tree[0].heirarchyID = "0";
maxLevel = 0;
calcTotals = true;
+ addSeqToTree("unknown", "unknown;");
}
catch(exception& e) {
m->errorOut(e, "PhyloTree", "PhyloTree");
istringstream iss (tempBuf,istringstream::in);
delete buffer;
- iss >> numNodes; gobble(iss);
+ //read version
+ m->getline(iss); m->gobble(iss);
+
+ iss >> numNodes; m->gobble(iss);
tree.resize(numNodes);
for (int i = 0; i < tree.size(); i++) {
- iss >> tree[i].name >> tree[i].level >> tree[i].parent; gobble(iss);
+ iss >> tree[i].name >> tree[i].level >> tree[i].parent; m->gobble(iss);
}
//read genus nodes
int numGenus = 0;
- iss >> numGenus; gobble(iss);
+ iss >> numGenus; m->gobble(iss);
int gnode, gsize;
totals.clear();
for (int i = 0; i < numGenus; i++) {
- iss >> gnode >> gsize; gobble(iss);
+ iss >> gnode >> gsize; m->gobble(iss);
uniqueTaxonomies[gnode] = gnode;
totals.push_back(gsize);
MPI_File_close(&inMPI);
#else
- in >> numNodes; gobble(in);
+ //read version
+ string line = m->getline(in); m->gobble(in);
+
+ in >> numNodes; m->gobble(in);
tree.resize(numNodes);
for (int i = 0; i < tree.size(); i++) {
- in >> tree[i].name >> tree[i].level >> tree[i].parent; gobble(in);
+ in >> tree[i].name >> tree[i].level >> tree[i].parent; m->gobble(in);
}
//read genus nodes
int numGenus = 0;
- in >> numGenus; gobble(in);
+ in >> numGenus; m->gobble(in);
int gnode, gsize;
totals.clear();
for (int i = 0; i < numGenus; i++) {
- in >> gnode >> gsize; gobble(in);
+ in >> gnode >> gsize; m->gobble(in);
uniqueTaxonomies[gnode] = gnode;
totals.push_back(gsize);
maxLevel = 0;
calcTotals = true;
string name, tax;
-
#ifdef USE_MPI
int pid, num, processors;
- vector<unsigned long int> positions;
+ vector<unsigned long long> positions;
MPI_Status status;
MPI_File inMPI;
MPI_File_open(MPI_COMM_WORLD, inFileName, MPI_MODE_RDONLY, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer
if (pid == 0) {
- positions = setFilePosEachLine(tfile, num);
+ positions = m->setFilePosEachLine(tfile, num);
//send file positions to all processes
for(int i = 1; i < processors; i++) {
#else
ifstream in;
- openInputFile(tfile, in);
+ m->openInputFile(tfile, in);
//read in users taxonomy file and add sequences to tree
while(!in.eof()){
- in >> name >> tax; gobble(in);
-
+ in >> name >> tax; m->gobble(in);
+
addSeqToTree(name, tax);
}
in.close();
#endif
-
+
assignHeirarchyIDs(0);
-
+
+
+ string unknownTax = "unknown;";
+ //added last taxon until you get desired level
+ for (int i = 1; i < maxLevel; i++) {
+ unknownTax += "unclassfied;";
+ }
+
+ addSeqToTree("unknown", unknownTax);
+
//create file for summary if needed
setUp(tfile);
}
int PhyloTree::addSeqToTree(string seqName, string seqTaxonomy){
try {
-
numSeqs++;
map<string, int>::iterator childPointer;
int level = 1;
tree[0].accessions.push_back(seqName);
+ m->removeConfidences(seqTaxonomy);
+
string taxon;// = getNextTaxon(seqTaxonomy);
while(seqTaxonomy != ""){
if (seqTaxonomy == "") { uniqueTaxonomies[currentNode] = currentNode; }
}
+
+ return 0;
}
catch(exception& e) {
m->errorOut(e, "PhyloTree", "addSeqToTree");
if (pid == 0) { binUnclassified(taxFileNameTest); }
#else
- //create file needed for summary if it doesn't exist
- ifstream FileTest(taxFileNameTest.c_str());
-
- if (!FileTest) {
- binUnclassified(taxFileNameTest);
- }
+ binUnclassified(taxFileNameTest);
#endif
}
catch(exception& e) {
try {
ofstream out;
- openOutputFile(file, out);
+ m->openOutputFile(file, out);
map<string, int>::iterator itBin;
map<string, int>::iterator childPointer;
vector<TaxNode> copy = tree;
-
+
//fill out tree
fillOutTree(0, copy);
leafNodes[i] = i;
}
}
- //cout << "maxLevel = " << maxLevel << endl;
- int copyNodes = copy.size();
+ int copyNodes = copy.size();
+
//go through the seqs and if a sequence finest taxon is not the same level as the most finely defined taxon then classify it as unclassified where necessary
map<int, int>::iterator itLeaf;
for (itLeaf = leafNodes.begin(); itLeaf != leafNodes.end(); itLeaf++) {
/**************************************************************************************************/
void PhyloTree::fillOutTree(int index, vector<TaxNode>& copy) {
try {
+
map<string,int>::iterator it;
- if (copy[index].level < maxLevel) {
- it = copy[index].children.find("unclassified");
- if (it == copy[index].children.end()) { //no unclassified at this level
- string taxon = "unclassified";
- copy.push_back(TaxNode(taxon));
- copy[index].children[taxon] = copy.size()-1;
- copy[copy.size()-1].parent = index;
- copy[copy.size()-1].level = copy[index].level + 1;
- }
+ it = copy[index].children.find("unclassified");
+ if (it == copy[index].children.end()) { //no unclassified at this level
+ string taxon = "unclassified";
+ copy.push_back(TaxNode(taxon));
+ copy[index].children[taxon] = copy.size()-1;
+ copy[copy.size()-1].parent = index;
+ copy[copy.size()-1].level = copy[index].level + 1;
+ }
-
- for(it=copy[index].children.begin();it!=copy[index].children.end();it++){ //check your children
+ if (tree[index].level < maxLevel) {
+ for(it=tree[index].children.begin();it!=tree[index].children.end();it++){ //check your children
fillOutTree(it->second, copy);
}
-
}
-
+
}
catch(exception& e) {
m->errorOut(e, "PhyloTree", "fillOutTree");
void PhyloTree::print(ofstream& out, vector<TaxNode>& copy){
try {
+
+ //output mothur version
+ out << "#" << m->getVersion() << endl;
+
out << copy.size() << endl;
+ out << maxLevel << endl;
+
for (int i = 0; i < copy.size(); i++) {
-
+
out << copy[i].level << '\t'<< copy[i].name << '\t' << copy[i].children.size() << '\t';
map<string,int>::iterator it;
#endif
ofstream outTree;
- openOutputFile(treefilename, outTree);
+ m->openOutputFile(treefilename, outTree);
+
+ //output mothur version
+ outTree << "#" << m->getVersion() << endl;
//print treenodes
outTree << tree.size() << endl;
try {
bool okay = true;
+ templateFileNames.push_back("unknown");
map<string, int>::iterator itFind;
map<string, int> taxonomyFileNames = name2Taxonomy;
okay = false;
}
- templateFileNames.erase(templateFileNames.begin()+i);
- i--;
+ //templateFileNames.erase(templateFileNames.begin()+i);
+ //i--;
}
+ templateFileNames.clear();
if (taxonomyFileNames.size() > 0) { //there are names in tax file that are not in template
okay = false;