X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=libshuffcommand.cpp;h=46f1a53a475767d379059ca431f0d01e8596afa8;hp=219b0bf5fcb7149ff20ee7d65563ec1932147bf0;hb=cf9987b67aa49777a4c91c2d21f96e58bf17aa82;hpb=aa9238c0a9e6e7aa0ed8b8b606b08ad4fd7dcfe3 diff --git a/libshuffcommand.cpp b/libshuffcommand.cpp index 219b0bf..46f1a53 100644 --- a/libshuffcommand.cpp +++ b/libshuffcommand.cpp @@ -18,41 +18,160 @@ #include "slibshuff.h" #include "dlibshuff.h" + + +//********************************************************************************************************************** +vector LibShuffCommand::setParameters(){ + try { + CommandParameter pphylip("phylip", "InputTypes", "", "", "none", "none", "none","coverage-libshuffsummary",false,true,true); parameters.push_back(pphylip); + CommandParameter pgroup("group", "InputTypes", "", "", "none", "none", "none","",false,true,true); parameters.push_back(pgroup); + CommandParameter pgroups("groups", "String", "", "", "", "", "","",false,false); parameters.push_back(pgroups); + CommandParameter piters("iters", "Number", "", "10000", "", "", "","",false,false); parameters.push_back(piters); + CommandParameter pstep("step", "Number", "", "0.01", "", "", "","",false,false); parameters.push_back(pstep); + CommandParameter pcutoff("cutoff", "Number", "", "1.0", "", "", "","",false,false); parameters.push_back(pcutoff); + CommandParameter pform("form", "Multiple", "discrete-integral", "integral", "", "", "","",false,false); parameters.push_back(pform); + CommandParameter psim("sim", "Boolean", "", "F", "", "", "","",false,false); parameters.push_back(psim); + 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, "LibShuffCommand", "setParameters"); + exit(1); + } +} //********************************************************************************************************************** +string LibShuffCommand::getHelpString(){ + try { + string helpString = ""; + helpString += "The libshuff command parameters are phylip, group, sim, groups, iters, step, form and cutoff. phylip and group parameters are required, unless you have valid current files.\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 2 valid groups.\n"; + helpString += "The group names are separated by dashes. The iters parameter allows you to specify how many random matrices you would like compared to your matrix.\n"; + helpString += "The step parameter allows you to specify change in distance you would like between each output if you are using the discrete form.\n"; + helpString += "The form parameter allows you to specify if you would like to analyze your matrix using the discrete or integral form. Your options are integral or discrete.\n"; + helpString += "The libshuff command should be in the following format: libshuff(groups=yourGroups, iters=yourIters, cutOff=yourCutOff, form=yourForm, step=yourStep).\n"; + helpString += "Example libshuff(groups=A-B-C, iters=500, form=discrete, step=0.01, cutOff=2.0).\n"; + helpString += "The default value for groups is all the groups in your groupfile, iters is 10000, cutoff is 1.0, form is integral and step is 0.01.\n"; + helpString += "The libshuff command output two files: .coverage and .slsummary their descriptions are in the manual.\n"; + helpString += "Note: No spaces between parameter labels (i.e. iters), '=' and parameters (i.e.yourIters).\n"; + return helpString; + } + catch(exception& e) { + m->errorOut(e, "LibShuffCommand", "getHelpString"); + exit(1); + } +} +//********************************************************************************************************************** +string LibShuffCommand::getOutputPattern(string type) { + try { + string pattern = ""; + + if (type == "coverage") { pattern = "[filename],libshuff.coverage"; } + else if (type == "libshuffsummary") { pattern = "[filename],libshuff.summary"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "LibShuffCommand", "getOutputPattern"); + exit(1); + } +} +//********************************************************************************************************************** +LibShuffCommand::LibShuffCommand(){ + try { + abort = true; calledHelp = true; + setParameters(); + vector tempOutNames; + outputTypes["coverage"] = tempOutNames; + outputTypes["libshuffsummary"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "LibShuffCommand", "LibShuffCommand"); + exit(1); + } +} +//********************************************************************************************************************** LibShuffCommand::LibShuffCommand(string option) { try { - globaldata = GlobalData::getInstance(); - abort = false; - Groups.clear(); + 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[] = {"iters","groups","step","form","cutoff","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; } } + //initialize outputTypes + vector tempOutNames; + outputTypes["coverage"] = tempOutNames; + outputTypes["libshuffsummary"] = tempOutNames; + + string inputDir = validParameter.validFile(parameters, "inputdir", false); + if (inputDir == "not found"){ inputDir = ""; } + else { + string path; + it = parameters.find("phylip"); + //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["phylip"] = 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; } + } + } + + //check for required parameters + phylipfile = validParameter.validFile(parameters, "phylip", true); + if (phylipfile == "not open") { phylipfile = ""; abort = true; } + else if (phylipfile == "not found") { + phylipfile = m->getPhylipFile(); + if (phylipfile != "") { m->mothurOut("Using " + phylipfile + " as input file for the phylip parameter."); m->mothurOutEndLine(); } + else { + m->mothurOut("You must provide a phylip file."); m->mothurOutEndLine(); + abort = true; + } + }else { m->setPhylipFile(phylipfile); } + + //check for required parameters + groupfile = validParameter.validFile(parameters, "group", true); + if (groupfile == "not open") { groupfile = ""; abort = true; } + else if (groupfile == "not found") { + groupfile = m->getGroupFile(); + if (groupfile != "") { m->mothurOut("Using " + groupfile + " as input file for the group parameter."); m->mothurOutEndLine(); } + else { + m->mothurOut("You must provide a group file."); m->mothurOutEndLine(); + abort = true; + } + }else { m->setGroupFile(groupfile); } + //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"){ outputDir = ""; - outputDir += hasPath(globaldata->getPhylipFile()); //if user entered a file with a path then preserve it - } - - //make sure the user has already run the read.dist command - if ((globaldata->gMatrix == NULL) || (globaldata->gGroupmap == NULL)) { - m->mothurOut("You must read in a matrix and groupfile using the read.dist command, before you use the libshuff command. "); m->mothurOutEndLine(); abort = true;; + outputDir += m->hasPath(phylipfile); //if user entered a file with a path then preserve it } //check for optional parameter and set defaults @@ -61,44 +180,24 @@ LibShuffCommand::LibShuffCommand(string option) { if (groups == "not found") { groups = ""; savegroups = groups; } else { savegroups = groups; - splitAtDash(groups, Groups); - globaldata->Groups = Groups; + m->splitAtDash(groups, Groups); + m->setGroups(Groups); } string temp; temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "10000"; } - convert(temp, iters); + m->mothurConvert(temp, iters); temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "1.0"; } - convert(temp, cutOff); + m->mothurConvert(temp, cutOff); temp = validParameter.validFile(parameters, "step", false); if (temp == "not found") { temp = "0.01"; } - convert(temp, step); - - userform = validParameter.validFile(parameters, "form", false); if (userform == "not found") { userform = "integral"; } + m->mothurConvert(temp, step); - if (abort == false) { - - matrix = globaldata->gMatrix; //get the distance matrix - setGroups(); //set the groups to be analyzed and sorts them - - /********************************************************************************************/ - //this is needed because when we read the matrix we sort it into groups in alphabetical order - //the rest of the command and the classes used in this command assume specific order - /********************************************************************************************/ - matrix->setGroups(globaldata->gGroupmap->namesOfGroups); - vector sizes; - for (int i = 0; i < globaldata->gGroupmap->namesOfGroups.size(); i++) { sizes.push_back(globaldata->gGroupmap->getNumSeqs(globaldata->gGroupmap->namesOfGroups[i])); } - matrix->setSizes(sizes); + temp = validParameter.validFile(parameters, "sim", false); if (temp == "not found") { temp = "F"; } + sim = m->isTrue(temp); - - if(userform == "discrete"){ - form = new DLibshuff(matrix, iters, step, cutOff); - } - else{ - form = new SLibshuff(matrix, iters, cutOff); - } - } + userform = validParameter.validFile(parameters, "form", false); if (userform == "not found") { userform = "integral"; } } @@ -110,65 +209,110 @@ LibShuffCommand::LibShuffCommand(string option) { } //********************************************************************************************************************** -void LibShuffCommand::help(){ - try { - m->mothurOut("The libshuff command can only be executed after a successful read.dist command including a groupfile.\n"); - m->mothurOut("The libshuff command parameters are groups, iters, step, form and cutoff. 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 2 valid groups.\n"); - m->mothurOut("The group names are separated by dashes. The iters parameter allows you to specify how many random matrices you would like compared to your matrix.\n"); - m->mothurOut("The step parameter allows you to specify change in distance you would like between each output if you are using the discrete form.\n"); - m->mothurOut("The form parameter allows you to specify if you would like to analyze your matrix using the discrete or integral form. Your options are integral or discrete.\n"); - m->mothurOut("The libshuff command should be in the following format: libshuff(groups=yourGroups, iters=yourIters, cutOff=yourCutOff, form=yourForm, step=yourStep).\n"); - m->mothurOut("Example libshuff(groups=A-B-C, iters=500, form=discrete, step=0.01, cutOff=2.0).\n"); - m->mothurOut("The default value for groups is all the groups in your groupfile, iters is 10000, cutoff is 1.0, form is integral and step is 0.01.\n"); - m->mothurOut("The libshuff command output two files: .coverage and .slsummary their descriptions are in the manual.\n"); - m->mothurOut("Note: No spaces between parameter labels (i.e. iters), '=' and parameters (i.e.yourIters).\n\n"); - } - catch(exception& e) { - m->errorOut(e, "LibShuffCommand", "help"); - exit(1); - } -} - -//********************************************************************************************************************** - int LibShuffCommand::execute(){ try { - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } + + //read files + groupMap = new GroupMap(groupfile); + int error = groupMap->readMap(); + if (error == 1) { delete groupMap; return 0; } + + ifstream in; + m->openInputFile(phylipfile, in); + matrix = new FullMatrix(in, groupMap, sim); //reads the matrix file + in.close(); + + if (m->control_pressed) { delete groupMap; delete matrix; return 0; } + + //if files don't match... + if (matrix->getNumSeqs() < groupMap->getNumSeqs()) { + m->mothurOut("Your distance file contains " + toString(matrix->getNumSeqs()) + " sequences, and your group file contains " + toString(groupMap->getNumSeqs()) + " sequences."); m->mothurOutEndLine(); + //create new group file + if(outputDir == "") { outputDir += m->hasPath(groupfile); } + + string newGroupFile = outputDir + m->getRootName(m->getSimpleName(groupfile)) + "editted.groups"; + outputNames.push_back(newGroupFile); + ofstream outGroups; + m->openOutputFile(newGroupFile, outGroups); + + for (int i = 0; i < matrix->getNumSeqs(); i++) { + if (m->control_pressed) { delete groupMap; delete matrix; outGroups.close(); m->mothurRemove(newGroupFile); return 0; } + + Names temp = matrix->getRowInfo(i); + outGroups << temp.seqName << '\t' << temp.groupName << endl; + } + outGroups.close(); + + m->mothurOut(newGroupFile + " is a new group file containing only the sequence that are in your distance file. I will read this file instead."); m->mothurOutEndLine(); + + //read new groupfile + delete groupMap; + groupfile = newGroupFile; + + groupMap = new GroupMap(groupfile); + groupMap->readMap(); + + if (m->control_pressed) { delete groupMap; delete matrix; m->mothurRemove(newGroupFile); return 0; } + } + + + setGroups(); //set the groups to be analyzed and sorts them + + if (numGroups < 2) { m->mothurOut("[ERROR]: libshuff requires at least 2 groups, you only have " + toString(numGroups) + ", aborting."); m->mothurOutEndLine(); m->control_pressed = true; } + + if (m->control_pressed) { delete groupMap; delete matrix; return 0; } + + /********************************************************************************************/ + //this is needed because when we read the matrix we sort it into groups in alphabetical order + //the rest of the command and the classes used in this command assume specific order + /********************************************************************************************/ + matrix->setGroups(groupMap->getNamesOfGroups()); + vector sizes; + for (int i = 0; i < (groupMap->getNamesOfGroups()).size(); i++) { sizes.push_back(groupMap->getNumSeqs((groupMap->getNamesOfGroups())[i])); } + matrix->setSizes(sizes); + + + if(userform == "discrete"){ + form = new DLibshuff(matrix, iters, step, cutOff); + } + else{ + form = new SLibshuff(matrix, iters, cutOff); + } savedDXYValues = form->evaluateAll(); savedMinValues = form->getSavedMins(); - if (m->control_pressed) { delete form; globaldata->Groups.clear(); delete globaldata->gMatrix; globaldata->gMatrix = NULL; return 0; } + if (m->control_pressed) { delete form; m->clearGroups(); delete matrix; delete groupMap; return 0; } pValueCounts.resize(numGroups); for(int i=0;icontrol_pressed) { delete form; globaldata->Groups.clear(); delete globaldata->gMatrix; globaldata->gMatrix = NULL; return 0; } + if (m->control_pressed) { outputTypes.clear(); delete form; m->clearGroups(); delete matrix; delete groupMap; return 0; } Progress* reading = new Progress(); for(int i=0;icontrol_pressed) { delete form; globaldata->Groups.clear(); delete globaldata->gMatrix; globaldata->gMatrix = NULL; delete reading; return 0; } + if (m->control_pressed) { outputTypes.clear(); delete form; m->clearGroups(); delete matrix; delete groupMap; delete reading; return 0; } reading->newLine(groupNames[i]+'-'+groupNames[j], iters); - int spoti = globaldata->gGroupmap->groupIndex[groupNames[i]]; //neccessary in case user selects groups so you know where they are in the matrix - int spotj = globaldata->gGroupmap->groupIndex[groupNames[j]]; + int spoti = groupMap->groupIndex[groupNames[i]]; //neccessary in case user selects groups so you know where they are in the matrix + int spotj = groupMap->groupIndex[groupNames[j]]; for(int p=0;pcontrol_pressed) { delete form; globaldata->Groups.clear(); delete globaldata->gMatrix; globaldata->gMatrix = NULL; delete reading; return 0; } + if (m->control_pressed) { outputTypes.clear(); delete form; m->clearGroups(); delete matrix; delete groupMap; delete reading; return 0; } form->randomizeGroups(spoti,spotj); if(form->evaluatePair(spoti,spotj) >= savedDXYValues[spoti][spotj]) { pValueCounts[i][j]++; } if(form->evaluatePair(spotj,spoti) >= savedDXYValues[spotj][spoti]) { pValueCounts[j][i]++; } - if (m->control_pressed) { delete form; globaldata->Groups.clear(); delete globaldata->gMatrix; globaldata->gMatrix = NULL; delete reading; return 0; } + if (m->control_pressed) { outputTypes.clear(); delete form; m->clearGroups(); delete matrix; delete groupMap; delete reading; return 0; } reading->update(p); } @@ -177,7 +321,7 @@ int LibShuffCommand::execute(){ } } - if (m->control_pressed) { delete form; globaldata->Groups.clear(); delete globaldata->gMatrix; globaldata->gMatrix = NULL; delete reading; return 0; } + if (m->control_pressed) { outputTypes.clear(); delete form; m->clearGroups(); delete matrix; delete groupMap; delete reading; return 0; } reading->finish(); delete reading; @@ -187,15 +331,13 @@ int LibShuffCommand::execute(){ printCoverageFile(); //clear out users groups - globaldata->Groups.clear(); + m->clearGroups(); delete form; - //delete globaldata's copy of the gmatrix to free up memory - delete globaldata->gMatrix; globaldata->gMatrix = NULL; + delete matrix; delete groupMap; - if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } return 0; } + if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } - m->mothurOutEndLine(); m->mothurOut("Output File Names: "); m->mothurOutEndLine(); for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } @@ -215,9 +357,11 @@ int LibShuffCommand::printCoverageFile() { try { ofstream outCov; - summaryFile = outputDir + getRootName(getSimpleName(globaldata->getPhylipFile())) + "libshuff.coverage"; - openOutputFile(summaryFile, outCov); - outputNames.push_back(summaryFile); + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(phylipfile)); + summaryFile = getOutputFileName("coverage", variables); + m->openOutputFile(summaryFile, outCov); + outputNames.push_back(summaryFile); outputTypes["coverage"].push_back(summaryFile); outCov.setf(ios::fixed, ios::floatfield); outCov.setf(ios::showpoint); //cout.setf(ios::fixed, ios::floatfield); cout.setf(ios::showpoint); @@ -233,8 +377,8 @@ int LibShuffCommand::printCoverageFile() { for(int j=0;jgGroupmap->groupIndex[groupNames[i]]; //neccessary in case user selects groups so you know where they are in the matrix - int spotj = globaldata->gGroupmap->groupIndex[groupNames[j]]; + int spoti = groupMap->groupIndex[groupNames[i]]; //neccessary in case user selects groups so you know where they are in the matrix + int spotj = groupMap->groupIndex[groupNames[j]]; for(int k=0;kgetPhylipFile())) + "libshuff.summary"; - openOutputFile(summaryFile, outSum); - outputNames.push_back(summaryFile); + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(phylipfile)); + summaryFile = getOutputFileName("libshuffsummary",variables); + m->openOutputFile(summaryFile, outSum); + outputNames.push_back(summaryFile); outputTypes["libshuffsummary"].push_back(summaryFile); outSum.setf(ios::fixed, ios::floatfield); outSum.setf(ios::showpoint); cout.setf(ios::fixed, ios::floatfield); cout.setf(ios::showpoint); @@ -327,8 +473,8 @@ int LibShuffCommand::printSummaryFile() { for(int j=i+1;jcontrol_pressed) { outSum.close(); return 0; } - int spoti = globaldata->gGroupmap->groupIndex[groupNames[i]]; //neccessary in case user selects groups so you know where they are in the matrix - int spotj = globaldata->gGroupmap->groupIndex[groupNames[j]]; + int spoti = groupMap->groupIndex[groupNames[i]]; //neccessary in case user selects groups so you know where they are in the matrix + int spotj = groupMap->groupIndex[groupNames[j]]; if(pValueCounts[i][j]){ cout << setw(20) << left << groupNames[i]+'-'+groupNames[j] << '\t' << setprecision(8) << savedDXYValues[spoti][spotj] << '\t' << setprecision(precision) << pValueCounts[i][j]/(float)iters << endl; @@ -366,49 +512,51 @@ int LibShuffCommand::printSummaryFile() { void LibShuffCommand::setGroups() { try { + vector myGroups = m->getGroups(); //if the user has not entered specific groups to analyze then do them all - if (globaldata->Groups.size() == 0) { - numGroups = globaldata->gGroupmap->getNumGroups(); + if (m->getNumGroups() == 0) { + numGroups = groupMap->getNumGroups(); for (int i=0; i < numGroups; i++) { - globaldata->Groups.push_back(globaldata->gGroupmap->namesOfGroups[i]); + myGroups.push_back((groupMap->getNamesOfGroups())[i]); } } else { if (savegroups != "all") { //check that groups are valid - for (int i = 0; i < globaldata->Groups.size(); i++) { - if (globaldata->gGroupmap->isValidGroup(globaldata->Groups[i]) != true) { - m->mothurOut(globaldata->Groups[i] + " is not a valid group, and will be disregarded."); m->mothurOutEndLine(); + for (int i = 0; i < myGroups.size(); i++) { + if (groupMap->isValidGroup(myGroups[i]) != true) { + m->mothurOut(myGroups[i] + " is not a valid group, and will be disregarded."); m->mothurOutEndLine(); // erase the invalid group from globaldata->Groups - globaldata->Groups.erase(globaldata->Groups.begin()+i); + myGroups.erase(myGroups.begin()+i); } } //if the user only entered invalid groups - if ((globaldata->Groups.size() == 0) || (globaldata->Groups.size() == 1)) { - numGroups = globaldata->gGroupmap->getNumGroups(); + if ((myGroups.size() == 0) || (myGroups.size() == 1)) { + numGroups = groupMap->getNumGroups(); for (int i=0; i < numGroups; i++) { - globaldata->Groups.push_back(globaldata->gGroupmap->namesOfGroups[i]); + myGroups.push_back((groupMap->getNamesOfGroups())[i]); } m->mothurOut("When using the groups parameter you must have at least 2 valid groups. I will run the command using all the groups in your groupfile."); m->mothurOutEndLine(); - } else { numGroups = globaldata->Groups.size(); } + } else { numGroups = myGroups.size(); } } else { //users wants all groups - numGroups = globaldata->gGroupmap->getNumGroups(); - globaldata->Groups.clear(); + numGroups = groupMap->getNumGroups(); + myGroups.clear(); for (int i=0; i < numGroups; i++) { - globaldata->Groups.push_back(globaldata->gGroupmap->namesOfGroups[i]); + myGroups.push_back((groupMap->getNamesOfGroups())[i]); } } } //sort so labels match - sort(globaldata->Groups.begin(), globaldata->Groups.end()); + sort(myGroups.begin(), myGroups.end()); //sort - sort(globaldata->gGroupmap->namesOfGroups.begin(), globaldata->gGroupmap->namesOfGroups.end()); + //sort(groupMap->namesOfGroups.begin(), groupMap->namesOfGroups.end()); - for (int i = 0; i < globaldata->gGroupmap->namesOfGroups.size(); i++) { globaldata->gGroupmap->groupIndex[globaldata->gGroupmap->namesOfGroups[i]] = i; } + for (int i = 0; i < (groupMap->getNamesOfGroups()).size(); i++) { groupMap->groupIndex[(groupMap->getNamesOfGroups())[i]] = i; } - groupNames = globaldata->Groups; + groupNames = myGroups; + m->setGroups(myGroups); } catch(exception& e) {