X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=summarysharedcommand.cpp;h=59c12a03dae958f4bb199a6aad94056264c0426a;hp=6147832f70a750e1424939c9928a50db643fb8b8;hb=b206f634aae1b4ce13978d203247fb64757d5482;hpb=f687723a8357916e86a05116978e6869b039ce36 diff --git a/summarysharedcommand.cpp b/summarysharedcommand.cpp index 6147832..59c12a0 100644 --- a/summarysharedcommand.cpp +++ b/summarysharedcommand.cpp @@ -13,18 +13,18 @@ //********************************************************************************************************************** vector SummarySharedCommand::setParameters(){ try { - CommandParameter pshared("shared", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pshared); - CommandParameter plabel("label", "String", "", "", "", "", "",false,false); parameters.push_back(plabel); - CommandParameter psubsample("subsample", "String", "", "", "", "", "",false,false); parameters.push_back(psubsample); - CommandParameter pdistance("distance", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pdistance); - CommandParameter pcalc("calc", "Multiple", "sharedchao-sharedsobs-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan-kstest-whittaker-sharednseqs-ochiai-anderberg-kulczynski-kulczynskicody-lennon-morisitahorn-braycurtis-odum-canberra-structeuclidean-structchord-hellinger-manhattan-structpearson-soergel-spearman-structkulczynski-speciesprofile-structchi2-hamming-gower-memchi2-memchord-memeuclidean-mempearson", "sharedsobs-sharedchao-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan", "", "", "",true,false); parameters.push_back(pcalc); - CommandParameter poutput("output", "Multiple", "lt-square", "lt", "", "", "",false,false); parameters.push_back(poutput); - CommandParameter pall("all", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(pall); - CommandParameter piters("iters", "Number", "", "1000", "", "", "",false,false); parameters.push_back(piters); - CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors); - CommandParameter pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups); - CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir); - CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir); + CommandParameter pshared("shared", "InputTypes", "", "", "none", "none", "none","summary",false,true,true); parameters.push_back(pshared); + CommandParameter plabel("label", "String", "", "", "", "", "","",false,false); parameters.push_back(plabel); + CommandParameter psubsample("subsample", "String", "", "", "", "", "","phylip",false,false); parameters.push_back(psubsample); + CommandParameter pdistance("distance", "Boolean", "", "F", "", "", "","phylip",false,false); parameters.push_back(pdistance); + CommandParameter pcalc("calc", "Multiple", "sharedchao-sharedsobs-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan-kstest-whittaker-sharednseqs-ochiai-anderberg-kulczynski-kulczynskicody-lennon-morisitahorn-braycurtis-odum-canberra-structeuclidean-structchord-hellinger-manhattan-structpearson-soergel-spearman-structkulczynski-speciesprofile-structchi2-hamming-gower-memchi2-memchord-memeuclidean-mempearson-jsd-rjsd", "sharedsobs-sharedchao-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan", "", "", "","",true,false,true); parameters.push_back(pcalc); + CommandParameter poutput("output", "Multiple", "lt-square", "lt", "", "", "","",false,false); parameters.push_back(poutput); + CommandParameter pall("all", "Boolean", "", "F", "", "", "","",false,false); parameters.push_back(pall); + CommandParameter piters("iters", "Number", "", "1000", "", "", "","",false,false); parameters.push_back(piters); + CommandParameter pprocessors("processors", "Number", "", "1", "", "", "","",false,false,true); parameters.push_back(pprocessors); + CommandParameter pgroups("groups", "String", "", "", "", "", "","",false,false); parameters.push_back(pgroups); + 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); } @@ -64,25 +64,20 @@ string SummarySharedCommand::getHelpString(){ } } //********************************************************************************************************************** -string SummarySharedCommand::getOutputFileNameTag(string type, string inputName=""){ - try { - string outputFileName = ""; - map >::iterator it; +string SummarySharedCommand::getOutputPattern(string type) { + try { + string pattern = ""; - //is this a type this command creates - it = outputTypes.find(type); - if (it == outputTypes.end()) { m->mothurOut("[ERROR]: this command doesn't create a " + type + " output file.\n"); } - else { - if (type == "summary") { outputFileName = "shared.summary"; } - else if (type == "phylip") { outputFileName = "dist"; } - else { m->mothurOut("[ERROR]: No definition for type " + type + " output file tag.\n"); m->control_pressed = true; } - } - return outputFileName; - } - catch(exception& e) { - m->errorOut(e, "SummarySharedCommand", "getOutputFileNameTag"); - exit(1); - } + if (type == "summary") { pattern = "[filename],summary-[filename],[tag],summary"; } + else if (type == "phylip") { pattern = "[filename],[calc],[distance],[outputtag],[tag2],dist"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "SummarySharedCommand", "getOutputPattern"); + exit(1); + } } //********************************************************************************************************************** SummarySharedCommand::SummarySharedCommand(){ @@ -189,13 +184,12 @@ SummarySharedCommand::SummarySharedCommand(string option) { string temp = validParameter.validFile(parameters, "all", false); if (temp == "not found") { temp = "false"; } all = m->isTrue(temp); - temp = validParameter.validFile(parameters, "distance", false); if (temp == "not found") { temp = "false"; } - createPhylip = m->isTrue(temp); - temp = validParameter.validFile(parameters, "iters", false); if (temp == "not found") { temp = "1000"; } m->mothurConvert(temp, iters); - output = validParameter.validFile(parameters, "output", false); if(output == "not found"){ output = "lt"; } + output = validParameter.validFile(parameters, "output", false); + if(output == "not found"){ output = "lt"; } + else { createPhylip = true; } if ((output != "lt") && (output != "square")) { m->mothurOut(output + " is not a valid output form. Options are lt and square. I will use lt."); m->mothurOutEndLine(); output = "lt"; } temp = validParameter.validFile(parameters, "subsample", false); if (temp == "not found") { temp = "F"; } @@ -205,7 +199,11 @@ SummarySharedCommand::SummarySharedCommand(string option) { else { subsample = false; } } - if (subsample == false) { iters = 1; } + if (subsample == false) { iters = 0; } + + temp = validParameter.validFile(parameters, "distance", false); if (temp == "not found") { temp = "false"; } + createPhylip = m->isTrue(temp); + if (subsample) { createPhylip = true; } temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); } m->setProcessors(temp); @@ -296,6 +294,10 @@ SummarySharedCommand::SummarySharedCommand(string option) { sumCalculators.push_back(new MemEuclidean()); }else if (Estimators[i] == "mempearson") { sumCalculators.push_back(new MemPearson()); + }else if (Estimators[i] == "jsd") { + sumCalculators.push_back(new JSD()); + }else if (Estimators[i] == "rjsd") { + sumCalculators.push_back(new RJSD()); } } } @@ -317,7 +319,9 @@ int SummarySharedCommand::execute(){ if (abort == true) { if (calledHelp) { return 0; } return 2; } ofstream outputFileHandle, outAll; - string outputFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + getOutputFileNameTag("summary"); + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile)); + string outputFileName = getOutputFileName("summary",variables); //if the users entered no valid calculators don't execute command if (sumCalculators.size() == 0) { return 0; } @@ -348,7 +352,8 @@ int SummarySharedCommand::execute(){ outputFileHandle.close(); //create file and put column headers for multiple groups file - string outAllFileName = ((m->getRootName(sharedfile)) + "multiple." + getOutputFileNameTag("summary")); + variables["[tag]"]= "multiple"; + string outAllFileName = getOutputFileName("summary",variables); if (mult == true) { m->openOutputFile(outAllFileName, outAll); outputNames.push_back(outAllFileName); @@ -703,7 +708,7 @@ int SummarySharedCommand::process(vector thisLookup, string vector pDataArray; DWORD dwThreadIdArray[processors-1]; - HANDLE hThreadArray[processors-1]; + HANDLE hThreadArray[processors-1]; //Create processor worker threads. for( int i=1; i thisLookup, string temp->setGroup(thisLookup[k]->getGroup()); newLookup.push_back(temp); } + //for each bin for (int k = 0; k < thisLookup[0]->getNumBins(); k++) { @@ -742,6 +748,9 @@ int SummarySharedCommand::process(vector thisLookup, string //Close all thread handles and free memory allocations. for(int i=0; i < pDataArray.size(); i++){ + if (pDataArray[i]->count != (pDataArray[i]->end-pDataArray[i]->start)) { + m->mothurOut("[ERROR]: process " + toString(i) + " only processed " + toString(pDataArray[i]->count) + " of " + toString(pDataArray[i]->end-pDataArray[i]->start) + " groups assigned to it, quitting. \n"); m->control_pressed = true; + } m->appendFiles((sumFileName + toString(processIDS[i]) + ".temp"), sumFileName); m->mothurRemove((sumFileName + toString(processIDS[i]) + ".temp")); @@ -786,7 +795,13 @@ int SummarySharedCommand::process(vector thisLookup, string matrix[column][row] = dist; } - string distFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + sumCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + "." + output + "." + getOutputFileNameTag("phylip"); + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile)); + variables["[calc]"] = sumCalculators[i]->getName(); + variables["[distance]"] = thisLookup[0]->getLabel(); + variables["[outputtag]"] = output; + variables["[tag2]"] = ""; + string distFileName = getOutputFileName("phylip",variables); outputNames.push_back(distFileName); outputTypes["phylip"].push_back(distFileName); ofstream outDist; m->openOutputFile(distFileName, outDist); @@ -801,60 +816,12 @@ int SummarySharedCommand::process(vector thisLookup, string for (int i = 0; i < calcDists.size(); i++) { calcDists[i].clear(); } } - if (iters != 1) { + if (iters != 0) { //we need to find the average distance and standard deviation for each groups distance - - vector< vector > calcAverages; calcAverages.resize(sumCalculators.size()); - for (int i = 0; i < calcAverages.size(); i++) { //initialize sums to zero. - calcAverages[i].resize(calcDistsTotals[0][i].size()); - - for (int j = 0; j < calcAverages[i].size(); j++) { - calcAverages[i][j].seq1 = calcDists[i][j].seq1; - calcAverages[i][j].seq2 = calcDists[i][j].seq2; - calcAverages[i][j].dist = 0.0; - } - } - - for (int thisIter = 0; thisIter < iters; thisIter++) { //sum all groups dists for each calculator - for (int i = 0; i < calcAverages.size(); i++) { //initialize sums to zero. - for (int j = 0; j < calcAverages[i].size(); j++) { - calcAverages[i][j].dist += calcDistsTotals[thisIter][i][j].dist; - } - } - } - - for (int i = 0; i < calcAverages.size(); i++) { //finds average. - for (int j = 0; j < calcAverages[i].size(); j++) { - calcAverages[i][j].dist /= (float) iters; - } - } + vector< vector > calcAverages = m->getAverages(calcDistsTotals); //find standard deviation - vector< vector > stdDev; stdDev.resize(sumCalculators.size()); - for (int i = 0; i < stdDev.size(); i++) { //initialize sums to zero. - stdDev[i].resize(calcDistsTotals[0][i].size()); - - for (int j = 0; j < stdDev[i].size(); j++) { - stdDev[i][j].seq1 = calcDists[i][j].seq1; - stdDev[i][j].seq2 = calcDists[i][j].seq2; - stdDev[i][j].dist = 0.0; - } - } - - for (int thisIter = 0; thisIter < iters; thisIter++) { //compute the difference of each dist from the mean, and square the result of each - for (int i = 0; i < stdDev.size(); i++) { - for (int j = 0; j < stdDev[i].size(); j++) { - stdDev[i][j].dist += ((calcDistsTotals[thisIter][i][j].dist - calcAverages[i][j].dist) * (calcDistsTotals[thisIter][i][j].dist - calcAverages[i][j].dist)); - } - } - } - - for (int i = 0; i < stdDev.size(); i++) { //finds average. - for (int j = 0; j < stdDev[i].size(); j++) { - stdDev[i][j].dist /= (float) iters; - stdDev[i][j].dist = sqrt(stdDev[i][j].dist); - } - } + vector< vector > stdDev = m->getStandardDeviation(calcDistsTotals, calcAverages); //print results for (int i = 0; i < calcDists.size(); i++) { @@ -879,7 +846,13 @@ int SummarySharedCommand::process(vector thisLookup, string stdmatrix[column][row] = stdDist; } - string distFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + sumCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + "." + output + ".ave." + getOutputFileNameTag("phylip"); + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile)); + variables["[calc]"] = sumCalculators[i]->getName(); + variables["[distance]"] = thisLookup[0]->getLabel(); + variables["[outputtag]"] = output; + variables["[tag2]"] = "ave"; + string distFileName = getOutputFileName("phylip",variables); outputNames.push_back(distFileName); outputTypes["phylip"].push_back(distFileName); ofstream outAve; m->openOutputFile(distFileName, outAve); @@ -889,7 +862,8 @@ int SummarySharedCommand::process(vector thisLookup, string outAve.close(); - distFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + sumCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + "." + output + ".std." + getOutputFileNameTag("phylip"); + variables["[tag2]"] = "std"; + distFileName = getOutputFileName("phylip",variables); outputNames.push_back(distFileName); outputTypes["phylip"].push_back(distFileName); ofstream outSTD; m->openOutputFile(distFileName, outSTD);