X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=summarysharedcommand.cpp;h=1a3ecf058c898529b17f2d7c67209949722ccb0b;hb=163b300cfd7d4ca4e70c454be20f07b1d8346650;hp=f31f7e10e1f19e47781d2d47dd9db09ebafa05cf;hpb=09a01fb51eea9e81409d367410c831394193fd86;p=mothur.git diff --git a/summarysharedcommand.cpp b/summarysharedcommand.cpp index f31f7e1..1a3ecf0 100644 --- a/summarysharedcommand.cpp +++ b/summarysharedcommand.cpp @@ -11,6 +11,7 @@ #include "sharedsobscollectsummary.h" #include "sharedchao1.h" #include "sharedace.h" +#include "sharednseqs.h" #include "sharedjabund.h" #include "sharedsorabund.h" #include "sharedjclass.h" @@ -19,49 +20,150 @@ #include "sharedsorest.h" #include "sharedthetayc.h" #include "sharedthetan.h" +#include "sharedkstest.h" +#include "whittaker.h" +#include "sharedochiai.h" +#include "sharedanderbergs.h" +#include "sharedkulczynski.h" +#include "sharedkulczynskicody.h" +#include "sharedlennon.h" +#include "sharedmorisitahorn.h" +#include "sharedbraycurtis.h" +#include "sharedjackknife.h" +#include "whittaker.h" + //********************************************************************************************************************** -SummarySharedCommand::SummarySharedCommand(){ +SummarySharedCommand::SummarySharedCommand(string option){ try { globaldata = GlobalData::getInstance(); - outputFileName = ((getRootName(globaldata->inputFileName)) + "shared.summary"); - openOutputFile(outputFileName, outputFileHandle); - format = globaldata->getFormat(); - validCalculator = new ValidCalculators(); - - int i; - for (i=0; iEstimators.size(); i++) { - if (validCalculator->isValidCalculator("sharedsummary", globaldata->Estimators[i]) == true) { - if (globaldata->Estimators[i] == "sharedsobs") { - sumCalculators.push_back(new SharedSobsCS()); - }else if (globaldata->Estimators[i] == "sharedchao") { - sumCalculators.push_back(new SharedChao1()); - }else if (globaldata->Estimators[i] == "sharedace") { - sumCalculators.push_back(new SharedAce()); - }else if (globaldata->Estimators[i] == "sharedjabund") { - sumCalculators.push_back(new SharedJAbund()); - }else if (globaldata->Estimators[i] == "sharedsorensonabund") { - sumCalculators.push_back(new SharedSorAbund()); - }else if (globaldata->Estimators[i] == "sharedjclass") { - sumCalculators.push_back(new SharedJclass()); - }else if (globaldata->Estimators[i] == "sharedsorclass") { - sumCalculators.push_back(new SharedSorClass()); - }else if (globaldata->Estimators[i] == "sharedjest") { - sumCalculators.push_back(new SharedJest()); - }else if (globaldata->Estimators[i] == "sharedsorest") { - sumCalculators.push_back(new SharedSorEst()); - }else if (globaldata->Estimators[i] == "sharedthetayc") { - sumCalculators.push_back(new SharedThetaYC()); - }else if (globaldata->Estimators[i] == "sharedthetan") { - sumCalculators.push_back(new SharedThetaN()); + abort = false; + allLines = 1; + lines.clear(); + labels.clear(); + Estimators.clear(); + + //allow user to run help + if(option == "help") { validCalculator = new ValidCalculators(); help(); abort = true; } + + else { + //valid paramters for this command + string Array[] = {"line","label","calc","groups"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + + OptionParser parser(option); + map parameters = parser.getParameters(); + + ValidParameters validParameter; + + //check to make sure all parameters are valid for command + for (map::iterator it = parameters.begin(); it != parameters.end(); it++) { + if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; } + } + + //make sure the user has already run the read.otu command + if (globaldata->getSharedFile() == "") { + cout << "You must read a list and a group, or a shared before you can use the summary.shared command." << endl; abort = true; + } + + //check for optional parameter and set defaults + // ...at some point should added some additional type checking... + line = validParameter.validFile(parameters, "line", false); + if (line == "not found") { line = ""; } + else { + if(line != "all") { splitAtDash(line, lines); allLines = 0; } + else { allLines = 1; } + } + + label = validParameter.validFile(parameters, "label", false); + if (label == "not found") { label = ""; } + else { + if(label != "all") { splitAtDash(label, labels); allLines = 0; } + else { allLines = 1; } + } + + //make sure user did not use both the line and label parameters + if ((line != "") && (label != "")) { cout << "You cannot use both the line and label parameters at the same time. " << endl; abort = true; } + //if the user has not specified any line or labels use the ones from read.otu + else if((line == "") && (label == "")) { + allLines = globaldata->allLines; + labels = globaldata->labels; + lines = globaldata->lines; + } + + calc = validParameter.validFile(parameters, "calc", false); + if (calc == "not found") { calc = "sharedsobs-sharedchao-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan"; } + else { + if (calc == "default") { calc = "sharedsobs-sharedchao-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan"; } + } + splitAtDash(calc, Estimators); + + groups = validParameter.validFile(parameters, "groups", false); + if (groups == "not found") { groups = ""; } + else { + splitAtDash(groups, Groups); + globaldata->Groups = Groups; + } + + if (abort == false) { + + validCalculator = new ValidCalculators(); + int i; + + for (i=0; iisValidCalculator("sharedsummary", Estimators[i]) == true) { + if (Estimators[i] == "sharedsobs") { + sumCalculators.push_back(new SharedSobsCS()); + }else if (Estimators[i] == "sharedchao") { + sumCalculators.push_back(new SharedChao1()); + }else if (Estimators[i] == "sharedace") { + sumCalculators.push_back(new SharedAce()); + }else if (Estimators[i] == "jabund") { + sumCalculators.push_back(new JAbund()); + }else if (Estimators[i] == "sorabund") { + sumCalculators.push_back(new SorAbund()); + }else if (Estimators[i] == "jclass") { + sumCalculators.push_back(new Jclass()); + }else if (Estimators[i] == "sorclass") { + sumCalculators.push_back(new SorClass()); + }else if (Estimators[i] == "jest") { + sumCalculators.push_back(new Jest()); + }else if (Estimators[i] == "sorest") { + sumCalculators.push_back(new SorEst()); + }else if (Estimators[i] == "thetayc") { + sumCalculators.push_back(new ThetaYC()); + }else if (Estimators[i] == "thetan") { + sumCalculators.push_back(new ThetaN()); + }else if (Estimators[i] == "kstest") { + sumCalculators.push_back(new KSTest()); + }else if (Estimators[i] == "sharednseqs") { + sumCalculators.push_back(new SharedNSeqs()); + }else if (Estimators[i] == "ochiai") { + sumCalculators.push_back(new Ochiai()); + }else if (Estimators[i] == "anderberg") { + sumCalculators.push_back(new Anderberg()); + }else if (Estimators[i] == "kulczynski") { + sumCalculators.push_back(new Kulczynski()); + }else if (Estimators[i] == "kulczynskicody") { + sumCalculators.push_back(new KulczynskiCody()); + }else if (Estimators[i] == "lennon") { + sumCalculators.push_back(new Lennon()); + }else if (Estimators[i] == "morisitahorn") { + sumCalculators.push_back(new MorHorn()); + }else if (Estimators[i] == "braycurtis") { + sumCalculators.push_back(new BrayCurtis()); + }else if (Estimators[i] == "whittaker") { + sumCalculators.push_back(new Whittaker()); + } + } } + + outputFileName = ((getRootName(globaldata->inputFileName)) + "shared.summary"); + openOutputFile(outputFileName, outputFileHandle); + mult = false; } } - - //reset calc for next command - globaldata->setCalc(""); - } catch(exception& e) { cout << "Standard Error: " << e.what() << " has occurred in the SummarySharedCommand class Function SummarySharedCommand. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; @@ -72,85 +174,171 @@ SummarySharedCommand::SummarySharedCommand(){ exit(1); } } + +//********************************************************************************************************************** + +void SummarySharedCommand::help(){ + try { + cout << "The summary.shared command can only be executed after a successful read.otu command." << "\n"; + cout << "The summary.shared command parameters are label, line and calc. No parameters are required, but you may not use " << "\n"; + cout << "both the line and label parameters at the same time. The summary.shared command should be in the following format: " << "\n"; + cout << "summary.shared(label=yourLabel, line=yourLines, calc=yourEstimators, groups=yourGroups)." << "\n"; + cout << "Example summary.shared(label=unique-.01-.03, line=0,5,10, groups=B-C, calc=sharedchao-sharedace-jabund-sorensonabund-jclass-sorclass-jest-sorest-thetayc-thetan)." << "\n"; + validCalculator->printCalc("sharedsummary", cout); + cout << "The default value for calc is sharedsobs-sharedchao-sharedace-jabund-sorensonabund-jclass-sorclass-jest-sorest-thetayc-thetan" << "\n"; + cout << "The default value for groups is all the groups in your groupfile." << "\n"; + cout << "The label and line parameters are used to analyze specific lines in your input." << "\n"; + cout << "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"; + cout << "Note: No spaces between parameter labels (i.e. line), '=' and parameters (i.e.yourLines)." << "\n" << "\n"; + } + catch(exception& e) { + cout << "Standard Error: " << e.what() << " has occurred in the SummarySharedCommand class Function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } + catch(...) { + cout << "An unknown error has occurred in the SummarySharedCommand class function help. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + exit(1); + } +} + //********************************************************************************************************************** SummarySharedCommand::~SummarySharedCommand(){ - delete input; - delete read; + if (abort == false) { + delete read; + delete validCalculator; + } } //********************************************************************************************************************** int SummarySharedCommand::execute(){ try { - int count = 1; + + if (abort == true) { return 0; } + int count = 1; + //if the users entered no valid calculators don't execute command if (sumCalculators.size() == 0) { return 0; } - - if (format == "sharedfile") { - read = new ReadPhilFile(globaldata->inputFileName); - read->read(&*globaldata); - - input = globaldata->ginput; - order = input->getSharedOrderVector(); - }else { - //you are using a list and a groupfile - read = new ReadPhilFile(globaldata->inputFileName); - read->read(&*globaldata); - - input = globaldata->ginput; - SharedList = globaldata->gSharedList; - order = SharedList->getSharedOrderVector(); + //check if any calcs can do multiples + else{ + for (int i = 0; i < sumCalculators.size(); i++) { + if (sumCalculators[i]->getMultiple() == true) { mult = true; } + } } + //read first line + read = new ReadOTUFile(globaldata->inputFileName); + read->read(&*globaldata); + + input = globaldata->ginput; + lookup = input->getSharedRAbundVectors(); + vector lastLookup = lookup; + //output estimator names as column headers outputFileHandle << "label" <<'\t' << "comparison" << '\t'; for(int i=0;igetName(); } outputFileHandle << endl; - - while(order != NULL){ - if(globaldata->allLines == 1 || globaldata->lines.count(count) == 1 || globaldata->labels.count(order->getLabel()) == 1){ - - cout << order->getLabel() << '\t' << count << endl; - getSharedVectors(); //fills group vectors from order vector. - - //randomize group order - if (globaldata->getJumble() == "1") { random_shuffle(lookup.begin(), lookup.end()); } - - int n = 1; - for (int k = 0; k < (lookup.size() - 1); k++) { // pass cdd each set of groups to commpare - for (int l = n; l < lookup.size(); l++) { - outputFileHandle << order->getLabel() << '\t' << (lookup[k]->getGroup() + lookup[l]->getGroup()) << '\t' << '\t'; //print out label and group - for(int i=0;igetValues(lookup[k], lookup[l]); //saves the calculator outputs - outputFileHandle << '\t'; - sumCalculators[i]->print(outputFileHandle); - } - outputFileHandle << endl; - } - n++; + //create file and put column headers for multiple groups file + if (mult == true) { + outAllFileName = ((getRootName(globaldata->inputFileName)) + "sharedmultiple.summary"); + openOutputFile(outAllFileName, outAll); + + outAll << "label" <<'\t' << "comparison" << '\t'; + for(int i=0;igetMultiple() == true) { + outAll << '\t' << sumCalculators[i]->getName(); } } + outAll << endl; + } + + if (lookup.size() < 2) { + cout << "I cannot run the command without at least 2 valid groups."; + for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } + + //close files and clean up + outputFileHandle.close(); remove(outputFileName.c_str()); + if (mult == true) { outAll.close(); remove(outAllFileName.c_str()); } + return 0; + //if you only have 2 groups you don't need a .sharedmultiple file + }else if ((lookup.size() == 2) && (mult == true)) { + mult = false; + outAll.close(); + remove(outAllFileName.c_str()); + } + + //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label. + set processedLabels; + set userLabels = labels; + set userLines = lines; + + //as long as you are not at the end of the file or done wih the lines you want + while((lookup[0] != NULL) && ((allLines == 1) || (userLabels.size() != 0) || (userLines.size() != 0))) { + + if(allLines == 1 || lines.count(count) == 1 || labels.count(lookup[0]->getLabel()) == 1){ + cout << lookup[0]->getLabel() << '\t' << count << endl; + process(lookup); + + processedLabels.insert(lookup[0]->getLabel()); + userLabels.erase(lookup[0]->getLabel()); + userLines.erase(count); + } + + if ((anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLookup[0]->getLabel()) != 1)) { + cout << lastLookup[0]->getLabel() << '\t' << count << endl; + process(lastLookup); + + processedLabels.insert(lastLookup[0]->getLabel()); + userLabels.erase(lastLookup[0]->getLabel()); + } + + //prevent memory leak + if (count != 1) { for (int i = 0; i < lastLookup.size(); i++) { delete lastLookup[i]; } } + lastLookup = lookup; + //get next line to process - if (format == "sharedfile") { - order = input->getSharedOrderVector(); + lookup = input->getSharedRAbundVectors(); + count++; + } + + //output error messages about any remaining user labels + set::iterator it; + bool needToRun = false; + for (it = userLabels.begin(); it != userLabels.end(); it++) { + cout << "Your file does not include the label "<< *it; + if (processedLabels.count(lastLookup[0]->getLabel()) != 1) { + cout << ". I will use " << lastLookup[0]->getLabel() << "." << endl; + needToRun = true; }else { - //you are using a list and a groupfile - SharedList = input->getSharedListVector(); //get new list vector to process - if (SharedList != NULL) { - order = SharedList->getSharedOrderVector(); //gets new order vector with group info. - }else { - break; - } + cout << ". Please refer to " << lastLookup[0]->getLabel() << "." << endl; } - count++; } - + + //run last line if you need to + if (needToRun == true) { + cout << lastLookup[0]->getLabel() << '\t' << count << endl; + process(lastLookup); + } + + for (int i = 0; i < lastLookup.size(); i++) { delete lastLookup[i]; } + + //reset groups parameter + globaldata->Groups.clear(); + + //close files + outputFileHandle.close(); + if (mult == true) { outAll.close(); } + + for(int i=0;iginput = NULL; + return 0; } catch(exception& e) { @@ -163,46 +351,70 @@ int SummarySharedCommand::execute(){ } } -//********************************************************************************************************************** - -void SummarySharedCommand::getSharedVectors(){ -try { - lookup.clear(); - //create and initialize vector of sharedvectors, one for each group - for (int i = 0; i < globaldata->gGroupmap->getNumGroups(); i++) { - SharedRAbundVector* temp = new SharedRAbundVector(order->getNumBins()); - temp->setLabel(order->getLabel()); - temp->setGroup(globaldata->gGroupmap->namesOfGroups[i]); - lookup.push_back(temp); - } - - int numSeqs = order->size(); - //sample all the members - for(int i=0;iget(i); - int abundance; +/***********************************************************/ +void SummarySharedCommand::process(vector thisLookup) { + try { + //loop through calculators and add to file all for all calcs that can do mutiple groups + if (mult == true) { + //output label + outAll << thisLookup[0]->getLabel() << '\t'; + + //output groups names + string outNames = ""; + for (int j = 0; j < thisLookup.size(); j++) { + outNames += thisLookup[j]->getGroup() + "-"; + } + outNames = outNames.substr(0, outNames.length()-1); //rip off extra '-'; + outAll << outNames << '\t'; - //set info for sharedvector in chosens group - for (int j = 0; j < lookup.size(); j++) { - if (chosen.group == lookup[j]->getGroup()) { - abundance = lookup[j]->getAbundance(chosen.bin); - lookup[j]->set(chosen.bin, (abundance + 1), chosen.group); - break; + for(int i=0;igetMultiple() == true) { + sumCalculators[i]->getValues(thisLookup); + outAll << '\t'; + sumCalculators[i]->print(outAll); + } + } + outAll << endl; } - } - - } + + int n = 1; + vector subset; + for (int k = 0; k < (thisLookup.size() - 1); k++) { // pass cdd each set of groups to commpare + for (int l = n; l < thisLookup.size(); l++) { + + outputFileHandle << thisLookup[0]->getLabel() << '\t'; + + subset.clear(); //clear out old pair of sharedrabunds + //add new pair of sharedrabunds + subset.push_back(thisLookup[k]); subset.push_back(thisLookup[l]); + + //sort groups to be alphanumeric + if (thisLookup[k]->getGroup() > thisLookup[l]->getGroup()) { + outputFileHandle << (thisLookup[l]->getGroup() +'\t' + thisLookup[k]->getGroup()) << '\t'; //print out groups + }else{ + outputFileHandle << (thisLookup[k]->getGroup() +'\t' + thisLookup[l]->getGroup()) << '\t'; //print out groups + } + + for(int i=0;igetValues(subset); //saves the calculator outputs + outputFileHandle << '\t'; + sumCalculators[i]->print(outputFileHandle); + } + outputFileHandle << endl; + } + n++; + } + } catch(exception& e) { - cout << "Standard Error: " << e.what() << " has occurred in the SummarySharedCommand class Function getSharedVectors. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + cout << "Standard Error: " << e.what() << " has occurred in the SummarySharedCommand class Function process. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; exit(1); } catch(...) { - cout << "An unknown error has occurred in the SummarySharedCommand class function getSharedVectors. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; + cout << "An unknown error has occurred in the SummarySharedCommand class function process. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n"; exit(1); - } - + } } -//********************************************************************************************************************** +/***********************************************************/