//**********************************************************************************************************************
vector<string> TreeGroupCommand::setParameters(){
try {
- CommandParameter pshared("shared", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "none",false,false); parameters.push_back(pshared);
- CommandParameter pphylip("phylip", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "none",false,false); parameters.push_back(pphylip);
- CommandParameter pname("name", "InputTypes", "", "", "none", "none", "ColumnName",false,false); parameters.push_back(pname);
- CommandParameter pcolumn("column", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "ColumnName",false,false); parameters.push_back(pcolumn);
- CommandParameter piters("iters", "Number", "", "1000", "", "", "",false,false); parameters.push_back(piters);
- CommandParameter psubsample("subsample", "String", "", "", "", "", "",false,false); parameters.push_back(psubsample);
- CommandParameter pcutoff("cutoff", "Number", "", "10", "", "", "",false,false); parameters.push_back(pcutoff);
- CommandParameter pprecision("precision", "Number", "", "100", "", "", "",false,false); parameters.push_back(pprecision);
- CommandParameter plabel("label", "String", "", "", "", "", "",false,false); parameters.push_back(plabel);
- CommandParameter pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups);
- CommandParameter pcalc("calc", "Multiple", "sharedsobs-sharedchao-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan-kstest-sharednseqs-ochiai-anderberg-kulczynski-kulczynskicody-lennon-morisitahorn-braycurtis-whittaker-odum-canberra-structeuclidean-structchord-hellinger-manhattan-structpearson-soergel-spearman-structkulczynski-speciesprofile-hamming-structchi2-gower-memchi2-memchord-memeuclidean-mempearson", "jclass-thetayc", "", "", "",true,false); parameters.push_back(pcalc);
+ CommandParameter pshared("shared", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "none","tree",false,false,true); parameters.push_back(pshared);
+ CommandParameter pphylip("phylip", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "none","tree",false,false); parameters.push_back(pphylip);
+ CommandParameter pname("name", "InputTypes", "", "", "NameCount", "none", "ColumnName","",false,false); parameters.push_back(pname);
+ CommandParameter pcount("count", "InputTypes", "", "", "NameCount", "none", "countcolumn","",false,false); parameters.push_back(pcount);
+ CommandParameter pcolumn("column", "InputTypes", "", "", "PhylipColumnShared", "PhylipColumnShared", "ColumnName-countcolumn","tree",false,false); parameters.push_back(pcolumn);
+ CommandParameter piters("iters", "Number", "", "1000", "", "", "","",false,false); parameters.push_back(piters);
+ CommandParameter psubsample("subsample", "String", "", "", "", "", "","",false,false); parameters.push_back(psubsample);
+ CommandParameter pcutoff("cutoff", "Number", "", "10", "", "", "","",false,false); parameters.push_back(pcutoff);
+ CommandParameter pprecision("precision", "Number", "", "100", "", "", "","",false,false); parameters.push_back(pprecision);
+ CommandParameter plabel("label", "String", "", "", "", "", "","",false,false); parameters.push_back(plabel);
+ CommandParameter pgroups("groups", "String", "", "", "", "", "","",false,false); parameters.push_back(pgroups);
+ CommandParameter pcalc("calc", "Multiple", "sharedsobs-sharedchao-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan-kstest-sharednseqs-ochiai-anderberg-kulczynski-kulczynskicody-lennon-morisitahorn-braycurtis-whittaker-odum-canberra-structeuclidean-structchord-hellinger-manhattan-structpearson-soergel-spearman-structkulczynski-speciesprofile-hamming-structchi2-gower-memchi2-memchord-memeuclidean-mempearson", "jclass-thetayc", "", "", "","",true,false,true); parameters.push_back(pcalc);
- CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors);
- CommandParameter poutput("output", "Multiple", "lt-square", "lt", "", "", "",false,false); parameters.push_back(poutput);
- CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
- CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
+ CommandParameter pprocessors("processors", "Number", "", "1", "", "", "","",false,false,true); parameters.push_back(pprocessors);
+//CommandParameter poutput("output", "Multiple", "lt-square", "lt", "", "", "",false,false); parameters.push_back(poutput);
+ CommandParameter pinputdir("inputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(pinputdir);
+ CommandParameter poutputdir("outputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(poutputdir);
vector<string> myArray;
for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
}
}
//**********************************************************************************************************************
+string TreeGroupCommand::getOutputPattern(string type) {
+ try {
+ string pattern = "";
+
+ if (type == "tree") { pattern = "[filename],[calc],[distance],[tag],tre-[filename],tre"; }
+ else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; }
+
+ return pattern;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "TreeGroupCommand", "getOutputPattern");
+ exit(1);
+ }
+}
+//**********************************************************************************************************************
TreeGroupCommand::TreeGroupCommand(){
try {
abort = true; calledHelp = true;
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["name"] = inputDir + it->second; }
}
+
+ it = parameters.find("count");
+ //user has given a template file
+ if(it != parameters.end()){
+ path = m->hasPath(it->second);
+ //if the user has not given a path then, add inputdir. else leave path alone.
+ if (path == "") { parameters["count"] = inputDir + it->second; }
+ }
}
//check for required parameters
if (namefile == "not open") { abort = true; }
else if (namefile == "not found") { namefile = ""; }
else { m->setNameFile(namefile); }
+
+ countfile = validParameter.validFile(parameters, "count", true);
+ if (countfile == "not open") { abort = true; countfile = ""; }
+ else if (countfile == "not found") { countfile = ""; }
+ else { m->setCountTableFile(countfile); }
if ((phylipfile == "") && (columnfile == "") && (sharedfile == "")) {
//is there are current file available for either of these?
else if ((phylipfile != "") && (columnfile != "")) { m->mothurOut("When running the tree.shared command with a distance file you may not use both the column and the phylip parameters."); m->mothurOutEndLine(); abort = true; }
if (columnfile != "") {
- if (namefile == "") {
+ if ((namefile == "") && (countfile == "")){
namefile = m->getNameFile();
if (namefile != "") { m->mothurOut("Using " + namefile + " as input file for the name parameter."); m->mothurOutEndLine(); }
else {
- m->mothurOut("You need to provide a namefile if you are going to use the column format."); m->mothurOutEndLine();
- abort = true;
+ countfile = m->getCountTableFile();
+ if (countfile != "") { m->mothurOut("Using " + countfile + " as input file for the count parameter."); m->mothurOutEndLine(); }
+ else {
+ m->mothurOut("You need to provide a namefile or countfile if you are going to use the column format."); m->mothurOutEndLine();
+ abort = true;
+ }
}
}
}
+
//check for optional parameter and set defaults
// ...at some point should added some additional type checking...
TreeGroupCommand::~TreeGroupCommand(){
if (abort == false) {
if (format == "sharedfile") { delete input; }
- else { delete readMatrix; delete matrix; delete list; }
- delete tmap;
+ else { delete list; }
+ delete ct;
}
}
m->runParse = false;
//create treemap class from groupmap for tree class to use
- tmap = new TreeMap();
- tmap->makeSim(m->getAllGroups());
+ ct = new CountTable();
+ set<string> nameMap;
+ map<string, string> groupMap;
+ set<string> gps;
+ for (int i = 0; i < m->getAllGroups().size(); i++) {
+ nameMap.insert(m->getAllGroups()[i]);
+ gps.insert(m->getAllGroups()[i]);
+ groupMap[m->getAllGroups()[i]] = m->getAllGroups()[i];
+ }
+ ct->createTable(nameMap, groupMap, gps);
//clear globaldatas old tree names if any
m->Treenames.clear();
//fills globaldatas tree names
- m->Treenames = m->getGroups();
+ //m->Treenames = m->getGroups();
+ for (int k = 0; k < lookup.size(); k++) {
+ m->Treenames.push_back(lookup[k]->getGroup());
+ }
if (m->control_pressed) { return 0; }
}else{
//read in dist file
filename = inputfile;
-
+
+ ReadMatrix* readMatrix;
if (format == "column") { readMatrix = new ReadColumnMatrix(filename); }
else if (format == "phylip") { readMatrix = new ReadPhylipMatrix(filename); }
readMatrix->setCutoff(cutoff);
- if(namefile != ""){
- nameMap = new NameAssignment(namefile);
- nameMap->readMap();
- }
- else{
- nameMap = NULL;
- }
-
- readMatrix->read(nameMap);
- list = readMatrix->getListVector();
- matrix = readMatrix->getMatrix();
+ ct = NULL;
+ nameMap = NULL;
+ if(namefile != ""){
+ nameMap = new NameAssignment(namefile);
+ nameMap->readMap();
+ readMatrix->read(nameMap);
+ }else if (countfile != "") {
+ ct = new CountTable();
+ ct->readTable(countfile, true, false);
+ readMatrix->read(ct);
+ }else {
+ readMatrix->read(nameMap);
+ }
+ list = readMatrix->getListVector();
+ SparseDistanceMatrix* dMatrix = readMatrix->getDMatrix();
+
+ //clear globaldatas old tree names if any
+ m->Treenames.clear();
+
//make treemap
- tmap = new TreeMap();
-
- if (m->control_pressed) { return 0; }
-
- tmap->makeSim(list);
+ if (ct != NULL) { delete ct; }
+ ct = new CountTable();
+ set<string> nameMap;
+ map<string, string> groupMap;
+ set<string> gps;
+ for (int i = 0; i < list->getNumBins(); i++) {
+ string bin = list->get(i);
+ nameMap.insert(bin);
+ gps.insert(bin);
+ groupMap[bin] = bin;
+ m->Treenames.push_back(bin);
+ }
+ ct->createTable(nameMap, groupMap, gps);
- vector<string> namesGroups = tmap->getNamesOfGroups();
+ vector<string> namesGroups = ct->getNamesOfGroups();
m->setGroups(namesGroups);
-
- //clear globaldatas old tree names if any
- m->Treenames.clear();
-
- //fills globaldatas tree names
- m->Treenames = m->getGroups();
//used in tree constructor
m->runParse = false;
if (m->control_pressed) { return 0; }
- vector< vector<double> > matrix = makeSimsDist();
+ vector< vector<double> > matrix = makeSimsDist(dMatrix);
+ delete readMatrix;
+ delete dMatrix;
if (m->control_pressed) { return 0; }
//create a new filename
- string outputFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + "tre";
+ map<string, string> variables;
+ variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(inputfile));
+ string outputFile = getOutputFileName("tree",variables);
outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
Tree* newTree = createTree(matrix);
Tree* TreeGroupCommand::createTree(vector< vector<double> >& simMatrix){
try {
//create tree
- t = new Tree(tmap, simMatrix);
+ t = new Tree(ct, simMatrix);
if (m->control_pressed) { delete t; t = NULL; return t; }
//assemble tree
- map<string, string> empty;
- t->assembleTree(empty);
+ t->assembleTree();
return t;
}
}
}
/***********************************************************/
-vector< vector<double> > TreeGroupCommand::makeSimsDist() {
+vector< vector<double> > TreeGroupCommand::makeSimsDist(SparseDistanceMatrix* matrix) {
try {
numGroups = list->size();
//go through sparse matrix and fill sims
//go through each cell in the sparsematrix
- for(MatData currentCell = matrix->begin(); currentCell != matrix->end(); currentCell++){
- //similairity = -(distance-1)
- simMatrix[currentCell->row][currentCell->column] = -(currentCell->dist -1.0);
- simMatrix[currentCell->column][currentCell->row] = -(currentCell->dist -1.0);
-
- if (m->control_pressed) { return simMatrix; }
+ for (int i = 0; i < matrix->seqVec.size(); i++) {
+ for (int j = 0; j < matrix->seqVec[i].size(); j++) {
+
+ //already checked everyone else in row
+ if (i < matrix->seqVec[i][j].index) {
+ simMatrix[i][matrix->seqVec[i][j].index] = -(matrix->seqVec[i][j].dist -1.0);
+ simMatrix[matrix->seqVec[i][j].index][i] = -(matrix->seqVec[i][j].dist -1.0);
+ if (m->control_pressed) { return simMatrix; }
+ }
+ }
}
return simMatrix;
int TreeGroupCommand::makeSimsShared() {
try {
- numGroups = lookup.size();
- lines.resize(processors);
- for (int i = 0; i < processors; i++) {
- lines[i].start = int (sqrt(float(i)/float(processors)) * numGroups);
- lines[i].end = int (sqrt(float(i+1)/float(processors)) * numGroups);
- }
-
if (subsample) {
if (subsampleSize == -1) { //user has not set size, set size = smallest samples size
subsampleSize = lookup[0]->getNumSeqs();
}else {
m->clearGroups();
Groups.clear();
+ m->Treenames.clear();
vector<SharedRAbundVector*> temp;
for (int i = 0; i < lookup.size(); i++) {
if (lookup[i]->getNumSeqs() < subsampleSize) {
}else {
Groups.push_back(lookup[i]->getGroup());
temp.push_back(lookup[i]);
+ m->Treenames.push_back(lookup[i]->getGroup());
}
}
lookup = temp;
m->setGroups(Groups);
}
+
+ if (lookup.size() < 2) { m->mothurOut("You have not provided enough valid groups. I cannot run the command."); m->mothurOutEndLine(); m->control_pressed = true; return 0; }
+ }
+ numGroups = lookup.size();
+
+ //sanity check to make sure processors < numComparisions
+ int numDists = 0;
+ for(int i=0;i<numGroups;i++){
+ for(int j=0;j<i;j++){
+ numDists++;
+ if (numDists > processors) { break; }
+ }
}
+ if (numDists < processors) { processors = numDists; }
+
+ lines.resize(processors);
+ for (int i = 0; i < processors; i++) {
+ lines[i].start = int (sqrt(float(i)/float(processors)) * numGroups);
+ lines[i].end = int (sqrt(float(i+1)/float(processors)) * numGroups);
+ }
set<string> processedLabels;
set<string> userLabels = labels;
//Close all thread handles and free memory allocations.
for(int i=0; i < pDataArray.size(); i++){
+ if (pDataArray[i]->count != (pDataArray[i]->end-pDataArray[i]->start)) {
+ m->mothurOut("[ERROR]: process " + toString(i) + " only processed " + toString(pDataArray[i]->count) + " of " + toString(pDataArray[i]->end-pDataArray[i]->start) + " groups assigned to it, quitting. \n"); m->control_pressed = true;
+ }
for (int j = 0; j < pDataArray[i]->thisLookup.size(); j++) { delete pDataArray[i]->thisLookup[j]; }
for (int k = 0; k < calcDists.size(); k++) {
thisItersLookup.clear();
for (int i = 0; i < calcDists.size(); i++) { calcDists[i].clear(); }
}
+
+ if (m->debug) { m->mothurOut("[DEBUG]: iter = " + toString(thisIter) + ".\n"); }
}
-
+
+ if (m->debug) { m->mothurOut("[DEBUG]: done with iters.\n"); }
+
if (iters != 1) {
//we need to find the average distance and standard deviation for each groups distance
+ vector< vector<seqDist> > calcAverages = m->getAverages(calcDistsTotals);
- vector< vector<seqDist> > calcAverages; calcAverages.resize(treeCalculators.size());
- for (int i = 0; i < calcAverages.size(); i++) { //initialize sums to zero.
- calcAverages[i].resize(calcDistsTotals[0][i].size());
-
- for (int j = 0; j < calcAverages[i].size(); j++) {
- calcAverages[i][j].seq1 = calcDists[i][j].seq1;
- calcAverages[i][j].seq2 = calcDists[i][j].seq2;
- calcAverages[i][j].dist = 0.0;
- }
- }
-
- for (int thisIter = 0; thisIter < iters; thisIter++) { //sum all groups dists for each calculator
- for (int i = 0; i < calcAverages.size(); i++) { //initialize sums to zero.
- for (int j = 0; j < calcAverages[i].size(); j++) {
- calcAverages[i][j].dist += calcDistsTotals[thisIter][i][j].dist;
- }
- }
- }
-
- for (int i = 0; i < calcAverages.size(); i++) { //finds average.
- for (int j = 0; j < calcAverages[i].size(); j++) {
- calcAverages[i][j].dist /= (float) iters;
- }
- }
+ if (m->debug) { m->mothurOut("[DEBUG]: found averages.\n"); }
//create average tree for each calc
for (int i = 0; i < calcDists.size(); i++) {
}
//create a new filename
- string outputFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".ave.tre";
+ map<string, string> variables;
+ variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(inputfile));
+ variables["[calc]"] = treeCalculators[i]->getName();
+ variables["[distance]"] = thisLookup[0]->getLabel();
+ variables["[tag]"] = "ave";
+ string outputFile = getOutputFileName("tree",variables);
outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
//creates tree from similarity matrix and write out file
if (newTree != NULL) { writeTree(outputFile, newTree); }
}
+ if (m->debug) { m->mothurOut("[DEBUG]: done averages trees.\n"); }
+
//create all trees for each calc and find their consensus tree
for (int i = 0; i < calcDists.size(); i++) {
if (m->control_pressed) { break; }
//create a new filename
- string outputFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".all.tre";
+ //create a new filename
+ map<string, string> variables;
+ variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(inputfile));
+ variables["[calc]"] = treeCalculators[i]->getName();
+ variables["[distance]"] = thisLookup[0]->getLabel();
+ variables["[tag]"] = "all";
+ string outputFile = getOutputFileName("tree",variables);
outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
ofstream outAll;
int row = calcDistsTotals[myIter][i][j].seq1;
int column = calcDistsTotals[myIter][i][j].seq2;
double dist = calcDistsTotals[myIter][i][j].dist;
-
+
matrix[row][column] = dist;
matrix[column][row] = dist;
}
outAll.close();
if (m->control_pressed) { for (int k = 0; k < trees.size(); k++) { delete trees[k]; } }
+ if (m->debug) { m->mothurOut("[DEBUG]: done all trees.\n"); }
+
Consensus consensus;
//clear old tree names if any
m->Treenames.clear(); m->Treenames = m->getGroups(); //may have changed if subsample eliminated groups
Tree* conTree = consensus.getTree(trees);
+ if (m->debug) { m->mothurOut("[DEBUG]: done cons tree.\n"); }
+
//create a new filename
- string conFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".cons.tre";
+ variables["[tag]"] = "cons";
+ string conFile = getOutputFileName("tree",variables);
outputNames.push_back(conFile); outputTypes["tree"].push_back(conFile);
ofstream outTree;
m->openOutputFile(conFile, outTree);
}
//create a new filename
- string outputFile = outputDir + m->getRootName(m->getSimpleName(inputfile)) + treeCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + ".tre";
+ map<string, string> variables;
+ variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(inputfile));
+ variables["[calc]"] = treeCalculators[i]->getName();
+ variables["[distance]"] = thisLookup[0]->getLabel();
+ variables["[tag]"] = "";
+ string outputFile = getOutputFileName("tree",variables);
outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
//creates tree from similarity matrix and write out file
projects / mothur.git / blobdiff