static char *tree_getNewickTreeStringP(stTree *tree) {
char *cA, *cA2;
if(stTree_getChildNumber(tree) > 0) {
int32_t i;
cA = stString_copy("(");
for(i=0; i<stTree_getChildNumber(tree); i++) {
cA2 = tree_getNewickTreeStringP(stTree_getChild(tree, i));
char *cA3 = stString_print((i+1 < stTree_getChildNumber(tree) ? "%s%s," : "%s%s"), cA, cA2);
free(cA);
free(cA2);
cA = cA3;
}
cA2 = stString_print("%s)", cA);
free(cA);
cA = cA2;
}
else {
cA = stString_copy("");
}
if(stTree_getLabel(tree) != NULL) {
cA2 = stString_print("%s%s", cA, stTree_getLabel(tree));
free(cA);
cA = cA2;
}
if(stTree_getBranchLength(tree) != INFINITY) {
char *cA2 = stString_print("%s:%g", cA, stTree_getBranchLength(tree));
free(cA);
cA = cA2;
}
return cA;
}
// clone this node, make its supertree a child, and clone all children
// other than oldNode, leaving this node as a child of nodeToAddTo
static void tree_cloneFlippedTree(stTree *node, stTree *oldNode,
stTree *nodeToAddTo,
double branchLength) {
if(stTree_getParent(node) != NULL) {
// This node isn't the root
stTree *clonedNode = stTree_cloneNode(node);
stTree_setParent(clonedNode, nodeToAddTo);
stTree_setBranchLength(clonedNode, branchLength);
// Clone its children (other than oldNode) and their subtrees
for(int64_t i = 0; i < stTree_getChildNumber(node); i++) {
stTree *child = stTree_getChild(node, i);
if(child != oldNode) {
stTree *clonedChild = stTree_clone(child);
stTree_setParent(clonedChild, clonedNode);
}
}
// Recurse on the parent of this node.
tree_cloneFlippedTree(stTree_getParent(node), node,
clonedNode, stTree_getBranchLength(node));
} else {
// We have to treat the root specially, because we're going to
// eliminate it. Just add all the other children of the root
// as children of nodeToAddTo.
for(int64_t i = 0; i < stTree_getChildNumber(node); i++) {
stTree *child = stTree_getChild(node, i);
if(child != oldNode) {
stTree *clonedChild = stTree_clone(child);
stTree_setParent(clonedChild, nodeToAddTo);
stTree_setBranchLength(clonedChild, stTree_getBranchLength(child) + branchLength);
}
}
}
}
static void assignEventsAndSequences(Event *parentEvent, stTree *tree,
stSet *outgroupNameSet,
char *argv[], int64_t *j) {
Event *myEvent = NULL; // To distinguish from the global "event" variable.
assert(tree != NULL);
totalEventNumber++;
if (stTree_getChildNumber(tree) > 0) {
myEvent = event_construct3(stTree_getLabel(tree),
stTree_getBranchLength(tree), parentEvent,
eventTree);
for (int64_t i = 0; i < stTree_getChildNumber(tree); i++) {
assignEventsAndSequences(myEvent, stTree_getChild(tree, i),
outgroupNameSet, argv, j);
}
}
if (stTree_getChildNumber(tree) == 0 || (stTree_getLabel(tree) != NULL && (stSet_search(outgroupNameSet, (char *)stTree_getLabel(tree)) != NULL))) {
// This event is a leaf and/or an outgroup, so it has
// associated sequence.
assert(stTree_getLabel(tree) != NULL);
assert(stTree_getBranchLength(tree) != INFINITY);
if (stTree_getChildNumber(tree) == 0) {
// Construct the leaf event
myEvent = event_construct3(stTree_getLabel(tree), stTree_getBranchLength(tree), parentEvent, eventTree);
}
char *fileName = argv[*j];
if (!stFile_exists(fileName)) {
st_errAbort("File does not exist: %s\n", fileName);
}
// Set the global "event" variable, which is needed for the
// function provided to fastaReadToFunction.
event = myEvent;
if (stFile_isDir(fileName)) {
st_logInfo("Processing directory: %s\n", fileName);
stList *filesInDir = stFile_getFileNamesInDirectory(fileName);
for (int64_t i = 0; i < stList_length(filesInDir); i++) {
char *absChildFileName = stFile_pathJoin(fileName, stList_get(filesInDir, i));
assert(stFile_exists(absChildFileName));
setCompleteStatus(absChildFileName); //decide if the sequences in the file should be free or attached.
FILE *fileHandle = fopen(absChildFileName, "r");
fastaReadToFunction(fileHandle, processSequence);
fclose(fileHandle);
free(absChildFileName);
}
stList_destruct(filesInDir);
} else {
st_logInfo("Processing file: %s\n", fileName);
setCompleteStatus(fileName); //decide if the sequences in the file should be free or attached.
FILE *fileHandle = fopen(fileName, "r");
fastaReadToFunction(fileHandle, processSequence);
fclose(fileHandle);
}
(*j)++;
}
}
/* recursive clone children of node */
static void subrangeCloneChildren(stTree *srcParent, struct malnCompCompMap *srcDestCompMap, stList *pendingSubtrees) {
// children are in on-stack array and then passed up pendingSubtrees
int numDestChildren = 0;
stTree *destChildren[stTree_getChildNumber(srcParent)];
for (int i = 0; i < stTree_getChildNumber(srcParent); i++) {
stTree *destNode = subrangeCloneNode(stTree_getChild(srcParent, i), srcDestCompMap, pendingSubtrees);
if (destNode != NULL) {
destChildren[numDestChildren++] = destNode;
}
}
subrangeSavePendingChildren(numDestChildren, destChildren, pendingSubtrees);
}
int stTree_getNumNodes(stTree *root) {
int cnt = 1; // this node
for (int i = 0; i < stTree_getChildNumber(root); i++) {
cnt += stTree_getNumNodes(stTree_getChild(root, i));
}
return cnt;
}
/* recursively clone a tree */
static stTree *cloneTree(stTree *srcNode, stTree *destParent, struct malnCompCompMap *srcDestCompMap) {
stTree *destNode = cloneNode(srcNode, srcDestCompMap);
stTree_setParent(destNode, destParent);
for (int i = 0; i < stTree_getChildNumber(srcNode); i++) {
cloneTree(stTree_getChild(srcNode, i), destNode, srcDestCompMap);
}
return destNode;
}
// Set client data to NULL (optionally recursively).
static void stTree_clearClientData(stTree *tree, bool recursive) {
stTree_setClientData(tree, NULL);
if (recursive) {
for (int64_t i = 0; i < stTree_getChildNumber(tree); i++) {
stTree_clearClientData(stTree_getChild(tree, i), true);
}
}
}
/* recursively clone a tree */
static stTree *tree_clonetree(stTree *node, stTree *parent2) {
stTree *node2 = stTree_cloneNode(node);
stTree_setParent(node2, parent2);
for (int i = 0; i < stTree_getChildNumber(node); i++) {
tree_clonetree(stTree_getChild(node, i), node2);
}
return node2;
}
void stTree_sortChildren(stTree *root, int cmpFn(stTree *a, stTree *b)) {
sortChildrenCmpFn = cmpFn;
stList_sort(root->nodes, sortChildrenListCmpFn);
sortChildrenCmpFn = NULL;
for (int i = 0; i < stTree_getChildNumber(root); i++) {
stTree_sortChildren(stTree_getChild(root, i), cmpFn);
}
}
void makeEventHeadersAlphaNumericFn(stTree *tree) {
char *cA = makeAlphaNumeric(stTree_getLabel(tree));
stTree_setLabel(tree, cA);
free(cA);
for (int64_t i = 0; i < stTree_getChildNumber(tree); i++) {
makeEventHeadersAlphaNumericFn(stTree_getChild(tree, i));
}
}
void checkBranchLengthsAreDefined(stTree *tree) {
if (isinf(stTree_getBranchLength(tree))) {
st_errAbort("Got a non defined branch length in the input tree: %s.\n", stTree_getNewickTreeString(tree));
}
for (int64_t i = 0; i < stTree_getChildNumber(tree); i++) {
checkBranchLengthsAreDefined(stTree_getChild(tree, i));
}
}
/* get location type in tree */
enum mafTreeLoc mafTreeNodeCompLink_getLoc(struct mafTreeNodeCompLink *ncLink) {
if (stTree_getParent(ncLink->node) == NULL) {
return mafTreeLocRoot;
} else if (stTree_getChildNumber(ncLink->node) == 0) {
return mafTreeLocLeaf;
} else {
return mafTreeLocInternal;
}
}
/* assert that the tree has no loops (same genome at two levels) */
static void assertNoLoops(struct Genome *genome, stTree *root) {
for (int i = 0; i < stTree_getChildNumber(root); i++) {
stTree *subNode = stTree_getChild(root, i);
if (getNodeComp(subNode)->seq->genome == genome) {
errAbort("genome occurs at two levels in the tree: %s", genome->name);
}
assertNoLoops(genome, subNode);
}
}
bool stTree_equals(stTree *tree1, stTree *tree2) {
if (stTree_getBranchLength(tree1) != stTree_getBranchLength(tree2)) {
return false;
}
if (!stString_eq(stTree_getLabel(tree1), stTree_getLabel(tree2))) {
return false;
}
int numChildren = stTree_getChildNumber(tree1);
if (stTree_getChildNumber(tree2) != numChildren) {
return false;
}
for (int i = 0; i < numChildren; i++) {
if (!stTree_equals(stTree_getChild(tree1, i), stTree_getChild(tree2, i))) {
return false;
}
}
return true;
}
/* DFS to fill in table of node links and link back with clientData */
static void fillNodeCompLinksDFS(mafTree *mTree, stTree *node, int *treeOrder, struct malnComp *treeComps[]) {
for (int i = 0; i < stTree_getChildNumber(node); i++) {
fillNodeCompLinksDFS(mTree, stTree_getChild(node, i), treeOrder, treeComps);
}
struct mafTreeNodeCompLink *ncLink = mafTreeNodeCompLink_construct(*treeOrder, node, treeComps[*treeOrder]);
(*treeOrder)++;
if (!sameString(ncLink->comp->seq->orgSeqName, stTree_getLabel(node))) {
errAbort("tree component name \"%s\" doesn't match tree node name \"%s\"", ncLink->comp->seq->orgSeqName, stTree_getLabel(node));
}
}
/* recursively find a species tree node *below* the specified node */
static stTree *speciesTreeFindBelow(stTree *speciesRoot, struct Genome *genome) {
stTree *genomeNode = NULL;
for (int i = 0; (genomeNode == NULL) && (i < stTree_getChildNumber(speciesRoot)); i++) {
stTree *sn = stTree_getChild(speciesRoot, i);
if (speciesTreeGetGenome(sn) == genome) {
genomeNode = sn;
} else {
genomeNode = speciesTreeFindBelow(sn, genome);
}
}
return genomeNode;
}
/* recursively search a tree for node linked to the specified component */
static stTree *searchByComp(stTree *node, struct malnComp *comp) {
struct mafTreeNodeCompLink *ncLink = getNodeCompLink(node);
if (ncLink->comp == comp) {
return node;
}
for (int i = 0; i < stTree_getChildNumber(node); i++) {
stTree *hit = searchByComp(stTree_getChild(node, i), comp);
if (hit != NULL) {
return hit;
}
}
return NULL;
}
/* recursive dump */
static void dumpSubtree(stTree *root, FILE *fh, int indent) {
fprintf(fh, "%*s", 4*indent, "");
struct malnComp *comp = getNodeComp(root);
if (comp == NULL) {
fprintf(fh, "%s", stTree_getLabel(root));
} else {
malnComp_prInfo(comp, fh);
}
fputc('\n', fh);
for (int i = 0; i < stTree_getChildNumber(root); i++) {
dumpSubtree(stTree_getChild(root, i), fh, indent+1);
}
}
/* DFS to set or check tree order after a join. */
static void setCheckTreeOrderDFS(mafTree *mTree, stTree *node, bool check, int *treeOrder) {
for (int i = 0; i < stTree_getChildNumber(node); i++) {
setCheckTreeOrderDFS(mTree, stTree_getChild(node, i), check, treeOrder);
}
struct mafTreeNodeCompLink *ncLink = getNodeCompLink(node);
if (!sameString(ncLink->comp->seq->orgSeqName, stTree_getLabel(node))) {
errAbort("tree component name \"%s\" doesn't match tree node name \"%s\"", ncLink->comp->seq->orgSeqName, stTree_getLabel(node));
}
if (!check) {
ncLink->treeOrder = *treeOrder;
} else if (ncLink->treeOrder != *treeOrder) {
errAbort("expected tree order (%d) doesn't match actual tree node order (%d) for \"%s\"", *treeOrder, ncLink->treeOrder, stTree_getLabel(node));
}
(*treeOrder)++;
}
/* Remove a node from the tree and free. Can't delete the root node. */
void mafTree_deleteNode(mafTree *mTree, struct mafTreeNodeCompLink *ncLink) {
stTree *node = ncLink->node;
stTree *parent = stTree_getParent(node);
if (parent == NULL) {
errAbort("BUG: can't remove tree root node");
}
stTree_setParent(node, NULL);
// setParent changes node children
while (stTree_getChildNumber(node) > 0) {
stTree_setParent(stTree_getChild(node, 0), parent);
}
freeMafTreeNodeCompLinks(node);
stTree_destruct(node);
setCheckTreeOrder(mTree, false);
}
/* recursively verify the tree */
static void speciesTreeBlkTreeVerify(stTree *speciesNode, stTree *blkNode) {
speciesNode = speciesTreeFindAtBelow(speciesNode, getNodeComp(blkNode)->seq->genome);
if (speciesNode == NULL) {
speciesTreeMismatchError(blkNode);
} else {
for (int i = 0; i < stTree_getChildNumber(blkNode); i++) {
stTree *blkSubNode = stTree_getChild(blkNode, i);
stTree *speciesSubNode = speciesTreeFindAtBelow(speciesNode, getNodeComp(blkSubNode)->seq->genome);
if (speciesSubNode == NULL) {
speciesTreeMismatchError(blkNode);
} else {
speciesTreeBlkTreeVerify(speciesSubNode, blkSubNode);
}
}
}
}
void stTree_setChild(stTree *tree, int64_t i, stTree *child) {
assert(i >= 0);
assert(i < stTree_getChildNumber(tree));
stList_set(tree->nodes, i, child);
}
/* recursively free mafTreeNodeCompLink objects */
static void freeMafTreeNodeCompLinks(stTree *node) {
mafTreeNodeCompLink_destruct(getNodeCompLink(node));
for (int i = 0; i < stTree_getChildNumber(node); i++) {
freeMafTreeNodeCompLinks(stTree_getChild(node, i));
}
}
/* add genome objects as client data */
static void speciesTreeAddLinks(stTree *speciesNode, struct Genomes *genomes) {
stTree_setClientData(speciesNode, genomesGetGenome(genomes, stTree_getLabel(speciesNode)));
for (int i = 0; i < stTree_getChildNumber(speciesNode); i++) {
speciesTreeAddLinks(stTree_getChild(speciesNode, i), genomes);
}
}
/* clone child and append clones to a give parent node */
static void cloneChildren(stTree *srcParent, stTree *destParent, struct malnCompCompMap *srcDestCompMap) {
for (int i = 0; i < stTree_getChildNumber(srcParent); i++) {
stTree_setParent(cloneTree(stTree_getChild(srcParent, i), destParent, srcDestCompMap), destParent);
}
}