X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=unifracweightedcommand.cpp;h=b2fd628b42218a38db6b61366ba7374b4ea4603a;hb=5694c92fbf646fe01abc87bde2af59e14a9a56b6;hp=e8add90a259bac0da04bf2b8d52be247e9fc1fd1;hpb=a537fefa3bd2c2eba7d532249375b6a11b295e25;p=mothur.git diff --git a/unifracweightedcommand.cpp b/unifracweightedcommand.cpp index e8add90..b2fd628 100644 --- a/unifracweightedcommand.cpp +++ b/unifracweightedcommand.cpp @@ -9,6 +9,58 @@ #include "unifracweightedcommand.h" +//********************************************************************************************************************** +vector UnifracWeightedCommand::getValidParameters(){ + try { + string Array[] = {"groups","iters","distance","random","processors","outputdir","inputdir"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "UnifracWeightedCommand", "getValidParameters"); + exit(1); + } +} +//********************************************************************************************************************** +UnifracWeightedCommand::UnifracWeightedCommand(){ + try { + abort = true; + //initialize outputTypes + vector tempOutNames; + outputTypes["weighted"] = tempOutNames; + outputTypes["wsummary"] = tempOutNames; + outputTypes["phylip"] = tempOutNames; + outputTypes["column"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "UnifracWeightedCommand", "UnifracWeightedCommand"); + exit(1); + } +} +//********************************************************************************************************************** +vector UnifracWeightedCommand::getRequiredParameters(){ + try { + vector myArray; + return myArray; + } + catch(exception& e) { + m->errorOut(e, "UnifracWeightedCommand", "getRequiredParameters"); + exit(1); + } +} +//********************************************************************************************************************** +vector UnifracWeightedCommand::getRequiredFiles(){ + try { + string Array[] = {"tree","group"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + + return myArray; + } + catch(exception& e) { + m->errorOut(e, "UnifracWeightedCommand", "getRequiredFiles"); + exit(1); + } +} /***********************************************************/ UnifracWeightedCommand::UnifracWeightedCommand(string option) { try { @@ -34,6 +86,13 @@ UnifracWeightedCommand::UnifracWeightedCommand(string option) { if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; } } + //initialize outputTypes + vector tempOutNames; + outputTypes["weighted"] = tempOutNames; + outputTypes["wsummary"] = tempOutNames; + outputTypes["phylip"] = tempOutNames; + outputTypes["column"] = tempOutNames; + if (globaldata->gTree.size() == 0) {//no trees were read m->mothurOut("You must execute the read.tree command, before you may execute the unifrac.weighted command."); m->mothurOutEndLine(); abort = true; } @@ -55,9 +114,13 @@ UnifracWeightedCommand::UnifracWeightedCommand(string option) { itersString = validParameter.validFile(parameters, "iters", false); if (itersString == "not found") { itersString = "1000"; } convert(itersString, iters); - string temp = validParameter.validFile(parameters, "distance", false); if (temp == "not found") { temp = "false"; } - phylip = m->isTrue(temp); - + string temp = validParameter.validFile(parameters, "distance", false); + if (temp == "not found") { phylip = false; outputForm = ""; } + else{ + if ((temp == "lt") || (temp == "column") || (temp == "square")) { phylip = true; outputForm = temp; } + else { m->mothurOut("Options for distance are: lt, square, or column. Using lt."); m->mothurOutEndLine(); phylip = true; outputForm = "lt"; } + } + temp = validParameter.validFile(parameters, "random", false); if (temp == "not found") { temp = "F"; } random = m->isTrue(temp); @@ -72,7 +135,7 @@ UnifracWeightedCommand::UnifracWeightedCommand(string option) { tmap = globaldata->gTreemap; sumFile = outputDir + m->getSimpleName(globaldata->getTreeFile()) + ".wsummary"; m->openOutputFile(sumFile, outSum); - outputNames.push_back(sumFile); + outputNames.push_back(sumFile); outputTypes["wsummary"].push_back(sumFile); util = new SharedUtil(); string s; //to make work with setgroups @@ -138,6 +201,7 @@ int UnifracWeightedCommand::execute() { if (random) { output = new ColumnFile(outputDir + m->getSimpleName(globaldata->getTreeFile()) + toString(i+1) + ".weighted", itersString); outputNames.push_back(outputDir + m->getSimpleName(globaldata->getTreeFile()) + toString(i+1) + ".weighted"); + outputTypes["weighted"].push_back(outputDir + m->getSimpleName(globaldata->getTreeFile()) + toString(i+1) + ".weighted"); } userData = weighted->getValues(T[i], processors, outputDir); //userData[0] = weightedscore @@ -154,7 +218,6 @@ int UnifracWeightedCommand::execute() { } if (random) { - vector sums = weighted->getBranchLengthSums(T[i]); //calculate number of comparisons i.e. with groups A,B,C = AB, AC, BC = 3; vector< vector > namesOfGroupCombos; @@ -188,12 +251,12 @@ int UnifracWeightedCommand::execute() { #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) if(processors == 1){ - driver(T[i], namesOfGroupCombos, 0, namesOfGroupCombos.size(), sums, rScores); + driver(T[i], namesOfGroupCombos, 0, namesOfGroupCombos.size(), rScores); }else{ - createProcesses(T[i], namesOfGroupCombos, sums, rScores); + createProcesses(T[i], namesOfGroupCombos, rScores); } #else - driver(T[i], namesOfGroupCombos, 0, namesOfGroupCombos.size(), sums, rScores); + driver(T[i], namesOfGroupCombos, 0, namesOfGroupCombos.size(), rScores); #endif if (m->control_pressed) { delete output; outSum.close(); for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } @@ -265,11 +328,10 @@ int UnifracWeightedCommand::execute() { } /**************************************************************************************************/ -int UnifracWeightedCommand::createProcesses(Tree* t, vector< vector > namesOfGroupCombos, vector& sums, vector< vector >& scores) { +int UnifracWeightedCommand::createProcesses(Tree* t, vector< vector > namesOfGroupCombos, vector< vector >& scores) { try { #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) int process = 1; - int num = 0; vector processIDS; EstOutput results; @@ -282,7 +344,7 @@ int UnifracWeightedCommand::createProcesses(Tree* t, vector< vector > na processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later process++; }else if (pid == 0){ - driver(t, namesOfGroupCombos, lines[process].start, lines[process].num, sums, scores); + driver(t, namesOfGroupCombos, lines[process].start, lines[process].num, scores); //pass numSeqs to parent ofstream out; @@ -292,10 +354,14 @@ int UnifracWeightedCommand::createProcesses(Tree* t, vector< vector > na out.close(); exit(0); - }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); } + }else { + m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine(); + for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); } + exit(0); + } } - driver(t, namesOfGroupCombos, lines[0].start, lines[0].num, sums, scores); + driver(t, namesOfGroupCombos, lines[0].start, lines[0].num, scores); //force parent to wait until all the processes are done for (int i=0;i<(processors-1);i++) { @@ -326,7 +392,7 @@ int UnifracWeightedCommand::createProcesses(Tree* t, vector< vector > na } /**************************************************************************************************/ -int UnifracWeightedCommand::driver(Tree* t, vector< vector > namesOfGroupCombos, int start, int num, vector& sums, vector< vector >& scores) { +int UnifracWeightedCommand::driver(Tree* t, vector< vector > namesOfGroupCombos, int start, int num, vector< vector >& scores) { try { Tree* randT = new Tree(); @@ -347,7 +413,7 @@ int UnifracWeightedCommand::driver(Tree* t, vector< vector > namesOfGrou if (m->control_pressed) { delete randT; return 0; } //get wscore of random tree - EstOutput randomData = weighted->getValues(randT, groupA, groupB, sums); + EstOutput randomData = weighted->getValues(randT, groupA, groupB); if (m->control_pressed) { delete randT; return 0; } @@ -436,14 +502,23 @@ void UnifracWeightedCommand::createPhylipFile() { //for each tree for (int i = 0; i < T.size(); i++) { - string phylipFileName = outputDir + m->getSimpleName(globaldata->getTreeFile()) + toString(i+1) + ".weighted.dist"; - outputNames.push_back(phylipFileName); + string phylipFileName; + if ((outputForm == "lt") || (outputForm == "square")) { + phylipFileName = outputDir + m->getSimpleName(globaldata->getTreeFile()) + toString(i+1) + ".weighted.phylip.dist"; + outputNames.push_back(phylipFileName); outputTypes["phylip"].push_back(phylipFileName); + }else { //column + phylipFileName = outputDir + m->getSimpleName(globaldata->getTreeFile()) + toString(i+1) + ".weighted.column.dist"; + outputNames.push_back(phylipFileName); outputTypes["column"].push_back(phylipFileName); + } + ofstream out; m->openOutputFile(phylipFileName, out); - //output numSeqs - out << globaldata->Groups.size() << endl; - + if ((outputForm == "lt") || (outputForm == "square")) { + //output numSeqs + out << globaldata->Groups.size() << endl; + } + //make matrix with scores in it vector< vector > dists; dists.resize(globaldata->Groups.size()); for (int i = 0; i < globaldata->Groups.size(); i++) { @@ -466,11 +541,30 @@ void UnifracWeightedCommand::createPhylipFile() { if (name.length() < 10) { //pad with spaces to make compatible while (name.length() < 10) { name += " "; } } - out << name << '\t'; - //output distances - for (int l = 0; l < r; l++) { out << dists[r][l] << '\t'; } - out << endl; + if (outputForm == "lt") { + out << name << '\t'; + + //output distances + for (int l = 0; l < r; l++) { out << dists[r][l] << '\t'; } + out << endl; + }else if (outputForm == "square") { + out << name << '\t'; + + //output distances + for (int l = 0; l < globaldata->Groups.size(); l++) { out << dists[r][l] << '\t'; } + out << endl; + }else{ + //output distances + for (int l = 0; l < r; l++) { + string otherName = globaldata->Groups[l]; + if (otherName.length() < 10) { //pad with spaces to make compatible + while (otherName.length() < 10) { otherName += " "; } + } + + out << name << '\t' << otherName << dists[r][l] << endl; + } + } } out.close(); }