNodeSizes",
no_argument, 0, 'f' }, { "nameLabels", no_argument, 0, 'g' }, {
"help", no_argument, 0, 'h' }, { 0, 0, 0, 0 } };
int option_index = 0;
int key = getopt_long(argc, argv, "a:c:d:e:fgh", long_options,
&option_index);
if (key == -1) {
break;
}
switch (key) {
case 'a':
logLevelString = stString_copy(optarg);
break;
case 'c':
cactusDiskDatabaseString = stString_copy(optarg);
break;
case 'd':
flowerName = stString_copy(optarg);
break;
case 'e':
outputFile = stString_copy(optarg);
break;
case 'f':
scaleNodeSizes = !scaleNodeSizes;
break;
case 'g':
nameLabels = !nameLabels;
break;
case 'h':
usage();
return 0;
default:
usage();
return 1;
}
}
///////////////////////////////////////////////////////////////////////////
// (0) Check the inputs.
///////////////////////////////////////////////////////////////////////////
assert(cactusDiskDatabaseString != NULL);
assert(flowerName != NULL);
assert(outputFile != NULL);
//////////////////////////////////////////////
//Set up logging
//////////////////////////////////////////////
st_setLogLevelFromString(logLevelString);
//////////////////////////////////////////////
//Log (some of) the inputs
//////////////////////////////////////////////
st_logInfo("Flower name : %s\n", flowerName);
st_logInfo("Output graph file : %s\n", outputFile);
//////////////////////////////////////////////
//Load the database
//////////////////////////////////////////////
stKVDatabaseConf *kvDatabaseConf = stKVDatabaseConf_constructFromString(cactusDiskDatabaseString);
CactusDisk *cactusDisk = cactusDisk_construct(kvDatabaseConf, 0);
st_logInfo("Set up the flower disk\n");
///////////////////////////////////////////////////////////////////////////
// Parse the basic reconstruction problem
///////////////////////////////////////////////////////////////////////////
flower = cactusDisk_getFlower(cactusDisk, cactusMisc_stringToName(flowerName));
st_logInfo("Parsed the top level flower of the cactus tree to build\n");
///////////////////////////////////////////////////////////////////////////
// Build the graph.
///////////////////////////////////////////////////////////////////////////
totalProblemSize = flower_getTotalBaseLength(flower);
fileHandle = fopen(outputFile, "w");
graphViz_setupGraphFile(fileHandle);
makeCactusTree_flower(flower, fileHandle, NULL, NULL);
graphViz_finishGraphFile(fileHandle);
fclose(fileHandle);
st_logInfo("Written the tree to file\n");
///////////////////////////////////////////////////////////////////////////
// Clean up.
///////////////////////////////////////////////////////////////////////////
cactusDisk_destruct(cactusDisk);
stKVDatabaseConf_destruct(kvDatabaseConf);
return 0;
}
//============================== MAIN =========================================
int main(int argc, char *argv[]) {
Flower *flower;
/*
* Arguments/options
*/
char * st_logLevelString = NULL;
char * cactusDiskDatabaseString = NULL;
char * flowerName = "0";
char * outputFile = NULL;
char * species = NULL;
char * geneFile = NULL;
///////////////////////////////////////////////////////////////////////////
// (0) Parse the inputs handed by genomeCactus.py / setup stuff.
///////////////////////////////////////////////////////////////////////////
while(1) {
static struct option long_options[] = {
{ "genePslFile", required_argument, 0, 'g' },
{ "species", required_argument, 0, 's' },
{ "st_logLevel", required_argument, 0, 'a' },
{ "cactusDisk", required_argument, 0, 'c' },
{ "outputFile", required_argument, 0, 'o' },
{ "help", no_argument, 0, 'h' },
{ 0, 0, 0, 0 }
};
int option_index = 0;
int key = getopt_long(argc, argv, "s:g:o:a:c:h", long_options, &option_index);
if(key == -1) {
break;
}
switch(key) {
case 'a':
st_logLevelString = stString_copy(optarg);
break;
case 'c':
cactusDiskDatabaseString = stString_copy(optarg);
break;
case 'o':
outputFile = stString_copy(optarg);
break;
case 's':
species = stString_copy(optarg);
break;
case 'g':
geneFile = stString_copy(optarg);
break;
case 'h':
usage();
return 0;
default:
usage();
return 1;
}
}
///////////////////////////////////////////////////////////////////////////
// (0) Check the inputs.
///////////////////////////////////////////////////////////////////////////
assert(cactusDiskDatabaseString != NULL);
assert(outputFile != NULL);
assert(species != NULL);
assert(geneFile != NULL);
//////////////////////////////////////////////
//Set up st_logging
//////////////////////////////////////////////
st_setLogLevelFromString(st_logLevelString);
//////////////////////////////////////////////
//Log (some of) the inputs
//////////////////////////////////////////////
st_logInfo("Flower disk name : %s\n", cactusDiskDatabaseString);
st_logInfo("Output file : %s\n", outputFile);
st_logInfo("Species: %s\n", species);
st_logInfo("GenePslFile: %s\n", geneFile);
//////////////////////////////////////////////
//Load the database
//////////////////////////////////////////////
stKVDatabaseConf *kvDatabaseConf = stKVDatabaseConf_constructFromString(cactusDiskDatabaseString);
CactusDisk *cactusDisk = cactusDisk_construct(kvDatabaseConf, 0);
st_logInfo("Set up the flower disk\n");
///////////////////////////////////////////////////////////////////////////
// Parse the basic reconstruction problem
///////////////////////////////////////////////////////////////////////////
flower = cactusDisk_getFlower(cactusDisk, cactusMisc_stringToName(flowerName));
st_logInfo("Parsed the top level flower of the cactus tree to check\n");
///////////////////////////////////////////////////////////////////////////
// Recursive check the flowers.
///////////////////////////////////////////////////////////////////////////
int64_t startTime = time(NULL);
FILE *fileHandle = fopen(outputFile, "w");
struct bed *gene = bedLoadAll(geneFile);
mapGenes(flower, fileHandle, gene, species);
fclose(fileHandle);
st_logInfo("Map genes in %" PRIi64 " seconds/\n", time(NULL) - startTime);
///////////////////////////////////////////////////////////////////////////
// Clean up.
///////////////////////////////////////////////////////////////////////////
cactusDisk_destruct(cactusDisk);
return 0;
}
