X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=pcoacommand.cpp;h=2043a128bd94bc755cb7ef10f5a8b8d9bdb5a6be;hb=1a968f34ae2d2680eaf189a197d1a21b8dfd6c03;hp=9375144bcb342faf364a38eee6eca63226099659;hpb=605ab6fa594317a38f0df7bb6797740c735b2348;p=mothur.git diff --git a/pcoacommand.cpp b/pcoacommand.cpp index 9375144..2043a12 100644 --- a/pcoacommand.cpp +++ b/pcoacommand.cpp @@ -9,54 +9,71 @@ */ #include "pcoacommand.h" +#include "readphylipvector.h" + //********************************************************************************************************************** -vector PCOACommand::getValidParameters(){ +vector PCOACommand::setParameters(){ try { - string Array[] = {"phylip", "metric","outputdir","inputdir"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + CommandParameter pphylip("phylip", "InputTypes", "", "", "none", "none", "none","pcoa-loadings",false,true,true); parameters.push_back(pphylip); + CommandParameter pmetric("metric", "Boolean", "", "T", "", "", "","",false,false); parameters.push_back(pmetric); + CommandParameter pinputdir("inputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(pinputdir); + CommandParameter poutputdir("outputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(poutputdir); + + vector myArray; + for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); } return myArray; } catch(exception& e) { - m->errorOut(e, "PCOACommand", "getValidParameters"); + m->errorOut(e, "PCOACommand", "setParameters"); exit(1); } } //********************************************************************************************************************** -PCOACommand::PCOACommand(){ +string PCOACommand::getHelpString(){ try { - abort = true; - //initialize outputTypes - vector tempOutNames; - outputTypes["pcoa"] = tempOutNames; - outputTypes["loadings"] = tempOutNames; - outputTypes["corr"] = tempOutNames; + string helpString = ""; + helpString += "The pcoa command parameters are phylip and metric"; + helpString += "The phylip parameter allows you to enter your distance file."; + helpString += "The metric parameter allows indicate you if would like the pearson correlation coefficient calculated. Default=True"; + helpString += "Example pcoa(phylip=yourDistanceFile).\n"; + helpString += "Note: No spaces between parameter labels (i.e. phylip), '=' and parameters (i.e.yourDistanceFile).\n"; + return helpString; } catch(exception& e) { - m->errorOut(e, "PCOACommand", "PCOACommand"); + m->errorOut(e, "PCOACommand", "getHelpString"); exit(1); } } //********************************************************************************************************************** -vector PCOACommand::getRequiredParameters(){ - try { - string Array[] = {"phylip"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); - return myArray; - } - catch(exception& e) { - m->errorOut(e, "PCOACommand", "getRequiredParameters"); - exit(1); - } +string PCOACommand::getOutputPattern(string type) { + try { + string pattern = ""; + + if (type == "pcoa") { pattern = "[filename],pcoa.axes"; } + else if (type == "loadings") { pattern = "[filename],pcoa.loadings"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "PCOACommand", "getOutputPattern"); + exit(1); + } } + + //********************************************************************************************************************** -vector PCOACommand::getRequiredFiles(){ +PCOACommand::PCOACommand(){ try { - vector myArray; - return myArray; + abort = true; calledHelp = true; + setParameters(); + vector tempOutNames; + outputTypes["pcoa"] = tempOutNames; + outputTypes["loadings"] = tempOutNames; } catch(exception& e) { - m->errorOut(e, "PCOACommand", "getRequiredFiles"); + m->errorOut(e, "PCOACommand", "PCOACommand"); exit(1); } } @@ -64,15 +81,14 @@ vector PCOACommand::getRequiredFiles(){ PCOACommand::PCOACommand(string option) { try { - abort = false; + abort = false; calledHelp = false; //allow user to run help - if(option == "help") { help(); abort = true; } + if(option == "help") { help(); abort = true; calledHelp = true; } + else if(option == "citation") { citation(); abort = true; calledHelp = true;} else { - //valid paramters for this command - string Array[] = {"phylip","metric","outputdir", "inputdir"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + vector myArray = setParameters(); OptionParser parser(option); map parameters = parser. getParameters(); @@ -102,13 +118,18 @@ PCOACommand::PCOACommand(string option) { vector tempOutNames; outputTypes["pcoa"] = tempOutNames; outputTypes["loadings"] = tempOutNames; - outputTypes["corr"] = tempOutNames; //required parameters phylipfile = validParameter.validFile(parameters, "phylip", true); if (phylipfile == "not open") { abort = true; } - else if (phylipfile == "not found") { phylipfile = ""; abort = true; } - else { filename = phylipfile; } + else if (phylipfile == "not found") { + //if there is a current phylip file, use it + phylipfile = m->getPhylipFile(); + if (phylipfile != "") { m->mothurOut("Using " + phylipfile + " as input file for the phylip parameter."); m->mothurOutEndLine(); } + else { m->mothurOut("You have no current phylip file and the phylip parameter is required."); m->mothurOutEndLine(); abort = true; } + }else { m->setPhylipFile(phylipfile); } + + filename = phylipfile; //if the user changes the output directory command factory will send this info to us in the output parameter outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ @@ -116,9 +137,6 @@ PCOACommand::PCOACommand(string option) { outputDir += m->hasPath(phylipfile); //if user entered a file with a path then preserve it } - //error checking on files - if (phylipfile == "") { m->mothurOut("You must provide a distance file before running the pcoa command."); m->mothurOutEndLine(); abort = true; } - string temp = validParameter.validFile(parameters, "metric", false); if (temp == "not found"){ temp = "T"; } metric = m->isTrue(temp); } @@ -130,27 +148,10 @@ PCOACommand::PCOACommand(string option) { } } //********************************************************************************************************************** -void PCOACommand::help(){ - try { - - m->mothurOut("The pcoa command parameters are phylip and metric"); m->mothurOutEndLine(); - m->mothurOut("The phylip parameter allows you to enter your distance file."); m->mothurOutEndLine(); - m->mothurOut("The metric parameter allows indicate you if would like the pearson correlation coefficient calculated. Default=True"); m->mothurOutEndLine(); - m->mothurOut("Example pcoa(phylip=yourDistanceFile).\n"); - m->mothurOut("Note: No spaces between parameter labels (i.e. phylip), '=' and parameters (i.e.yourDistanceFile).\n\n"); - } - catch(exception& e) { - m->errorOut(e, "PCOACommand", "help"); - exit(1); - } -} -//********************************************************************************************************************** -PCOACommand::~PCOACommand(){} -//********************************************************************************************************************** int PCOACommand::execute(){ try { - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } cout.setf(ios::fixed, ios::floatfield); cout.setf(ios::showpoint); @@ -162,7 +163,8 @@ int PCOACommand::execute(){ fbase = outputDir + m->getRootName(m->getSimpleName(filename)); - read(filename, names, D); + ReadPhylipVector readFile(filename); + names = readFile.read(D); if (m->control_pressed) { return 0; } @@ -170,12 +172,12 @@ int PCOACommand::execute(){ vector d; vector e; vector > G = D; - vector > copy_G; + //vector > copy_G; - m->mothurOut("\nProcessing...\n\n"); + m->mothurOut("\nProcessing...\n"); for(int count=0;count<2;count++){ - recenter(offset, D, G); if (m->control_pressed) { return 0; } + linearCalc.recenter(offset, D, G); if (m->control_pressed) { return 0; } linearCalc.tred2(G, d, e); if (m->control_pressed) { return 0; } linearCalc.qtli(d, e, G); if (m->control_pressed) { return 0; } offset = d[d.size()-1]; @@ -186,21 +188,21 @@ int PCOACommand::execute(){ output(fbase, names, G, d); - if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } if (metric) { for (int i = 1; i < 4; i++) { - vector< vector > EuclidDists = calculateEuclidianDistance(G, i); //G is the pcoa file + vector< vector > EuclidDists = linearCalc.calculateEuclidianDistance(G, i); //G is the pcoa file - if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } - double corr = calcPearson(EuclidDists, D); //G is the pcoa file, D is the users distance matrix + double corr = linearCalc.calcPearson(EuclidDists, D); //G is the pcoa file, D is the users distance matrix - m->mothurOut("Pearson's coefficient using " + toString(i) + " axis: " + toString(corr)); m->mothurOutEndLine(); + m->mothurOut("Rsq " + toString(i) + " axis: " + toString(corr * corr)); m->mothurOutEndLine(); - if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } } } @@ -217,114 +219,6 @@ int PCOACommand::execute(){ } } /*********************************************************************************************************************************/ -vector< vector > PCOACommand::calculateEuclidianDistance(vector< vector >& axes, int dimensions){ - try { - //make square matrix - vector< vector > dists; dists.resize(axes.size()); - for (int i = 0; i < dists.size(); i++) { dists[i].resize(axes.size(), 0.0); } - - if (dimensions == 1) { //one dimension calc = abs(x-y) - - for (int i = 0; i < dists.size(); i++) { - - if (m->control_pressed) { return dists; } - - for (int j = 0; j < i; j++) { - dists[i][j] = abs(axes[i][0] - axes[j][0]); - dists[j][i] = dists[i][j]; - } - } - - }else if (dimensions == 2) { //two dimension calc = sqrt ((x1 - y1)^2 + (x2 - y2)^2) - - for (int i = 0; i < dists.size(); i++) { - - if (m->control_pressed) { return dists; } - - for (int j = 0; j < i; j++) { - double firstDim = ((axes[i][0] - axes[j][0]) * (axes[i][0] - axes[j][0])); - double secondDim = ((axes[i][1] - axes[j][1]) * (axes[i][1] - axes[j][1])); - - dists[i][j] = sqrt((firstDim + secondDim)); - dists[j][i] = dists[i][j]; - } - } - - }else if (dimensions == 3) { //two dimension calc = sqrt ((x1 - y1)^2 + (x2 - y2)^2 + (x3 - y3)^2) - - for (int i = 0; i < dists.size(); i++) { - - if (m->control_pressed) { return dists; } - - for (int j = 0; j < i; j++) { - double firstDim = ((axes[i][0] - axes[j][0]) * (axes[i][0] - axes[j][0])); - double secondDim = ((axes[i][1] - axes[j][1]) * (axes[i][1] - axes[j][1])); - double thirdDim = ((axes[i][2] - axes[j][2]) * (axes[i][2] - axes[j][2])); - - dists[i][j] = sqrt((firstDim + secondDim + thirdDim)); - dists[j][i] = dists[i][j]; - } - } - - }else { m->mothurOut("[ERROR]: too many dimensions, aborting."); m->mothurOutEndLine(); m->control_pressed = true; } - - return dists; - } - catch(exception& e) { - m->errorOut(e, "PCOACommand", "calculateEuclidianDistance"); - exit(1); - } -} -/*********************************************************************************************************************************/ -double PCOACommand::calcPearson(vector< vector >& euclidDists, vector< vector >& userDists){ - try { - - //find average for - X - vector averageEuclid; averageEuclid.resize(euclidDists.size(), 0.0); - for (int i = 0; i < euclidDists.size(); i++) { - for (int j = 0; j < euclidDists[i].size(); j++) { - averageEuclid[i] += euclidDists[i][j]; - } - } - for (int i = 0; i < averageEuclid.size(); i++) { averageEuclid[i] = averageEuclid[i] / (float) euclidDists.size(); } - - //find average for - Y - vector averageUser; averageUser.resize(userDists.size(), 0.0); - for (int i = 0; i < userDists.size(); i++) { - for (int j = 0; j < userDists[i].size(); j++) { - averageUser[i] += userDists[i][j]; - } - } - for (int i = 0; i < averageUser.size(); i++) { averageUser[i] = averageUser[i] / (float) userDists.size(); } - - double numerator = 0.0; - double denomTerm1 = 0.0; - double denomTerm2 = 0.0; - - for (int i = 0; i < euclidDists.size(); i++) { - - for (int k = 0; k < i; k++) { - - float Yi = userDists[i][k]; - float Xi = euclidDists[i][k]; - - numerator += ((Xi - averageEuclid[k]) * (Yi - averageUser[k])); - denomTerm1 += ((Xi - averageEuclid[k]) * (Xi - averageEuclid[k])); - denomTerm2 += ((Yi - averageUser[k]) * (Yi - averageUser[k])); - } - } - - double denom = (sqrt(denomTerm1) * sqrt(denomTerm2)); - double r = numerator / denom; - - return r; - } - catch(exception& e) { - m->errorOut(e, "PCOACommand", "calculateEuclidianDistance"); - exit(1); - } -} -/*********************************************************************************************************************************/ void PCOACommand::get_comment(istream& f, char begin, char end){ try { @@ -337,137 +231,6 @@ void PCOACommand::get_comment(istream& f, char begin, char end){ exit(1); } } - -/*********************************************************************************************************************************/ - -int PCOACommand::read_phylip(istream& f, int square_m, vector& name_list, vector >& d){ - try { - // int count1=0; - // int count2=0; - - int rank; - f >> rank; - - name_list.resize(rank); - d.resize(rank); - if(square_m == 1){ - for(int i=0;i> name_list[i]; - // cout << i << "\t" << name_list[i] << endl; - for(int j=0;jcontrol_pressed) { return 0; } - - f >> d[i][j]; - if (d[i][j] == -0.0000) - d[i][j] = 0.0000; - } - } - } - else if(square_m == 2){ - for(int i=0;i> name_list[0]; - for(int i=1;i> name_list[i]; - d[i][i]=0.0000; - for(int j=0;jcontrol_pressed) { return 0; } - f >> d[i][j]; - if (d[i][j] == -0.0000) - d[i][j] = 0.0000; - d[j][i]=d[i][j]; - } - } - } - - return 0; - } - catch(exception& e) { - m->errorOut(e, "PCOACommand", "read_phylip"); - exit(1); - } - -} - -/*********************************************************************************************************************************/ - -void PCOACommand::read(string fname, vector& names, vector >& D){ - try { - ifstream f; - m->openInputFile(fname, f); - - //check whether matrix is square - char d; - int q = 1; - int numSeqs; - string name; - - f >> numSeqs >> name; - - while((d=f.get()) != EOF){ - - //is d a number meaning its square - if(isalnum(d)){ - q = 1; - break; - } - - //is d a line return meaning its lower triangle - if(d == '\n'){ - q = 2; - break; - } - } - f.close(); - - //reopen to get back to beginning - m->openInputFile(fname, f); - read_phylip(f, q, names, D); - } - catch(exception& e) { - m->errorOut(e, "PCOACommand", "read"); - exit(1); - } -} - -/*********************************************************************************************************************************/ - -void PCOACommand::recenter(double offset, vector > D, vector >& G){ - try { - int rank = D.size(); - - vector > A(rank); - vector > C(rank); - for(int i=0;ierrorOut(e, "PCOACommand", "recenter"); - exit(1); - } - -} - /*********************************************************************************************************************************/ void PCOACommand::output(string fnameRoot, vector name_list, vector >& G, vector d) { @@ -482,17 +245,23 @@ void PCOACommand::output(string fnameRoot, vector name_list, vector variables; + variables["[filename]"] = fnameRoot; + string pcoaDataFile = getOutputFileName("pcoa",variables); + m->openOutputFile(pcoaDataFile, pcaData); pcaData.setf(ios::fixed, ios::floatfield); pcaData.setf(ios::showpoint); - outputNames.push_back(fnameRoot+"pcoa"); - outputTypes["pcoa"].push_back(fnameRoot+"pcoa"); + outputNames.push_back(pcoaDataFile); + outputTypes["pcoa"].push_back(pcoaDataFile); - ofstream pcaLoadings((fnameRoot+"pcoa.loadings").c_str(), ios::trunc); + ofstream pcaLoadings; + string loadingsFile = getOutputFileName("loadings",variables); + m->openOutputFile(loadingsFile, pcaLoadings); pcaLoadings.setf(ios::fixed, ios::floatfield); pcaLoadings.setf(ios::showpoint); - outputNames.push_back(fnameRoot+"pcoa.loadings"); - outputTypes["loadings"].push_back(fnameRoot+"pcoa.loadings"); + outputNames.push_back(loadingsFile); + outputTypes["loadings"].push_back(loadingsFile); pcaLoadings << "axis\tloading\n"; for(int i=0;i