X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=matrixoutputcommand.cpp;h=3844e18c1e687ef47fae1dda09d10e61973ab798;hb=4b54ce99af7db8019ea907cd7c2edf789369ada9;hp=ccd4f538ab03f7ea86c59b2d3381d2de1a2de52f;hpb=9b53f130ac9af5e95444ce2e817fce25ed19ff03;p=mothur.git diff --git a/matrixoutputcommand.cpp b/matrixoutputcommand.cpp index ccd4f53..3844e18 100644 --- a/matrixoutputcommand.cpp +++ b/matrixoutputcommand.cpp @@ -13,17 +13,17 @@ //********************************************************************************************************************** vector MatrixOutputCommand::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 pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups); - CommandParameter pcalc("calc", "Multiple", "sharedsobs-sharedchao-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan-kstest-sharednseqs-ochiai-anderberg-kulczynski-kulczynskicody-lennon-morisitahorn-braycurtis-whittaker-odum-canberra-structeuclidean-structchord-hellinger-manhattan-structpearson-soergel-spearman-structkulczynski-speciesprofile-hamming-structchi2-gower-memchi2-memchord-memeuclidean-mempearson", "jclass-thetayc", "", "", "",true,false); parameters.push_back(pcalc); - CommandParameter poutput("output", "Multiple", "lt-square", "lt", "", "", "",false,false); parameters.push_back(poutput); - CommandParameter pmode("mode", "Multiple", "average-median", "average", "", "", "",false,false); parameters.push_back(pmode); - CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors); - CommandParameter piters("iters", "Number", "", "1000", "", "", "",false,false); parameters.push_back(piters); - 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","phylip",false,true,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 pgroups("groups", "String", "", "", "", "", "","",false,false); parameters.push_back(pgroups); + CommandParameter pcalc("calc", "Multiple", "sharedsobs-sharedchao-sharedace-jabund-sorabund-jclass-sorclass-jest-sorest-thetayc-thetan-kstest-sharednseqs-ochiai-anderberg-kulczynski-kulczynskicody-lennon-morisitahorn-braycurtis-whittaker-odum-canberra-structeuclidean-structchord-hellinger-manhattan-structpearson-soergel-spearman-structkulczynski-speciesprofile-hamming-structchi2-gower-memchi2-memchord-memeuclidean-mempearson", "jclass-thetayc", "", "", "","",true,false,true); parameters.push_back(pcalc); + CommandParameter poutput("output", "Multiple", "lt-square", "lt", "", "", "","",false,false); parameters.push_back(poutput); + CommandParameter pmode("mode", "Multiple", "average-median", "average", "", "", "","",false,false); parameters.push_back(pmode); + CommandParameter pprocessors("processors", "Number", "", "1", "", "", "","",false,false,true); parameters.push_back(pprocessors); + CommandParameter piters("iters", "Number", "", "1000", "", "", "","",false,false); parameters.push_back(piters); + 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); } @@ -61,24 +61,19 @@ string MatrixOutputCommand::getHelpString(){ } } //********************************************************************************************************************** -string MatrixOutputCommand::getOutputFileNameTag(string type, string inputName=""){ - try { - string outputFileName = ""; - map >::iterator it; +string MatrixOutputCommand::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 == "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, "MatrixOutputCommand", "getOutputFileNameTag"); - exit(1); - } + if (type == "phylip") { pattern = "[filename],[calc],[distance],[outputtag],dist-[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, "MatrixOutputCommand", "getOutputPattern"); + exit(1); + } } //********************************************************************************************************************** MatrixOutputCommand::MatrixOutputCommand(){ @@ -486,8 +481,12 @@ int MatrixOutputCommand::process(vector thisLookup){ try { vector< vector< vector > > calcDistsTotals; //each iter, one for each calc, then each groupCombos dists. this will be used to make .dist files vector< vector > calcDists; calcDists.resize(matrixCalculators.size()); - + for (int thisIter = 0; thisIter < iters+1; thisIter++) { + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile)); + variables["[distance]"] = thisLookup[0]->getLabel(); + variables["[tag2]"] = ""; vector thisItersLookup = thisLookup; @@ -673,7 +672,9 @@ int MatrixOutputCommand::process(vector thisLookup){ matrix[column][row] = dist; } - string distFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + matrixCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + "." + output + "." + getOutputFileNameTag("phylip"); + variables["[outputtag]"] = output; + variables["[calc]"] = matrixCalculators[i]->getName(); + string distFileName = getOutputFileName("phylip",variables); outputNames.push_back(distFileName); outputTypes["phylip"].push_back(distFileName); ofstream outDist; @@ -690,70 +691,10 @@ int MatrixOutputCommand::process(vector thisLookup){ if (iters != 0) { //we need to find the average distance and standard deviation for each groups distance + vector< vector > calcAverages = m->getAverages(calcDistsTotals, mode); - vector< vector > calcAverages; calcAverages.resize(matrixCalculators.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 = calcDistsTotals[0][i][j].seq1; - calcAverages[i][j].seq2 = calcDistsTotals[0][i][j].seq2; - calcAverages[i][j].dist = 0.0; - } - } - if (mode == "average") { - 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; - if (m->debug) { m->mothurOut("[DEBUG]: Totaling for average calc: iter = " + toString(thisIter) + ", " + thisLookup[calcDistsTotals[thisIter][i][j].seq1]->getGroup() + " - " + thisLookup[calcDistsTotals[thisIter][i][j].seq2]->getGroup() + " distance = " + toString(calcDistsTotals[thisIter][i][j].dist) + ". New total = " + toString(calcAverages[i][j].dist) + ".\n"); } - } - } - } - - 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; - } - } - }else { //find median - for (int i = 0; i < calcAverages.size(); i++) { //for each calc - for (int j = 0; j < calcAverages[i].size(); j++) { //for each comparison - vector dists; - for (int thisIter = 0; thisIter < iters; thisIter++) { //for each subsample - dists.push_back(calcDistsTotals[thisIter][i][j].dist); - } - sort(dists.begin(), dists.end()); - calcAverages[i][j].dist = dists[(iters/2)]; - } - } - } //find standard deviation - vector< vector > stdDev; stdDev.resize(matrixCalculators.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 = calcDistsTotals[0][i][j].seq1; - stdDev[i][j].seq2 = calcDistsTotals[0][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++) { @@ -777,8 +718,14 @@ int MatrixOutputCommand::process(vector thisLookup){ stdmatrix[row][column] = stdDist; stdmatrix[column][row] = stdDist; } - - string distFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + matrixCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + "." + output + ".ave." + getOutputFileNameTag("phylip"); + + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile)); + variables["[distance]"] = thisLookup[0]->getLabel(); + variables["[outputtag]"] = output; + variables["[tag2]"] = "ave"; + variables["[calc]"] = matrixCalculators[i]->getName(); + string distFileName = getOutputFileName("phylip",variables); outputNames.push_back(distFileName); outputTypes["phylip"].push_back(distFileName); ofstream outAve; m->openOutputFile(distFileName, outAve); @@ -788,7 +735,8 @@ int MatrixOutputCommand::process(vector thisLookup){ outAve.close(); - distFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + matrixCalculators[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);