#include "sharedthetayc.h"
#include "sharedthetan.h"
#include "sharedmorisitahorn.h"
+#include "sharedbraycurtis.h"
//**********************************************************************************************************************
if (validCalculator->isValidCalculator("boot", globaldata->Estimators[i]) == true) {
if (globaldata->Estimators[i] == "jabund") {
treeCalculators.push_back(new JAbund());
- }else if (globaldata->Estimators[i] == "sorensonabund") {
+ }else if (globaldata->Estimators[i] == "sorabund") {
treeCalculators.push_back(new SorAbund());
}else if (globaldata->Estimators[i] == "jclass") {
treeCalculators.push_back(new Jclass());
treeCalculators.push_back(new ThetaN());
}else if (globaldata->Estimators[i] == "morisitahorn") {
treeCalculators.push_back(new MorHorn());
+ }else if (globaldata->Estimators[i] == "braycurtis") {
+ treeCalculators.push_back(new BrayCurtis());
}
}
}
int BootSharedCommand::execute(){
try {
int count = 1;
- EstOutput data;
//if the users entered no valid calculators don't execute command
if (treeCalculators.size() == 0) { return 0; }
-
- if (format == "sharedfile") {
- read = new ReadPhilFile(globaldata->inputFileName);
- read->read(&*globaldata);
-
- input = globaldata->ginput;
- order = input->getSharedOrderVector();
- }else {
- //you are using a list and a groupfile
- read = new ReadPhilFile(globaldata->inputFileName);
- read->read(&*globaldata);
- input = globaldata->ginput;
- SharedList = globaldata->gSharedList;
- order = SharedList->getSharedOrderVector();
- }
+ //read first line
+ read = new ReadOTUFile(globaldata->inputFileName);
+ read->read(&*globaldata);
+ input = globaldata->ginput;
+ order = input->getSharedOrderVector();
+ SharedOrderVector* lastOrder = order;
+ //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
+ set<string> processedLabels;
+ set<string> userLabels = globaldata->labels;
+ set<int> userLines = globaldata->lines;
+
//set users groups
util->setGroups(globaldata->Groups, globaldata->gGroupmap->namesOfGroups, "treegroup");
numGroups = globaldata->Groups.size();
tmap->makeSim(globaldata->gGroupmap);
globaldata->gTreemap = tmap;
- while(order != NULL){
+ while((order != NULL) && ((globaldata->allLines == 1) || (userLabels.size() != 0) || (userLines.size() != 0))) {
if(globaldata->allLines == 1 || globaldata->lines.count(count) == 1 || globaldata->labels.count(order->getLabel()) == 1){
cout << order->getLabel() << '\t' << count << endl;
+ process(order);
- //open an ostream for each calc to print to
- for (int z = 0; z < treeCalculators.size(); z++) {
- //create a new filename
- outputFile = getRootName(globaldata->inputFileName) + treeCalculators[z]->getName() + ".boot" + order->getLabel() + ".tre";
- openOutputFile(outputFile, *(out[z]));
- }
-
- //create a file for each calculator with the 1000 trees in it.
- for (int p = 0; p < iters; p++) {
-
- util->getSharedVectorswithReplacement(globaldata->Groups, lookup, order); //fills group vectors from order vector.
-
- //for each calculator
- for(int i = 0 ; i < treeCalculators.size(); i++) {
-
- //initialize simMatrix
- simMatrix.clear();
- simMatrix.resize(numGroups);
- for (int m = 0; m < simMatrix.size(); m++) {
- for (int j = 0; j < simMatrix.size(); j++) {
- simMatrix[m].push_back(0.0);
- }
- }
-
- //initialize index
- index.clear();
- for (int g = 0; g < numGroups; g++) { index[g] = g; }
-
- for (int k = 0; k < lookup.size(); k++) { // pass cdd each set of groups to commpare
- for (int l = k; l < lookup.size(); l++) {
- if (k != l) { //we dont need to similiarity of a groups to itself
- //get estimated similarity between 2 groups
- data = treeCalculators[i]->getValues(lookup[k], lookup[l]); //saves the calculator outputs
- //save values in similarity matrix
- simMatrix[k][l] = data[0];
- simMatrix[l][k] = data[0];
- }
- }
- }
-
- //creates tree from similarity matrix and write out file
- createTree(out[i]);
- }
- }
- //close ostream for each calc
- for (int z = 0; z < treeCalculators.size(); z++) { out[z]->close(); }
+ processedLabels.insert(order->getLabel());
+ userLabels.erase(order->getLabel());
+ userLines.erase(count);
}
-
+
+ //you have a label the user want that is smaller than this line and the last line has not already been processed
+ if ((anyLabelsToProcess(order->getLabel(), userLabels, "") == true) && (processedLabels.count(lastOrder->getLabel()) != 1)) {
+
+ cout << lastOrder->getLabel() << '\t' << count << endl;
+ process(lastOrder);
+
+ processedLabels.insert(lastOrder->getLabel());
+ userLabels.erase(lastOrder->getLabel());
+ }
+
+ if (count != 1) { delete lastOrder; }
+ lastOrder = order;
+
//get next line to process
- if (format == "sharedfile") {
- order = input->getSharedOrderVector();
+ order = input->getSharedOrderVector();
+ count++;
+ }
+
+ //output error messages about any remaining user labels
+ set<string>::iterator it;
+ bool needToRun = false;
+ for (it = userLabels.begin(); it != userLabels.end(); it++) {
+ cout << "Your file does not include the label "<< *it;
+ if (processedLabels.count(lastOrder->getLabel()) != 1) {
+ cout << ". I will use " << lastOrder->getLabel() << "." << endl;
+ needToRun = true;
}else {
- //you are using a list and a groupfile
- SharedList = input->getSharedListVector(); //get new list vector to process
- if (SharedList != NULL) {
- order = SharedList->getSharedOrderVector(); //gets new order vector with group info.
- }else {
- break;
- }
+ cout << ". Please refer to " << lastOrder->getLabel() << "." << endl;
}
- count++;
}
+ //run last line if you need to
+ if (needToRun == true) {
+ process(lastOrder);
+ cout << lastOrder->getLabel() << '\t' << count << endl;
+ }
+
+ delete lastOrder;
+
//reset groups parameter
globaldata->Groups.clear(); globaldata->setGroups("");
//do merges and create tree structure by setting parents and children
//there are numGroups - 1 merges to do
for (int i = 0; i < (numGroups - 1); i++) {
-
- float largest = 0.0;
+
+ float largest = -1.0;
int row, column;
//find largest value in sims matrix by searching lower triangle
for (int j = 1; j < simMatrix.size(); j++) {
//set non-leaf node info and update leaves to know their parents
//non-leaf
t->tree[numGroups + i].setChildren(index[row], index[column]);
-
+
//parents
t->tree[index[row]].setParent(numGroups + i);
t->tree[index[column]].setParent(numGroups + i);
//branchlengths
t->tree[index[row]].setBranchLength(blength - t->tree[index[row]].getLengthToLeaves());
t->tree[index[column]].setBranchLength(blength - t->tree[index[column]].getLengthToLeaves());
-
+
//set your length to leaves to your childs length plus branchlength
t->tree[numGroups + i].setLengthToLeaves(t->tree[index[row]].getLengthToLeaves() + t->tree[index[row]].getBranchLength());
-
+
//update index
index[row] = numGroups+i;
index[column] = numGroups+i;
//zero out highest value that caused the merge.
- simMatrix[row][column] = 0.0;
- simMatrix[column][row] = 0.0;
-
+ simMatrix[row][column] = -1.0;
+ simMatrix[column][row] = -1.0;
+
//merge values in simsMatrix
for (int n = 0; n < simMatrix.size(); n++) {
//row becomes merge of 2 groups
simMatrix[row][n] = (simMatrix[row][n] + simMatrix[column][n]) / 2;
simMatrix[n][row] = simMatrix[row][n];
//delete column
- simMatrix[column][n] = 0.0;
- simMatrix[n][column] = 0.0;
+ simMatrix[column][n] = -1.0;
+ simMatrix[n][column] = -1.0;
}
}
-
+
//assemble tree
t->assembleTree();
-
+
//print newick file
t->print(*out);
-
+
//delete tree
delete t;
}
}
/***********************************************************/
+void BootSharedCommand::process(SharedOrderVector* order) {
+ try{
+ EstOutput data;
+ vector<SharedRAbundVector*> subset;
+
+ //open an ostream for each calc to print to
+ for (int z = 0; z < treeCalculators.size(); z++) {
+ //create a new filename
+ outputFile = getRootName(globaldata->inputFileName) + treeCalculators[z]->getName() + ".boot" + order->getLabel() + ".tre";
+ openOutputFile(outputFile, *(out[z]));
+ }
+
+ //create a file for each calculator with the 1000 trees in it.
+ for (int p = 0; p < iters; p++) {
+
+ util->getSharedVectorswithReplacement(globaldata->Groups, lookup, order); //fills group vectors from order vector.
+
+ //for each calculator
+ for(int i = 0 ; i < treeCalculators.size(); i++) {
+
+ //initialize simMatrix
+ simMatrix.clear();
+ simMatrix.resize(numGroups);
+ for (int m = 0; m < simMatrix.size(); m++) {
+ for (int j = 0; j < simMatrix.size(); j++) {
+ simMatrix[m].push_back(0.0);
+ }
+ }
+
+ //initialize index
+ index.clear();
+ for (int g = 0; g < numGroups; g++) { index[g] = g; }
+
+ for (int k = 0; k < lookup.size(); k++) { // pass cdd each set of groups to commpare
+ for (int l = k; l < lookup.size(); l++) {
+ if (k != l) { //we dont need to similiarity of a groups to itself
+ subset.clear(); //clear out old pair of sharedrabunds
+ //add new pair of sharedrabunds
+ subset.push_back(lookup[k]); subset.push_back(lookup[l]);
+
+ //get estimated similarity between 2 groups
+ data = treeCalculators[i]->getValues(subset); //saves the calculator outputs
+ //save values in similarity matrix
+ simMatrix[k][l] = data[0];
+ simMatrix[l][k] = data[0];
+ }
+ }
+ }
+
+ //creates tree from similarity matrix and write out file
+ createTree(out[i]);
+ }
+ }
+ //close ostream for each calc
+ for (int z = 0; z < treeCalculators.size(); z++) { out[z]->close(); }
+
+ }
+ catch(exception& e) {
+ cout << "Standard Error: " << e.what() << " has occurred in the BootSharedCommand class Function process. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ exit(1);
+ }
+ catch(...) {
+ cout << "An unknown error has occurred in the BootSharedCommand class function process. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+ exit(1);
+ }
+}
+/***********************************************************/
+
projects / mothur.git / blobdiff