Group *flower_getParentGroup(Flower *flower) {
if (flower->parentFlowerName == NULL_NAME) {
return NULL;
}
Flower *flower2 = cactusDisk_getFlower(flower_getCactusDisk(flower), flower->parentFlowerName);
assert(flower2 != NULL);
return flower_getGroup(flower2, flower_getName(flower));
}
void flower_unloadParent(Flower *flower) {
Name parentName = flower->parentFlowerName;
if (parentName != NULL_NAME) {
CactusDisk *cactusDisk = flower_getCactusDisk(flower);
if (cactusDisk_flowerIsLoaded(cactusDisk, parentName)) {
Flower *parentFlower = cactusDisk_getFlower(cactusDisk, parentName);
flower_unload(parentFlower);
}
}
}
void testCactusDisk_getFlower(CuTest* testCase) {
cactusDiskTestSetup();
Flower *flower = flower_construct(cactusDisk);
Flower *flower2 = flower_construct(cactusDisk);
CuAssertTrue(testCase, cactusDisk_getFlower(cactusDisk, flower_getName(flower)) == flower);
CuAssertTrue(testCase, cactusDisk_getFlower(cactusDisk, flower_getName(flower2)) == flower2);
//now try closing the disk, then reloading it, to see if we get the same result.
Name name1 = flower_getName(flower);
Name name2 = flower_getName(flower2);
cactusDisk_write(cactusDisk);
cactusDisk_destruct(cactusDisk);
cactusDisk = cactusDisk_construct(conf, 0);
flower = cactusDisk_getFlower(cactusDisk, name1);
flower2 = cactusDisk_getFlower(cactusDisk, name2);
CuAssertTrue(testCase, flower != NULL);
CuAssertTrue(testCase, flower2 != NULL);
CuAssertTrue(testCase, flower_getName(flower) == name1);
CuAssertTrue(testCase, flower_getName(flower2) == name2);
cactusDiskTestTeardown();
}
int main(int argc, char *argv[])
{
char *cactusDiskString = NULL;
stKVDatabaseConf *kvDatabaseConf;
CactusDisk *cactusDisk;
Flower *flower;
Flower_SequenceIterator *flowerIt;
Sequence *sequence;
struct option longopts[] = { {"cactusDisk", required_argument, NULL, 'c' },
{0, 0, 0, 0} };
int flag;
while((flag = getopt_long(argc, argv, "", longopts, NULL)) != -1) {
switch(flag) {
case 'c':
cactusDiskString = stString_copy(optarg);
break;
case '?':
default:
usage();
return 1;
}
}
if (cactusDiskString == NULL) {
st_errAbort("--cactusDisk option must be provided");
}
kvDatabaseConf = stKVDatabaseConf_constructFromString(cactusDiskString);
cactusDisk = cactusDisk_construct(kvDatabaseConf, 0);
// Get top-level flower.
flower = cactusDisk_getFlower(cactusDisk, 0);
flowerIt = flower_getSequenceIterator(flower);
while((sequence = flower_getNextSequence(flowerIt)) != NULL) {
MetaSequence *metaSequence = sequence_getMetaSequence(sequence);
const char *header;
char *firstToken, *newHeader;
stList *tokens;
// Strip the ID token from the header (should be the first
// |-separated token) and complain if there isn't one.
header = metaSequence_getHeader(metaSequence);
tokens = fastaDecodeHeader(header);
assert(stList_length(tokens) > 1);
firstToken = stList_removeFirst(tokens);
assert(!strncmp(firstToken, "id=", 3));
free(firstToken);
newHeader = fastaEncodeHeader(tokens);
metaSequence_setHeader(metaSequence, newHeader);
}
cactusDisk_write(cactusDisk);
}
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;
}
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 main(int argc, char *argv[]) {
/*
* Open the database.
* Construct a flower.
* Construct an event tree representing the species tree.
* For each sequence contruct two ends each containing an cap.
* Make a file for the sequence.
* Link the two caps.
* Finish!
*/
int64_t key, j;
Group *group;
Flower_EndIterator *endIterator;
End *end;
bool makeEventHeadersAlphaNumeric = 0;
/*
* Arguments/options
*/
char * logLevelString = NULL;
char * speciesTree = NULL;
char * outgroupEvents = 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, 'b' }, {
"speciesTree", required_argument, 0, 'f' }, { "outgroupEvents", required_argument, 0, 'g' },
{ "help", no_argument, 0, 'h' }, { "makeEventHeadersAlphaNumeric", no_argument, 0, 'i' }, { 0, 0, 0, 0 } };
int option_index = 0;
key = getopt_long(argc, argv, "a:b:f:hg:i", long_options, &option_index);
if (key == -1) {
break;
}
switch (key) {
case 'a':
logLevelString = optarg;
break;
case 'b':
cactusDiskDatabaseString = optarg;
break;
case 'f':
speciesTree = optarg;
break;
case 'g':
outgroupEvents = optarg;
break;
case 'h':
usage();
return 0;
case 'i':
makeEventHeadersAlphaNumeric = 1;
break;
default:
usage();
return 1;
}
}
///////////////////////////////////////////////////////////////////////////
// (0) Check the inputs.
///////////////////////////////////////////////////////////////////////////
//assert(logLevelString == NULL || strcmp(logLevelString, "CRITICAL") == 0 || strcmp(logLevelString, "INFO") == 0 || strcmp(logLevelString, "DEBUG") == 0);
assert(cactusDiskDatabaseString != NULL);
assert(speciesTree != NULL);
//////////////////////////////////////////////
//Set up logging
//////////////////////////////////////////////
st_setLogLevelFromString(logLevelString);
//////////////////////////////////////////////
//Log (some of) the inputs
//////////////////////////////////////////////
st_logInfo("Flower disk name : %s\n", cactusDiskDatabaseString);
for (j = optind; j < argc; j++) {
st_logInfo("Sequence file/directory %s\n", argv[j]);
}
//////////////////////////////////////////////
//Load the database
//////////////////////////////////////////////
stKVDatabaseConf *kvDatabaseConf = kvDatabaseConf = stKVDatabaseConf_constructFromString(cactusDiskDatabaseString);
if (stKVDatabaseConf_getType(kvDatabaseConf) == stKVDatabaseTypeTokyoCabinet || stKVDatabaseConf_getType(kvDatabaseConf)
== stKVDatabaseTypeKyotoTycoon) {
assert(stKVDatabaseConf_getDir(kvDatabaseConf) != NULL);
cactusDisk = cactusDisk_construct2(kvDatabaseConf, "cactusSequences");
} else {
cactusDisk = cactusDisk_construct(kvDatabaseConf, 1);
}
st_logInfo("Set up the flower disk\n");
//////////////////////////////////////////////
//Construct the flower
//////////////////////////////////////////////
if (cactusDisk_getFlower(cactusDisk, 0) != NULL) {
cactusDisk_destruct(cactusDisk);
st_logInfo("The first flower already exists\n");
return 0;
}
flower = flower_construct2(0, cactusDisk);
assert(flower_getName(flower) == 0);
st_logInfo("Constructed the flower\n");
//////////////////////////////////////////////
//Construct the event tree
//////////////////////////////////////////////
st_logInfo("Going to build the event tree with newick string: %s\n", speciesTree);
stTree *tree = stTree_parseNewickString(speciesTree);
st_logInfo("Parsed the tree\n");
if (makeEventHeadersAlphaNumeric) {
makeEventHeadersAlphaNumericFn(tree);
}
stTree_setBranchLength(tree, INT64_MAX);
checkBranchLengthsAreDefined(tree);
eventTree = eventTree_construct2(flower); //creates the event tree and the root even
totalEventNumber = 1;
st_logInfo("Constructed the basic event tree\n");
// Construct a set of outgroup names so that ancestral outgroups
// get recognized.
stSet *outgroupNameSet = stSet_construct3(stHash_stringKey,
stHash_stringEqualKey,
free);
if(outgroupEvents != NULL) {
stList *outgroupNames = stString_split(outgroupEvents);
for(int64_t i = 0; i < stList_length(outgroupNames); i++) {
char *outgroupName = stList_get(outgroupNames, i);
stSet_insert(outgroupNameSet, stString_copy(outgroupName));
}
stList_destruct(outgroupNames);
}
//now traverse the tree
j = optind;
assignEventsAndSequences(eventTree_getRootEvent(eventTree), tree,
outgroupNameSet, argv, &j);
char *eventTreeString = eventTree_makeNewickString(eventTree);
st_logInfo(
"Constructed the initial flower with %" PRIi64 " sequences and %" PRIi64 " events with string: %s\n",
totalSequenceNumber, totalEventNumber, eventTreeString);
assert(event_getSubTreeBranchLength(eventTree_getRootEvent(eventTree)) >= 0.0);
free(eventTreeString);
//assert(0);
//////////////////////////////////////////////
//Label any outgroup events.
//////////////////////////////////////////////
if (outgroupEvents != NULL) {
stList *outgroupEventsList = stString_split(outgroupEvents);
for (int64_t i = 0; i < stList_length(outgroupEventsList); i++) {
char *outgroupEvent = makeEventHeadersAlphaNumeric ? makeAlphaNumeric(stList_get(outgroupEventsList, i)) : stString_copy(stList_get(outgroupEventsList, i));
Event *event = eventTree_getEventByHeader(eventTree, outgroupEvent);
if (event == NULL) {
st_errAbort("Got an outgroup string that does not match an event, outgroup string %s", outgroupEvent);
}
assert(!event_isOutgroup(event));
event_setOutgroupStatus(event, 1);
assert(event_isOutgroup(event));
free(outgroupEvent);
}
stList_destruct(outgroupEventsList);
}
//////////////////////////////////////////////
//Construct the terminal group.
//////////////////////////////////////////////
if (flower_getEndNumber(flower) > 0) {
group = group_construct2(flower);
endIterator = flower_getEndIterator(flower);
while ((end = flower_getNextEnd(endIterator)) != NULL) {
end_setGroup(end, group);
}
flower_destructEndIterator(endIterator);
assert(group_isLeaf(group));
// Create a one link chain if there is only one pair of attached ends..
group_constructChainForLink(group);
assert(!flower_builtBlocks(flower));
} else {
flower_setBuiltBlocks(flower, 1);
}
///////////////////////////////////////////////////////////////////////////
// Write the flower to disk.
///////////////////////////////////////////////////////////////////////////
//flower_check(flower);
cactusDisk_write(cactusDisk);
st_logInfo("Updated the flower on disk\n");
///////////////////////////////////////////////////////////////////////////
// Cleanup.
///////////////////////////////////////////////////////////////////////////
cactusDisk_destruct(cactusDisk);
return 0; //Exit without clean up is quicker, enable cleanup when doing memory leak detection.
stSet_destruct(outgroupNameSet);
stTree_destruct(tree);
stKVDatabaseConf_destruct(kvDatabaseConf);
return 0;
}
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;
}
void testFlower_getName(CuTest* testCase) {
cactusFlowerTestSetup();
CuAssertTrue(testCase, flower_getName(flower) != NULL_NAME);
CuAssertTrue(testCase, cactusDisk_getFlower(cactusDisk, flower_getName(flower)) == flower);
cactusFlowerTestTeardown();
}
int main(int argc, char *argv[]) {
char * logLevelString = NULL;
char * cactusDiskDatabaseString = NULL;
int64_t i, j;
int64_t spanningTrees = 10;
int64_t maximumLength = 1500;
bool useProgressiveMerging = 0;
float matchGamma = 0.5;
bool useBanding = 0;
int64_t k;
stList *listOfEndAlignmentFiles = NULL;
char *endAlignmentsToPrecomputeOutputFile = NULL;
bool calculateWhichEndsToComputeSeparately = 0;
int64_t largeEndSize = 1000000;
int64_t chainLengthForBigFlower = 1000000;
int64_t longChain = 2;
char *ingroupCoverageFilePath = NULL;
int64_t minimumSizeToRescue = 1;
double minimumCoverageToRescue = 0.0;
PairwiseAlignmentParameters *pairwiseAlignmentBandingParameters = pairwiseAlignmentBandingParameters_construct();
/*
* Setup the input parameters for cactus core.
*/
bool pruneOutStubAlignments = 0;
/*
* Parse the options.
*/
while (1) {
static struct option long_options[] = { { "logLevel", required_argument, 0, 'a' }, { "cactusDisk", required_argument, 0, 'b' }, {
"help", no_argument, 0, 'h' }, { "spanningTrees", required_argument, 0, 'i' },
{ "maximumLength", required_argument, 0, 'j' }, { "useBanding", no_argument, 0, 'k' },
{ "gapGamma", required_argument, 0, 'l' }, { "matchGamma", required_argument, 0, 'L' },
{ "splitMatrixBiggerThanThis", required_argument, 0, 'o' }, { "anchorMatrixBiggerThanThis",
required_argument, 0, 'p' }, { "repeatMaskMatrixBiggerThanThis", required_argument, 0, 'q' }, {
"diagonalExpansion", required_argument, 0, 'r' }, { "constraintDiagonalTrim", required_argument, 0, 't' }, {
"minimumDegree", required_argument, 0, 'u' }, { "alignAmbiguityCharacters", no_argument, 0, 'w' }, {
"pruneOutStubAlignments", no_argument, 0, 'y' }, {
"minimumIngroupDegree", required_argument, 0, 'A' }, { "minimumOutgroupDegree", required_argument, 0, 'B' },
{ "precomputedAlignments", required_argument, 0, 'D' }, {
"endAlignmentsToPrecomputeOutputFile", required_argument, 0, 'E' }, { "useProgressiveMerging",
no_argument, 0, 'F' }, { "calculateWhichEndsToComputeSeparately", no_argument, 0, 'G' }, { "largeEndSize",
required_argument, 0, 'I' },
{"ingroupCoverageFile", required_argument, 0, 'J'},
{"minimumSizeToRescue", required_argument, 0, 'K'},
{"minimumCoverageToRescue", required_argument, 0, 'M'},
{ "minimumNumberOfSpecies", required_argument, 0, 'N' },
{ 0, 0, 0, 0 } };
int option_index = 0;
int key = getopt_long(argc, argv, "a:b:hi:j:kl:o:p:q:r:t:u:wy:A:B:D:E:FGI:J:K:L:M:N:", long_options, &option_index);
if (key == -1) {
break;
}
switch (key) {
case 'a':
logLevelString = stString_copy(optarg);
st_setLogLevelFromString(logLevelString);
break;
case 'b':
cactusDiskDatabaseString = stString_copy(optarg);
break;
case 'h':
usage();
return 0;
case 'i':
i = sscanf(optarg, "%" PRIi64 "", &spanningTrees);
(void) i;
assert(i == 1);
assert(spanningTrees >= 0);
break;
case 'j':
i = sscanf(optarg, "%" PRIi64 "", &maximumLength);
assert(i == 1);
assert(maximumLength >= 0);
break;
case 'k':
useBanding = !useBanding;
break;
case 'l':
i = sscanf(optarg, "%f", &pairwiseAlignmentBandingParameters->gapGamma);
assert(i == 1);
assert(pairwiseAlignmentBandingParameters->gapGamma >= 0.0);
break;
case 'L':
i = sscanf(optarg, "%f", &matchGamma);
assert(i == 1);
assert(matchGamma >= 0.0);
break;
case 'o':
i = sscanf(optarg, "%" PRIi64 "", &k);
assert(i == 1);
assert(k >= 0);
pairwiseAlignmentBandingParameters->splitMatrixBiggerThanThis = (int64_t) k * k;
break;
case 'p':
i = sscanf(optarg, "%" PRIi64 "", &k);
assert(i == 1);
assert(k >= 0);
pairwiseAlignmentBandingParameters->anchorMatrixBiggerThanThis = (int64_t) k * k;
break;
case 'q':
i = sscanf(optarg, "%" PRIi64 "", &k);
assert(i == 1);
assert(k >= 0);
pairwiseAlignmentBandingParameters->repeatMaskMatrixBiggerThanThis = (int64_t) k * k;
break;
case 'r':
i = sscanf(optarg, "%" PRIi64 "", &pairwiseAlignmentBandingParameters->diagonalExpansion);
assert(i == 1);
assert(pairwiseAlignmentBandingParameters->diagonalExpansion >= 0);
assert(pairwiseAlignmentBandingParameters->diagonalExpansion % 2 == 0);
break;
case 't':
i = sscanf(optarg, "%" PRIi64 "", &pairwiseAlignmentBandingParameters->constraintDiagonalTrim);
assert(i == 1);
assert(pairwiseAlignmentBandingParameters->constraintDiagonalTrim >= 0);
break;
case 'u':
i = sscanf(optarg, "%" PRIi64 "", &minimumDegree);
assert(i == 1);
break;
case 'w':
pairwiseAlignmentBandingParameters->alignAmbiguityCharacters = 1;
break;
case 'y':
pruneOutStubAlignments = 1;
break;
case 'A':
i = sscanf(optarg, "%" PRIi64 "", &minimumIngroupDegree);
assert(i == 1);
break;
case 'B':
i = sscanf(optarg, "%" PRIi64 "", &minimumOutgroupDegree);
assert(i == 1);
break;
case 'D':
listOfEndAlignmentFiles = stString_split(optarg);
break;
case 'E':
endAlignmentsToPrecomputeOutputFile = stString_copy(optarg);
break;
case 'F':
useProgressiveMerging = 1;
break;
case 'G':
calculateWhichEndsToComputeSeparately = 1;
break;
case 'I':
i = sscanf(optarg, "%" PRIi64 "", &largeEndSize);
assert(i == 1);
break;
case 'J':
ingroupCoverageFilePath = stString_copy(optarg);
break;
case 'K':
i = sscanf(optarg, "%" PRIi64, &minimumSizeToRescue);
assert(i == 1);
break;
case 'M':
i = sscanf(optarg, "%lf", &minimumCoverageToRescue);
assert(i == 1);
break;
case 'N':
i = sscanf(optarg, "%" PRIi64, &minimumNumberOfSpecies);
if (i != 1) {
st_errAbort("Error parsing minimumNumberOfSpecies parameter");
}
break;
default:
usage();
return 1;
}
}
st_setLogLevelFromString(logLevelString);
/*
* Load the flowerdisk
*/
stKVDatabaseConf *kvDatabaseConf = stKVDatabaseConf_constructFromString(cactusDiskDatabaseString);
CactusDisk *cactusDisk = cactusDisk_construct(kvDatabaseConf, 0); //We precache the sequences
st_logInfo("Set up the flower disk\n");
/*
* Load the hmm
*/
StateMachine *sM = stateMachine5_construct(fiveState);
/*
* For each flower.
*/
if (calculateWhichEndsToComputeSeparately) {
stList *flowers = flowerWriter_parseFlowersFromStdin(cactusDisk);
if (stList_length(flowers) != 1) {
st_errAbort("We are breaking up a flower's end alignments for precomputation but we have %" PRIi64 " flowers.\n", stList_length(flowers));
}
stSortedSet *endsToAlignSeparately = getEndsToAlignSeparately(stList_get(flowers, 0), maximumLength, largeEndSize);
assert(stSortedSet_size(endsToAlignSeparately) != 1);
stSortedSetIterator *it = stSortedSet_getIterator(endsToAlignSeparately);
End *end;
while ((end = stSortedSet_getNext(it)) != NULL) {
fprintf(stdout, "%s\t%" PRIi64 "\t%" PRIi64 "\n", cactusMisc_nameToStringStatic(end_getName(end)), end_getInstanceNumber(end), getTotalAdjacencyLength(end));
}
return 0; //avoid cleanup costs
stSortedSet_destructIterator(it);
stSortedSet_destruct(endsToAlignSeparately);
} else if (endAlignmentsToPrecomputeOutputFile != NULL) {
/*
* In this case we will align a set of end and save the alignments in a file.
*/
stList *names = flowerWriter_parseNames(stdin);
Flower *flower = cactusDisk_getFlower(cactusDisk, *((Name *)stList_get(names, 0)));
FILE *fileHandle = fopen(endAlignmentsToPrecomputeOutputFile, "w");
for(int64_t i=1; i<stList_length(names); i++) {
End *end = flower_getEnd(flower, *((Name *)stList_get(names, i)));
if (end == NULL) {
st_errAbort("The end %" PRIi64 " was not found in the flower\n", *((Name *)stList_get(names, i)));
}
stSortedSet *endAlignment = makeEndAlignment(sM, end, spanningTrees, maximumLength, useProgressiveMerging,
matchGamma, pairwiseAlignmentBandingParameters);
writeEndAlignmentToDisk(end, endAlignment, fileHandle);
stSortedSet_destruct(endAlignment);
}
fclose(fileHandle);
return 0; //avoid cleanup costs
stList_destruct(names);
st_logInfo("Finished precomputing end alignments\n");
} else {
/*
* Compute complete flower alignments, possibly loading some precomputed alignments.
*/
bedRegion *bedRegions = NULL;
size_t numBeds = 0;
if (ingroupCoverageFilePath != NULL) {
// Pre-load the mmap for the coverage file.
FILE *coverageFile = fopen(ingroupCoverageFilePath, "rb");
if (coverageFile == NULL) {
st_errnoAbort("Opening coverage file %s failed",
ingroupCoverageFilePath);
}
fseek(coverageFile, 0, SEEK_END);
int64_t coverageFileLen = ftell(coverageFile);
assert(coverageFileLen >= 0);
assert(coverageFileLen % sizeof(bedRegion) == 0);
if (coverageFileLen == 0) {
// mmap doesn't like length-0 mappings, for obvious
// reasons. Pretend that the coverage file doesn't
// exist in this case, since it contains no data.
ingroupCoverageFilePath = NULL;
} else {
// Establish a memory mapping for the file.
bedRegions = mmap(NULL, coverageFileLen, PROT_READ, MAP_SHARED,
fileno(coverageFile), 0);
if (bedRegions == MAP_FAILED) {
st_errnoAbort("Failure mapping coverage file");
}
numBeds = coverageFileLen / sizeof(bedRegion);
}
fclose(coverageFile);
}
stList *flowers = flowerWriter_parseFlowersFromStdin(cactusDisk);
if (listOfEndAlignmentFiles != NULL && stList_length(flowers) != 1) {
st_errAbort("We have precomputed alignments but %" PRIi64 " flowers to align.\n", stList_length(flowers));
}
cactusDisk_preCacheStrings(cactusDisk, flowers);
for (j = 0; j < stList_length(flowers); j++) {
flower = stList_get(flowers, j);
st_logInfo("Processing a flower\n");
stSortedSet *alignedPairs = makeFlowerAlignment3(sM, flower, listOfEndAlignmentFiles, spanningTrees, maximumLength,
useProgressiveMerging, matchGamma, pairwiseAlignmentBandingParameters, pruneOutStubAlignments);
st_logInfo("Created the alignment: %" PRIi64 " pairs\n", stSortedSet_size(alignedPairs));
stPinchIterator *pinchIterator = stPinchIterator_constructFromAlignedPairs(alignedPairs, getNextAlignedPairAlignment);
/*
* Run the cactus caf functions to build cactus.
*/
stPinchThreadSet *threadSet = stCaf_setup(flower);
stCaf_anneal(threadSet, pinchIterator, NULL);
if (minimumDegree < 2) {
stCaf_makeDegreeOneBlocks(threadSet);
}
if (minimumIngroupDegree > 0 || minimumOutgroupDegree > 0 || minimumDegree > 1) {
stCaf_melt(flower, threadSet, blockFilterFn, 0, 0, 0, INT64_MAX);
}
if (ingroupCoverageFilePath != NULL) {
// Rescue any sequence that is covered by outgroups
// but currently unaligned into single-degree blocks.
stPinchThreadSetIt pinchIt = stPinchThreadSet_getIt(threadSet);
stPinchThread *thread;
while ((thread = stPinchThreadSetIt_getNext(&pinchIt)) != NULL) {
Cap *cap = flower_getCap(flower,
stPinchThread_getName(thread));
assert(cap != NULL);
Sequence *sequence = cap_getSequence(cap);
assert(sequence != NULL);
rescueCoveredRegions(thread, bedRegions, numBeds,
sequence_getName(sequence),
minimumSizeToRescue,
minimumCoverageToRescue);
}
stCaf_joinTrivialBoundaries(threadSet);
}
stCaf_finish(flower, threadSet, chainLengthForBigFlower, longChain, INT64_MAX, INT64_MAX); //Flower now destroyed.
stPinchThreadSet_destruct(threadSet);
st_logInfo("Ran the cactus core script.\n");
/*
* Cleanup
*/
//Clean up the sorted set after cleaning up the iterator
stPinchIterator_destruct(pinchIterator);
stSortedSet_destruct(alignedPairs);
st_logInfo("Finished filling in the alignments for the flower\n");
}
stList_destruct(flowers);
//st_errAbort("Done\n");
/*
* Write and close the cactusdisk.
*/
cactusDisk_write(cactusDisk);
return 0; //Exit without clean up is quicker, enable cleanup when doing memory leak detection.
if (bedRegions != NULL) {
// Clean up our mapping.
munmap(bedRegions, numBeds * sizeof(bedRegion));
}
}
///////////////////////////////////////////////////////////////////////////
// Cleanup
///////////////////////////////////////////////////////////////////////////
stateMachine_destruct(sM);
cactusDisk_destruct(cactusDisk);
stKVDatabaseConf_destruct(kvDatabaseConf);
//destructCactusCoreInputParameters(cCIP);
free(cactusDiskDatabaseString);
if (listOfEndAlignmentFiles != NULL) {
stList_destruct(listOfEndAlignmentFiles);
}
if (logLevelString != NULL) {
free(logLevelString);
}
st_logInfo("Finished with the flower disk for this flower.\n");
//while(1);
return 0;
}