X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=parsimonycommand.cpp;h=64d249831ad56f5501dd6dd4d44f17bda35daf78;hp=c0de504f32f379651d15fc95767821ec9abd56c7;hb=d1c97b8c04bb75faca1e76ffad60b37a4d789d3d;hpb=c69e2e9749626cfbf1d6cb0125ae94f869e00b18 diff --git a/parsimonycommand.cpp b/parsimonycommand.cpp index c0de504..64d2498 100644 --- a/parsimonycommand.cpp +++ b/parsimonycommand.cpp @@ -8,76 +8,102 @@ */ #include "parsimonycommand.h" +#include "treereader.h" //********************************************************************************************************************** -vector ParsimonyCommand::getValidParameters(){ +vector ParsimonyCommand::setParameters(){ try { - string Array[] = {"random","groups","iters","processors","outputdir","inputdir"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + CommandParameter ptree("tree", "InputTypes", "", "", "none", "none", "none","parsimony-psummary",false,true,true); parameters.push_back(ptree); + CommandParameter pname("name", "InputTypes", "", "", "NameCount", "none", "none","",false,false,true); parameters.push_back(pname); + CommandParameter pcount("count", "InputTypes", "", "", "NameCount-CountGroup", "none", "none","",false,false,true); parameters.push_back(pcount); + CommandParameter pgroup("group", "InputTypes", "", "", "CountGroup", "none", "none","",false,false,true); parameters.push_back(pgroup); + CommandParameter pgroups("groups", "String", "", "", "", "", "","",false,false); parameters.push_back(pgroups); + CommandParameter prandom("random", "String", "", "", "", "", "","",false,false); parameters.push_back(prandom); + CommandParameter piters("iters", "Number", "", "1000", "", "", "","",false,false); parameters.push_back(piters); + CommandParameter pprocessors("processors", "Number", "", "1", "", "", "","",false,false,true); parameters.push_back(pprocessors); + 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, "ParsimonyCommand", "getValidParameters"); + m->errorOut(e, "ParsimonyCommand", "setParameters"); exit(1); } } //********************************************************************************************************************** -ParsimonyCommand::ParsimonyCommand(){ +string ParsimonyCommand::getHelpString(){ try { - abort = true; calledHelp = true; - vector tempOutNames; - outputTypes["parsimony"] = tempOutNames; - outputTypes["psummary"] = tempOutNames; + string helpString = ""; + helpString += "The parsimony command parameters are tree, group, name, count, random, groups, processors and iters. tree parameter is required unless you have valid current tree file or are using random.\n"; + helpString += "The groups parameter allows you to specify which of the groups in your groupfile you would like analyzed. You must enter at least 1 valid group.\n"; + helpString += "The group names are separated by dashes. The iters parameter allows you to specify how many random trees you would like compared to your tree.\n"; + helpString += "The parsimony command should be in the following format: parsimony(random=yourOutputFilename, groups=yourGroups, iters=yourIters).\n"; + helpString += "The processors parameter allows you to specify the number of processors to use. The default is 1.\n"; + helpString += "Example parsimony(random=out, iters=500).\n"; + helpString += "The default value for random is "" (meaning you want to use the trees in your inputfile, randomtree=out means you just want the random distribution of trees outputted to out.rd_parsimony),\n"; + helpString += "and iters is 1000. The parsimony command output two files: .parsimony and .psummary their descriptions are in the manual.\n"; + helpString += "Note: No spaces between parameter labels (i.e. random), '=' and parameters (i.e.yourOutputFilename).\n"; + return helpString; } catch(exception& e) { - m->errorOut(e, "ParsimonyCommand", "ParsimonyCommand"); + m->errorOut(e, "ParsimonyCommand", "getHelpString"); exit(1); } } //********************************************************************************************************************** -vector ParsimonyCommand::getRequiredParameters(){ - try { - vector myArray; - return myArray; - } - catch(exception& e) { - m->errorOut(e, "ParsimonyCommand", "getRequiredParameters"); - exit(1); - } +string ParsimonyCommand::getOutputPattern(string type) { + try { + string pattern = ""; + + if (type == "parsimony") { pattern = "[filename],parsimony"; } + else if (type == "psummary") { pattern = "[filename],psummary"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "ParsimonyCommand", "getOutputPattern"); + exit(1); + } } //********************************************************************************************************************** -vector ParsimonyCommand::getRequiredFiles(){ +ParsimonyCommand::ParsimonyCommand(){ try { - vector myArray; - return myArray; + abort = true; calledHelp = true; + setParameters(); + vector tempOutNames; + outputTypes["parsimony"] = tempOutNames; + outputTypes["psummary"] = tempOutNames; } catch(exception& e) { - m->errorOut(e, "ParsimonyCommand", "getRequiredFiles"); + m->errorOut(e, "ParsimonyCommand", "ParsimonyCommand"); exit(1); } } /***********************************************************/ ParsimonyCommand::ParsimonyCommand(string option) { try { - globaldata = GlobalData::getInstance(); abort = false; calledHelp = false; Groups.clear(); //allow user to run help 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[] = {"random","groups","processors","iters","outputdir","inputdir"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + vector myArray = setParameters(); OptionParser parser(option); map parameters = parser.getParameters(); + map::iterator it; ValidParameters validParameter; //check to make sure all parameters are valid for command - for (map::iterator it = parameters.begin(); it != parameters.end(); it++) { + for (it = parameters.begin(); it != parameters.end(); it++) { if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; } } @@ -86,69 +112,110 @@ ParsimonyCommand::ParsimonyCommand(string option) { outputTypes["parsimony"] = tempOutNames; outputTypes["psummary"] = tempOutNames; + //if the user changes the input directory command factory will send this info to us in the output parameter + string inputDir = validParameter.validFile(parameters, "inputdir", false); + if (inputDir == "not found"){ inputDir = ""; } + else { + string path; + it = parameters.find("tree"); + //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["tree"] = inputDir + it->second; } + } + + it = parameters.find("group"); + //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["group"] = inputDir + it->second; } + } + + it = parameters.find("name"); + //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["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; } + } + } + + outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; } + randomtree = validParameter.validFile(parameters, "random", false); if (randomtree == "not found") { randomtree = ""; } //are you trying to use parsimony without reading a tree or saying you want random distribution if (randomtree == "") { - if (globaldata->gTree.size() == 0) { - m->mothurOut("You must read a treefile and a groupfile or set the randomtree parameter to the output filename you wish, before you may execute the parsimony command."); m->mothurOutEndLine(); abort = true; } + //check for required parameters + treefile = validParameter.validFile(parameters, "tree", true); + if (treefile == "not open") { treefile = ""; abort = true; } + else if (treefile == "not found") { //if there is a current design file, use it + treefile = m->getTreeFile(); + if (treefile != "") { m->mothurOut("Using " + treefile + " as input file for the tree parameter."); m->mothurOutEndLine(); } + else { m->mothurOut("You have no current tree file and the tree parameter is required."); m->mothurOutEndLine(); abort = true; } + }else { m->setTreeFile(treefile); } + + //check for required parameters + groupfile = validParameter.validFile(parameters, "group", true); + if (groupfile == "not open") { abort = true; } + else if (groupfile == "not found") { groupfile = ""; } + else { m->setGroupFile(groupfile); } + + namefile = validParameter.validFile(parameters, "name", true); + if (namefile == "not open") { namefile = ""; abort = true; } + else if (namefile == "not found") { namefile = ""; } + else { m->setNameFile(namefile); } + + countfile = validParameter.validFile(parameters, "count", true); + if (countfile == "not open") { countfile = ""; abort = true; } + else if (countfile == "not found") { countfile = ""; } + else { m->setCountTableFile(countfile); } + + if ((namefile != "") && (countfile != "")) { + m->mothurOut("[ERROR]: you may only use one of the following: name or count."); m->mothurOutEndLine(); abort = true; + } + + if ((groupfile != "") && (countfile != "")) { + m->mothurOut("[ERROR]: you may only use one of the following: group or count."); m->mothurOutEndLine(); abort=true; + } + } //if the user changes the output directory command factory will send this info to us in the output parameter - string outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; if (randomtree == "") { outputDir += m->hasPath(globaldata->inputFileName); } } + string outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; if (randomtree == "") { outputDir += m->hasPath(treefile); } } //check for optional parameter and set defaults // ...at some point should added some additional type checking... groups = validParameter.validFile(parameters, "groups", false); - if (groups == "not found") { groups = ""; globaldata->Groups.clear(); } + if (groups == "not found") { groups = ""; m->clearGroups(); } else { m->splitAtDash(groups, Groups); - globaldata->Groups = Groups; + m->setGroups(Groups); } itersString = validParameter.validFile(parameters, "iters", false); if (itersString == "not found") { itersString = "1000"; } - convert(itersString, iters); + m->mothurConvert(itersString, iters); - string temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = "1"; } - convert(temp, processors); - - if (abort == false) { - //randomtree will tell us if user had their own treefile or if they just want the random distribution - //user has entered their own tree - if (randomtree == "") { - T = globaldata->gTree; - tmap = globaldata->gTreemap; - - if(outputDir == "") { outputDir += m->hasPath(globaldata->getTreeFile()); } - output = new ColumnFile(outputDir + m->getSimpleName(globaldata->getTreeFile()) + ".parsimony", itersString); - outputNames.push_back(outputDir + m->getSimpleName(globaldata->getTreeFile()) + ".parsimony"); - outputTypes["parsimony"].push_back(outputDir + m->getSimpleName(globaldata->getTreeFile()) + ".parsimony"); - - sumFile = outputDir + m->getSimpleName(globaldata->getTreeFile()) + ".psummary"; - m->openOutputFile(sumFile, outSum); - outputNames.push_back(sumFile); - outputTypes["psummary"].push_back(sumFile); - }else { //user wants random distribution - savetmap = globaldata->gTreemap; - getUserInput(); - - if(outputDir == "") { outputDir += m->hasPath(randomtree); } - output = new ColumnFile(outputDir+ m->getSimpleName(randomtree), itersString); - outputNames.push_back(outputDir+ m->getSimpleName(randomtree)); - outputTypes["parsimony"].push_back(outputDir+ m->getSimpleName(randomtree)); - } - - //set users groups to analyze - util = new SharedUtil(); - util->setGroups(globaldata->Groups, tmap->namesOfGroups, allGroups, numGroups, "parsimony"); //sets the groups the user wants to analyze - util->getCombos(groupComb, globaldata->Groups, numComp); - - if (numGroups == 1) { numComp++; groupComb.push_back(allGroups); } - - pars = new Parsimony(tmap); - counter = 0; - - } + string temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); } + m->setProcessors(temp); + m->mothurConvert(temp, processors); + + if (countfile=="") { + if (namefile == "") { + vector files; files.push_back(treefile); + parser.getNameFile(files); + } + } } @@ -158,43 +225,69 @@ ParsimonyCommand::ParsimonyCommand(string option) { exit(1); } } - -//********************************************************************************************************************** - -void ParsimonyCommand::help(){ - try { - m->mothurOut("The parsimony command can only be executed after a successful read.tree command, unless you use the random parameter.\n"); - m->mothurOut("The parsimony command parameters are random, groups, processors and iters. No parameters are required.\n"); - m->mothurOut("The groups parameter allows you to specify which of the groups in your groupfile you would like analyzed. You must enter at least 1 valid group.\n"); - m->mothurOut("The group names are separated by dashes. The iters parameter allows you to specify how many random trees you would like compared to your tree.\n"); - m->mothurOut("The parsimony command should be in the following format: parsimony(random=yourOutputFilename, groups=yourGroups, iters=yourIters).\n"); - m->mothurOut("The processors parameter allows you to specify the number of processors to use. The default is 1.\n"); - m->mothurOut("Example parsimony(random=out, iters=500).\n"); - m->mothurOut("The default value for random is "" (meaning you want to use the trees in your inputfile, randomtree=out means you just want the random distribution of trees outputted to out.rd_parsimony),\n"); - m->mothurOut("and iters is 1000. The parsimony command output two files: .parsimony and .psummary their descriptions are in the manual.\n"); - m->mothurOut("Note: No spaces between parameter labels (i.e. random), '=' and parameters (i.e.yourOutputFilename).\n\n"); - } - catch(exception& e) { - m->errorOut(e, "ParsimonyCommand", "help"); - exit(1); - } -} - - /***********************************************************/ int ParsimonyCommand::execute() { try { if (abort == true) { if (calledHelp) { return 0; } return 2; } + + + //randomtree will tell us if user had their own treefile or if they just want the random distribution + //user has entered their own tree + if (randomtree == "") { + + m->setTreeFile(treefile); + + TreeReader* reader; + if (countfile == "") { reader = new TreeReader(treefile, groupfile, namefile); } + else { reader = new TreeReader(treefile, countfile); } + T = reader->getTrees(); + ct = T[0]->getCountTable(); + delete reader; + + if(outputDir == "") { outputDir += m->hasPath(treefile); } + map variables; + variables["[filename]"] = outputDir + m->getSimpleName(treefile) + "."; + + output = new ColumnFile(getOutputFileName("parsimony",variables), itersString); + outputNames.push_back(getOutputFileName("parsimony",variables)); + outputTypes["parsimony"].push_back(getOutputFileName("parsimony",variables)); + + sumFile = getOutputFileName("psummary",variables); + m->openOutputFile(sumFile, outSum); + outputNames.push_back(sumFile); + outputTypes["psummary"].push_back(sumFile); + }else { //user wants random distribution + getUserInput(); + + if(outputDir == "") { outputDir += m->hasPath(randomtree); } + output = new ColumnFile(outputDir+ m->getSimpleName(randomtree), itersString); + outputNames.push_back(outputDir+ m->getSimpleName(randomtree)); + outputTypes["parsimony"].push_back(outputDir+ m->getSimpleName(randomtree)); + } + + //set users groups to analyze + SharedUtil util; + vector mGroups = m->getGroups(); + vector tGroups = ct->getNamesOfGroups(); + util.setGroups(mGroups, tGroups, allGroups, numGroups, "parsimony"); //sets the groups the user wants to analyze + util.getCombos(groupComb, mGroups, numComp); + m->setGroups(mGroups); + + if (numGroups == 1) { numComp++; groupComb.push_back(allGroups); } + + Parsimony pars; + counter = 0; Progress* reading; reading = new Progress("Comparing to random:", iters); if (m->control_pressed) { - delete reading; delete pars; delete util; delete output; + delete reading; delete output; + delete ct; for (int i = 0; i < T.size(); i++) { delete T[i]; } if (randomtree == "") { outSum.close(); } - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); - globaldata->Groups.clear(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear(); + m->clearGroups(); return 0; } @@ -212,13 +305,14 @@ int ParsimonyCommand::execute() { if (randomtree == "") { //get pscores for users trees for (int i = 0; i < T.size(); i++) { - userData = pars->getValues(T[i], processors, outputDir); //data = AB, AC, BC, ABC. + userData = pars.getValues(T[i], processors, outputDir); //data = AB, AC, BC, ABC. if (m->control_pressed) { - delete reading; delete pars; delete util; delete output; + delete reading; delete output; + delete ct; for (int i = 0; i < T.size(); i++) { delete T[i]; } if (randomtree == "") { outSum.close(); } - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); - globaldata->Groups.clear(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear(); + m->clearGroups(); return 0; } @@ -244,19 +338,20 @@ int ParsimonyCommand::execute() { for (int j = 0; j < iters; j++) { //create new tree with same num nodes and leaves as users - randT = new Tree(); + randT = new Tree(ct); //create random relationships between nodes randT->assembleRandomTree(); //get pscore of random tree - randomData = pars->getValues(randT, processors, outputDir); + randomData = pars.getValues(randT, processors, outputDir); if (m->control_pressed) { - delete reading; delete pars; delete util; delete output; delete randT; + delete reading; delete output; delete randT; if (randomtree == "") { outSum.close(); } - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); - globaldata->Groups.clear(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear(); + delete ct; for (int i = 0; i < T.size(); i++) { delete T[i]; } + m->clearGroups(); return 0; } @@ -284,29 +379,23 @@ int ParsimonyCommand::execute() { for (int j = 0; j < iters; j++) { //create new tree with same num nodes and leaves as users - randT = new Tree(); + randT = new Tree(ct); //create random relationships between nodes randT->assembleRandomTree(); if (m->control_pressed) { - delete reading; delete pars; delete util; delete output; delete randT; - globaldata->gTreemap = savetmap; - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); - globaldata->Groups.clear(); - return 0; + delete reading; delete output; delete randT; delete ct; + for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear(); return 0; } //get pscore of random tree - randomData = pars->getValues(randT, processors, outputDir); + randomData = pars.getValues(randT, processors, outputDir); if (m->control_pressed) { - delete reading; delete pars; delete util; delete output; delete randT; - globaldata->gTreemap = savetmap; - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); - globaldata->Groups.clear(); - return 0; + delete reading; delete output; delete randT; delete ct; + for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear(); return 0; } for(int r = 0; r < numComp; r++) { @@ -358,37 +447,23 @@ int ParsimonyCommand::execute() { } if (m->control_pressed) { - delete reading; delete pars; delete util; delete output; + delete reading; delete output; + delete ct; for (int i = 0; i < T.size(); i++) { delete T[i]; } if (randomtree == "") { outSum.close(); } - else { globaldata->gTreemap = savetmap; } - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); - globaldata->Groups.clear(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear(); return 0; } //finish progress bar reading->finish(); delete reading; - printParsimonyFile(); if (randomtree == "") { printUSummaryFile(); } + + delete output; delete ct; for (int i = 0; i < T.size(); i++) { delete T[i]; } - //reset globaldata's treemap if you just did random distrib - if (randomtree != "") { - //memory leak prevention - //if (globaldata->gTreemap != NULL) { delete globaldata->gTreemap; } - globaldata->gTreemap = savetmap; - } - - //reset groups parameter - globaldata->Groups.clear(); - - if (m->control_pressed) { - delete pars; delete util; delete output; - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); - return 0; - } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear(); return 0;} m->mothurOutEndLine(); m->mothurOut("Output File Names: "); m->mothurOutEndLine(); @@ -478,7 +553,7 @@ void ParsimonyCommand::getUserInput() { try { //create treemap - tmap = new TreeMap(); + ct = new CountTable(); m->mothurOut("Please enter the number of groups you would like to analyze: "); cin >> numGroups; @@ -488,40 +563,38 @@ void ParsimonyCommand::getUserInput() { count = 1; numEachGroup.resize(numGroups, 0); + set nameMap; + map groupMap; + set gps; + for (int i = 1; i <= numGroups; i++) { m->mothurOut("Please enter the number of sequences in group " + toString(i) + ": "); cin >> num; m->mothurOutJustToLog(toString(num)); m->mothurOutEndLine(); - - //set tmaps seqsPerGroup - tmap->seqsPerGroup[toString(i)] = num; - tmap->namesOfGroups.push_back(toString(i)); + gps.insert(toString(i)); + //set tmaps namesOfSeqs for (int j = 0; j < num; j++) { - tmap->namesOfSeqs.push_back(toString(count)); - tmap->treemap[toString(count)].groupname = toString(i); + groupMap[toString(count)] = toString(i); + nameMap.insert(toString(count)); count++; } } - + ct->createTable(nameMap, groupMap, gps); + //clears buffer so next command doesn't have error string s; getline(cin, s); - //save tmap for later - //memory leak prevention - //if (globaldata->gTreemap != NULL) { delete globaldata->gTreemap; } - globaldata->gTreemap = tmap; - globaldata->Treenames = tmap->namesOfSeqs; - + m->Treenames = ct->getNamesOfSeqs(); + m->runParse = false; } catch(exception& e) { m->errorOut(e, "ParsimonyCommand", "getUserInput"); exit(1); } } - /***********************************************************/