try {
m = MothurOut::getInstance();
calcTotals = false;
+ numNodes = 0;
+ numSeqs = 0;
#ifdef USE_MPI
MPI_File inMPI;
#ifdef USE_MPI
int pid, num, processors;
- vector<long> positions;
+ vector<unsigned long int> positions;
MPI_Status status;
MPI_File inMPI;
/**************************************************************************************************/
-string PhyloTree::getNextTaxon(string& heirarchy){
+string PhyloTree::getNextTaxon(string& heirarchy, string seqname){
try {
string currentLevel = "";
if(heirarchy != ""){
int pos = heirarchy.find_first_of(';');
- currentLevel=heirarchy.substr(0,pos);
- if (pos != (heirarchy.length()-1)) { heirarchy=heirarchy.substr(pos+1); }
- else { heirarchy = ""; }
+
+ if (pos == -1) { //you can't find another ;
+ currentLevel = heirarchy;
+ heirarchy = "";
+ m->mothurOut(seqname + " is missing a ;, please check for other errors."); m->mothurOutEndLine();
+ }else{
+ currentLevel=heirarchy.substr(0,pos);
+ if (pos != (heirarchy.length()-1)) { heirarchy=heirarchy.substr(pos+1); }
+ else { heirarchy = ""; }
+ }
+
}
return currentLevel;
}
int PhyloTree::addSeqToTree(string seqName, string seqTaxonomy){
try {
+
numSeqs++;
map<string, int>::iterator childPointer;
tree[0].accessions.push_back(seqName);
string taxon;// = getNextTaxon(seqTaxonomy);
-
+
while(seqTaxonomy != ""){
level++;
-
+
if (m->control_pressed) { return 0; }
//somehow the parent is getting one too many accnos
//use print to reassign the taxa id
- taxon = getNextTaxon(seqTaxonomy);
+ taxon = getNextTaxon(seqTaxonomy, seqName);
+
+ if (taxon == "") { m->mothurOut(seqName + " has an error in the taxonomy. This may be due to a ;;"); m->mothurOutEndLine(); if (currentNode != 0) { uniqueTaxonomies[currentNode] = currentNode; } break; }
childPointer = tree[currentNode].children.find(taxon);
tree[currentNode].children[taxon] = numNodes-1;
tree[numNodes-1].parent = currentNode;
- // int numChildren = tree[currentNode].children.size();
- // string heirarchyID = tree[currentNode].heirarchyID;
- // tree[currentNode].accessions.push_back(seqName);
-
currentNode = tree[currentNode].children[taxon];
tree[currentNode].accessions.push_back(seqName);
name2Taxonomy[seqName] = currentNode;
- // tree[currentNode].level = level;
- // tree[currentNode].childNumber = numChildren;
- // tree[currentNode].heirarchyID = heirarchyID + '.' + toString(tree[currentNode].childNumber);
}
-
+
if (seqTaxonomy == "") { uniqueTaxonomies[currentNode] = currentNode; }
}
-
}
catch(exception& e) {
m->errorOut(e, "PhyloTree", "addSeqToTree");
map<string, int>::iterator childPointer;
vector<TaxNode> copy = tree;
-
+
//fill out tree
fillOutTree(0, copy);
-
- //get leaf nodes that may need externsion
+
+ //get leaf nodes that may need extension
for (int i = 0; i < copy.size(); i++) {
if (copy[i].children.size() == 0) {
leafNodes[i] = i;
}
}
-
+ //cout << "maxLevel = " << maxLevel << endl;
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
int currentNode = itLeaf->second;
//this sequence is unclassified at some levels
- while(level <= maxLevel){
+ while(level < maxLevel){
level++;
try {
map<string,int>::iterator it;
- 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;
- }
+ 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;
+ }
- if (tree[index].level <= maxLevel) {
- for(it=tree[index].children.begin();it!=tree[index].children.end();it++){ //check your children
+
+ for(it=copy[index].children.begin();it!=copy[index].children.end();it++){ //check your children
fillOutTree(it->second, copy);
}
+
}
}
for (it2=uniqueTaxonomies.begin(); it2!=uniqueTaxonomies.end(); it2++) { outTree << it2->first << '\t' << tree[it2->first].accessions.size() << endl; }
outTree << endl;
-
outTree.close();
#ifdef USE_MPI
}
}
/**************************************************************************************************/
+TaxNode PhyloTree::get(int i ){
+ try {
+ if (i < tree.size()) { return tree[i]; }
+ else { cout << i << '\t' << tree.size() << endl ; m->mothurOut("Mismatch with taxonomy and template files. Cannot continue."); m->mothurOutEndLine(); exit(1); }
+ }
+ catch(exception& e) {
+ m->errorOut(e, "PhyloTree", "get");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
+TaxNode PhyloTree::get(string seqName){
+ try {
+ map<string, int>::iterator itFind = name2Taxonomy.find(seqName);
+
+ if (itFind != name2Taxonomy.end()) { return tree[name2Taxonomy[seqName]]; }
+ else { m->mothurOut("Cannot find " + seqName + ". Mismatch with taxonomy and template files. Cannot continue."); m->mothurOutEndLine(); exit(1);}
+ }
+ catch(exception& e) {
+ m->errorOut(e, "PhyloTree", "get");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
+string PhyloTree::getName(int i ){
+ try {
+ if (i < tree.size()) { return tree[i].name; }
+ else { m->mothurOut("Mismatch with taxonomy and template files. Cannot continue."); m->mothurOutEndLine(); exit(1); }
+ }
+ catch(exception& e) {
+ m->errorOut(e, "PhyloTree", "get");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
+int PhyloTree::getIndex(string seqName){
+ try {
+ map<string, int>::iterator itFind = name2Taxonomy.find(seqName);
+
+ if (itFind != name2Taxonomy.end()) { return name2Taxonomy[seqName]; }
+ else { m->mothurOut("Cannot find " + seqName + ". Mismatch with taxonomy and template files. Cannot continue."); m->mothurOutEndLine(); exit(1);}
+ }
+ catch(exception& e) {
+ m->errorOut(e, "PhyloTree", "get");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
+bool PhyloTree::ErrorCheck(vector<string> templateFileNames){
+ try {
+
+ bool okay = true;
+
+ map<string, int>::iterator itFind;
+ map<string, int> taxonomyFileNames = name2Taxonomy;
+
+ for (int i = 0; i < templateFileNames.size(); i++) {
+ itFind = taxonomyFileNames.find(templateFileNames[i]);
+
+ if (itFind != taxonomyFileNames.end()) { //found it so erase it
+ taxonomyFileNames.erase(itFind);
+ }else {
+ m->mothurOut(templateFileNames[i] + " is in your template file and is not in your taxonomy file. Please correct."); m->mothurOutEndLine();
+ okay = false;
+ }
+
+ templateFileNames.erase(templateFileNames.begin()+i);
+ i--;
+ }
+
+ if (taxonomyFileNames.size() > 0) { //there are names in tax file that are not in template
+ okay = false;
+
+ for (itFind = taxonomyFileNames.begin(); itFind != taxonomyFileNames.end(); itFind++) {
+ m->mothurOut(itFind->first + " is in your taxonomy file and is not in your template file. Please correct."); m->mothurOutEndLine();
+ }
+ }
+
+ return okay;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "PhyloTree", "ErrorCheck");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
+