X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=clustercommand.cpp;h=528f71c11c24f7a47133bef2d60a6906249abe92;hb=5cbb9328400eabc6689da26738aceee394666dfa;hp=221646700817bdddf61c1f2b968328a1af590ac1;hpb=1d898dc6edaf9e9f287fab53bf1f21fb29757a17;p=mothur.git diff --git a/clustercommand.cpp b/clustercommand.cpp index 2216467..528f71c 100644 --- a/clustercommand.cpp +++ b/clustercommand.cpp @@ -11,6 +11,7 @@ #include "readphylip.h" #include "readcolumn.h" #include "readmatrix.hpp" +#include "clusterdoturcommand.h" //********************************************************************************************************************** vector ClusterCommand::setParameters(){ @@ -20,11 +21,11 @@ vector ClusterCommand::setParameters(){ CommandParameter pcolumn("column", "InputTypes", "", "", "PhylipColumn", "PhylipColumn", "ColumnName",false,false); parameters.push_back(pcolumn); CommandParameter pcutoff("cutoff", "Number", "", "10", "", "", "",false,false); parameters.push_back(pcutoff); CommandParameter pprecision("precision", "Number", "", "100", "", "", "",false,false); parameters.push_back(pprecision); - CommandParameter pmethod("method", "Multiple", "furthest-nearest-average-weighted", "furthest", "", "", "",false,false); parameters.push_back(pmethod); + CommandParameter pmethod("method", "Multiple", "furthest-nearest-average-weighted", "average", "", "", "",false,false); parameters.push_back(pmethod); CommandParameter pshowabund("showabund", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pshowabund); CommandParameter ptiming("timing", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(ptiming); CommandParameter psim("sim", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(psim); - CommandParameter phard("hard", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(phard); + CommandParameter phard("hard", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(phard); CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir); CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir); @@ -44,7 +45,7 @@ string ClusterCommand::getHelpString(){ helpString += "The cluster command parameter options are phylip, column, name, method, cuttoff, hard, precision, sim, showabund and timing. Phylip or column and name are required, unless you have a valid current file.\n"; helpString += "The cluster command should be in the following format: \n"; helpString += "cluster(method=yourMethod, cutoff=yourCutoff, precision=yourPrecision) \n"; - helpString += "The acceptable cluster methods are furthest, nearest, average and weighted. If no method is provided then furthest is assumed.\n"; + helpString += "The acceptable cluster methods are furthest, nearest, average and weighted. If no method is provided then average is assumed.\n"; return helpString; } catch(exception& e) { @@ -75,6 +76,7 @@ ClusterCommand::ClusterCommand(string option) { //allow user to run help if(option == "help") { help(); abort = true; calledHelp = true; } + else if(option == "citation") { citation(); abort = true; calledHelp = true;} else { vector myArray = setParameters(); @@ -134,25 +136,26 @@ ClusterCommand::ClusterCommand(string option) { phylipfile = validParameter.validFile(parameters, "phylip", true); if (phylipfile == "not open") { phylipfile = ""; abort = true; } else if (phylipfile == "not found") { phylipfile = ""; } - else { distfile = phylipfile; format = "phylip"; } + else { distfile = phylipfile; format = "phylip"; m->setPhylipFile(phylipfile); } columnfile = validParameter.validFile(parameters, "column", true); if (columnfile == "not open") { columnfile = ""; abort = true; } else if (columnfile == "not found") { columnfile = ""; } - else { distfile = columnfile; format = "column"; } + else { distfile = columnfile; format = "column"; m->setColumnFile(columnfile); } namefile = validParameter.validFile(parameters, "name", true); if (namefile == "not open") { abort = true; } else if (namefile == "not found") { namefile = ""; } + else { m->setNameFile(namefile); } if ((phylipfile == "") && (columnfile == "")) { //is there are current file available for either of these? //give priority to column, then phylip columnfile = m->getColumnFile(); - if (columnfile != "") { m->mothurOut("Using " + columnfile + " as input file for the column parameter."); m->mothurOutEndLine(); } + if (columnfile != "") { distfile = columnfile; format = "column"; m->mothurOut("Using " + columnfile + " as input file for the column parameter."); m->mothurOutEndLine(); } else { phylipfile = m->getPhylipFile(); - if (phylipfile != "") { m->mothurOut("Using " + phylipfile + " as input file for the phylip parameter."); m->mothurOutEndLine(); } + if (phylipfile != "") { distfile = phylipfile; format = "phylip"; m->mothurOut("Using " + phylipfile + " as input file for the phylip parameter."); m->mothurOutEndLine(); } else { m->mothurOut("No valid current files. You must provide a phylip or column file before you can use the cluster command."); m->mothurOutEndLine(); abort = true; @@ -180,9 +183,9 @@ ClusterCommand::ClusterCommand(string option) { if (temp == "not found") { temp = "100"; } //saves precision legnth for formatting below length = temp.length(); - convert(temp, precision); + m->mothurConvert(temp, precision); - temp = validParameter.validFile(parameters, "hard", false); if (temp == "not found") { temp = "F"; } + temp = validParameter.validFile(parameters, "hard", false); if (temp == "not found") { temp = "T"; } hard = m->isTrue(temp); temp = validParameter.validFile(parameters, "sim", false); if (temp == "not found") { temp = "F"; } @@ -190,11 +193,11 @@ ClusterCommand::ClusterCommand(string option) { temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "10"; } - convert(temp, cutoff); + m->mothurConvert(temp, cutoff); cutoff += (5 / (precision * 10.0)); method = validParameter.validFile(parameters, "method", false); - if (method == "not found") { method = "furthest"; } + if (method == "not found") { method = "average"; } if ((method == "furthest") || (method == "nearest") || (method == "average") || (method == "weighted")) { } else { m->mothurOut("Not a valid clustering method. Valid clustering algorithms are furthest, nearest, average, and weighted."); m->mothurOutEndLine(); abort = true; } @@ -205,7 +208,7 @@ ClusterCommand::ClusterCommand(string option) { timing = validParameter.validFile(parameters, "timing", false); if (timing == "not found") { timing = "F"; } - } + } } catch(exception& e) { m->errorOut(e, "ClusterCommand", "ClusterCommand"); @@ -221,6 +224,34 @@ int ClusterCommand::execute(){ if (abort == true) { if (calledHelp) { return 0; } return 2; } + //phylip file given and cutoff not given - use cluster.classic because it uses less memory and is faster + if ((format == "phylip") && (cutoff > 10.0)) { + m->mothurOutEndLine(); m->mothurOut("You are using a phylip file and no cutoff. I will run cluster.classic to save memory and time."); m->mothurOutEndLine(); + + //run unique.seqs for deconvolute results + string inputString = "phylip=" + distfile; + if (namefile != "") { inputString += ", name=" + namefile; } + inputString += ", precision=" + toString(precision); + inputString += ", method=" + method; + if (hard) { inputString += ", hard=T"; } + else { inputString += ", hard=F"; } + if (sim) { inputString += ", sim=T"; } + else { inputString += ", sim=F"; } + + + m->mothurOutEndLine(); + m->mothurOut("/------------------------------------------------------------/"); m->mothurOutEndLine(); + m->mothurOut("Running command: cluster.classic(" + inputString + ")"); m->mothurOutEndLine(); + + Command* clusterClassicCommand = new ClusterDoturCommand(inputString); + clusterClassicCommand->execute(); + delete clusterClassicCommand; + + m->mothurOut("/------------------------------------------------------------/"); m->mothurOutEndLine(); + + return 0; + } + ReadMatrix* read; if (format == "column") { read = new ReadColumnMatrix(columnfile, sim); } //sim indicates whether its a similarity matrix else if (format == "phylip") { read = new ReadPhylipMatrix(phylipfile, sim); } @@ -242,7 +273,7 @@ int ClusterCommand::execute(){ if (m->control_pressed) { //clean up delete list; delete matrix; delete rabund; sabundFile.close();rabundFile.close();listFile.close(); - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear(); return 0; } @@ -281,7 +312,7 @@ int ClusterCommand::execute(){ if (m->control_pressed) { //clean up delete list; delete matrix; delete rabund; delete cluster; sabundFile.close();rabundFile.close();listFile.close(); - for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } outputTypes.clear(); + for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } outputTypes.clear(); return 0; } @@ -324,7 +355,7 @@ int ClusterCommand::execute(){ cout.flush(); print_start = false; } - + if(previousDist <= 0.0000){ printData("unique"); }