X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=shhhercommand.cpp;h=537821134edebcf9dcad70f8c1f38be0da39b121;hb=731d19a6eaf81768ed35e4be25fbac5686a2a7a0;hp=469bd4d455a1d658a4ef3fe7ce69af86403a9e02;hpb=ef2b09798e381dbb9b9836e09c72f8af57825830;p=mothur.git diff --git a/shhhercommand.cpp b/shhhercommand.cpp index 469bd4d..5378211 100644 --- a/shhhercommand.cpp +++ b/shhhercommand.cpp @@ -18,20 +18,12 @@ #include "sparsematrix.hpp" #include -//********************************************************************************************************************** - -#define NUMBINS 1000 -#define HOMOPS 10 -#define MIN_COUNT 0.1 -#define MIN_WEIGHT 0.1 -#define MIN_TAU 0.0001 -#define MIN_ITER 10 //********************************************************************************************************************** vector ShhherCommand::setParameters(){ try { CommandParameter pflow("flow", "InputTypes", "", "", "none", "fileflow", "none",false,false); parameters.push_back(pflow); CommandParameter pfile("file", "InputTypes", "", "", "none", "fileflow", "none",false,false); parameters.push_back(pfile); - CommandParameter plookup("lookup", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(plookup); + CommandParameter plookup("lookup", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(plookup); CommandParameter pcutoff("cutoff", "Number", "", "0.01", "", "", "",false,false); parameters.push_back(pcutoff); CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors); CommandParameter pmaxiter("maxiter", "Number", "", "1000", "", "", "",false,false); parameters.push_back(pmaxiter); @@ -54,7 +46,7 @@ vector ShhherCommand::setParameters(){ string ShhherCommand::getHelpString(){ try { string helpString = ""; - helpString += "The shhh.seqs command reads a file containing flowgrams and creates a file of corrected sequences.\n"; + helpString += "The shhh.flows command reads a file containing flowgrams and creates a file of corrected sequences.\n"; return helpString; } catch(exception& e) { @@ -168,11 +160,11 @@ ShhherCommand::ShhherCommand(string option) { ofstream temp; //flow.files = 9 character offset - compositeFASTAFileName = flowFilesFileName.substr(0, flowFilesFileName.length()-9) + "shhh.fasta"; + compositeFASTAFileName = flowFilesFileName.substr(0, flowFilesFileName.length()-10) + "shhh.fasta"; m->openOutputFile(compositeFASTAFileName, temp); temp.close(); - compositeNamesFileName = flowFilesFileName.substr(0, flowFilesFileName.length()-9) + "shhh.names"; + compositeNamesFileName = flowFilesFileName.substr(0, flowFilesFileName.length()-10) + "shhh.names"; m->openOutputFile(compositeNamesFileName, temp); temp.close(); } @@ -259,19 +251,19 @@ ShhherCommand::ShhherCommand(string option) { temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); } m->setProcessors(temp); - convert(temp, processors); + m->mothurConvert(temp, processors); temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found"){ temp = "0.01"; } - convert(temp, cutoff); + m->mothurConvert(temp, cutoff); temp = validParameter.validFile(parameters, "mindelta", false); if (temp == "not found"){ temp = "0.000001"; } - convert(temp, minDelta); + m->mothurConvert(temp, minDelta); temp = validParameter.validFile(parameters, "maxiter", false); if (temp == "not found"){ temp = "1000"; } - convert(temp, maxIters); + m->mothurConvert(temp, maxIters); temp = validParameter.validFile(parameters, "sigma", false);if (temp == "not found") { temp = "60"; } - convert(temp, sigma); + m->mothurConvert(temp, sigma); flowOrder = validParameter.validFile(parameters, "order", false); if (flowOrder == "not found"){ flowOrder = "TACG"; } @@ -310,8 +302,8 @@ int ShhherCommand::execute(){ processors = ncpus; m->mothurOut("\nGetting preliminary data...\n"); - getSingleLookUp(); - getJointLookUp(); + getSingleLookUp(); if (m->control_pressed) { return 0; } + getJointLookUp(); if (m->control_pressed) { return 0; } vector flowFileVector; if(flowFilesFileName != "not found"){ @@ -320,7 +312,7 @@ int ShhherCommand::execute(){ ifstream flowFilesFile; m->openInputFile(flowFilesFileName, flowFilesFile); while(flowFilesFile){ - flowFilesFile >> fName; + fName = m->getline(flowFilesFile); flowFileVector.push_back(fName); m->gobble(flowFilesFile); } @@ -335,17 +327,24 @@ int ShhherCommand::execute(){ } for(int i=0;icontrol_pressed) { break; } + double begClock = clock(); - unsigned long int begTime = time(NULL); + unsigned long long begTime = time(NULL); flowFileName = flowFileVector[i]; m->mothurOut("\n>>>>>\tProcessing " + flowFileName + " (file " + toString(i+1) + " of " + toString(numFiles) + ")\t<<<<<\n"); m->mothurOut("Reading flowgrams...\n"); getFlowData(); + + if (m->control_pressed) { break; } m->mothurOut("Identifying unique flowgrams...\n"); getUniques(); + + if (m->control_pressed) { break; } m->mothurOut("Calculating distances between flowgrams...\n"); char fileName[1024]; @@ -368,27 +367,37 @@ int ShhherCommand::execute(){ string distFileName = flowDistMPI(0, int(sqrt(1.0/float(ncpus)) * numUniques)); + if (m->control_pressed) { break; } + int done; for(int i=1;iappendFiles((distFileName + ".temp." + toString(i)), distFileName); - remove((distFileName + ".temp." + toString(i)).c_str()); + m->mothurRemove((distFileName + ".temp." + toString(i))); } string namesFileName = createNamesFile(); + if (m->control_pressed) { break; } + m->mothurOut("\nClustering flowgrams...\n"); string listFileName = cluster(distFileName, namesFileName); - + + if (m->control_pressed) { break; } + getOTUData(listFileName); - remove(distFileName.c_str()); - remove(namesFileName.c_str()); - remove(listFileName.c_str()); + m->mothurRemove(distFileName); + m->mothurRemove(namesFileName); + m->mothurRemove(listFileName); + + if (m->control_pressed) { break; } initPyroCluster(); + if (m->control_pressed) { break; } + for(int i=1;icontrol_pressed) { break; } double maxDelta = 0; int iter = 0; @@ -406,10 +416,12 @@ int ShhherCommand::execute(){ m->mothurOut("iter\tmaxDelta\tnLL\t\tcycletime\n"); while((maxIters == 0 && maxDelta > minDelta) || iter < MIN_ITER || (maxDelta > minDelta && iter < maxIters)){ - + double cycClock = clock(); - unsigned long int cycTime = time(NULL); + unsigned long long cycTime = time(NULL); fill(); + + if (m->control_pressed) { break; } int total = singleTau.size(); for(int i=1;icontrol_pressed) { break; } + double nLL = getLikelihood(); if (m->control_pressed) { break; } + checkCentroids(); if (m->control_pressed) { break; } for(int i=1;icontrol_pressed) { break; } + m->mothurOut("\nFinalizing...\n"); fill(); + + if (m->control_pressed) { break; } + setOTUs(); vector otuCounts(numOTUs, 0); for(int i=0;icontrol_pressed) { break; } + + writeQualities(otuCounts); if (m->control_pressed) { break; } + writeSequences(otuCounts); if (m->control_pressed) { break; } + writeNames(otuCounts); if (m->control_pressed) { break; } + writeClusters(otuCounts); if (m->control_pressed) { break; } + writeGroups(); if (m->control_pressed) { break; } m->mothurOut("Total time to process " + toString(flowFileName) + ":\t" + toString(time(NULL) - begTime) + '\t' + toString((clock() - begClock)/(double)CLOCKS_PER_SEC) + '\n'); @@ -549,6 +568,9 @@ int ShhherCommand::execute(){ MPI_Recv(&numFiles, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status); for(int i=0;icontrol_pressed) { break; } + //Now into the pyrodist part bool live = 1; @@ -578,7 +600,9 @@ int ShhherCommand::execute(){ int flowDistEnd = int(sqrt(float(pid+1)/float(ncpus)) * numUniques); string distanceStringChild = flowDistMPI(flowDistStart, flowDistEnd); - + + if (m->control_pressed) { break; } + int done = 1; MPI_Send(&done, 1, MPI_INT, 0, tag, MPI_COMM_WORLD); @@ -608,6 +632,8 @@ int ShhherCommand::execute(){ while(live){ + if (m->control_pressed) { break; } + MPI_Recv(&total, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status); singleTau.assign(total, 0.0000); seqNumber.assign(total, 0); @@ -643,7 +669,10 @@ int ShhherCommand::execute(){ MPI_Recv(&live, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status); } } - } + } + + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } + MPI_Barrier(MPI_COMM_WORLD); @@ -680,6 +709,9 @@ string ShhherCommand::flowDistMPI(int startSeq, int stopSeq){ double begClock = clock(); for(int i=startSeq;icontrol_pressed) { break; } + for(int j=0;jmothurOut(toString(stopSeq) + '\t' + toString(time(NULL) - begTime) + '\t' + toString((clock()-begClock)/CLOCKS_PER_SEC) + '\n'); - string fDistFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".shhh.dist"; if(pid != 0){ fDistFileName += ".temp." + toString(pid); } + + if (m->control_pressed) { return fDistFileName; } + + m->mothurOut(toString(stopSeq) + '\t' + toString(time(NULL) - begTime) + '\t' + toString((clock()-begClock)/CLOCKS_PER_SEC) + '\n'); ofstream distFile(fDistFileName.c_str()); distFile << outStream.str(); @@ -719,9 +753,10 @@ int ShhherCommand::execute(){ try { if (abort == true) { return 0; } - getSingleLookUp(); - getJointLookUp(); + getSingleLookUp(); if (m->control_pressed) { return 0; } + getJointLookUp(); if (m->control_pressed) { return 0; } + vector flowFileVector; if(flowFilesFileName != "not found"){ string fName; @@ -729,7 +764,7 @@ int ShhherCommand::execute(){ ifstream flowFilesFile; m->openInputFile(flowFilesFileName, flowFilesFile); while(flowFilesFile){ - flowFilesFile >> fName; + fName = m->getline(flowFilesFile); flowFileVector.push_back(fName); m->gobble(flowFilesFile); } @@ -741,76 +776,111 @@ int ShhherCommand::execute(){ for(int i=0;icontrol_pressed) { break; } + flowFileName = flowFileVector[i]; m->mothurOut("\n>>>>>\tProcessing " + flowFileName + " (file " + toString(i+1) + " of " + toString(numFiles) + ")\t<<<<<\n"); m->mothurOut("Reading flowgrams...\n"); getFlowData(); + if (m->control_pressed) { break; } + m->mothurOut("Identifying unique flowgrams...\n"); getUniques(); + if (m->control_pressed) { break; } m->mothurOut("Calculating distances between flowgrams...\n"); string distFileName = createDistFile(processors); string namesFileName = createNamesFile(); - + + if (m->control_pressed) { break; } + m->mothurOut("\nClustering flowgrams...\n"); string listFileName = cluster(distFileName, namesFileName); + + if (m->control_pressed) { break; } + getOTUData(listFileName); - remove(distFileName.c_str()); - remove(namesFileName.c_str()); - remove(listFileName.c_str()); + if (m->control_pressed) { break; } + + m->mothurRemove(distFileName); + m->mothurRemove(namesFileName); + m->mothurRemove(listFileName); initPyroCluster(); + if (m->control_pressed) { break; } + double maxDelta = 0; int iter = 0; double begClock = clock(); - unsigned long int begTime = time(NULL); + unsigned long long begTime = time(NULL); m->mothurOut("\nDenoising flowgrams...\n"); m->mothurOut("iter\tmaxDelta\tnLL\t\tcycletime\n"); while((maxIters == 0 && maxDelta > minDelta) || iter < MIN_ITER || (maxDelta > minDelta && iter < maxIters)){ + if (m->control_pressed) { break; } + double cycClock = clock(); - unsigned long int cycTime = time(NULL); + unsigned long long cycTime = time(NULL); fill(); + if (m->control_pressed) { break; } + calcCentroids(); - maxDelta = getNewWeights(); - double nLL = getLikelihood(); + if (m->control_pressed) { break; } + + maxDelta = getNewWeights(); if (m->control_pressed) { break; } + double nLL = getLikelihood(); if (m->control_pressed) { break; } checkCentroids(); + if (m->control_pressed) { break; } + calcNewDistances(); - + + if (m->control_pressed) { break; } + iter++; m->mothurOut(toString(iter) + '\t' + toString(maxDelta) + '\t' + toString(nLL) + '\t' + toString(time(NULL) - cycTime) + '\t' + toString((clock() - cycClock)/(double)CLOCKS_PER_SEC) + '\n'); } + if (m->control_pressed) { break; } + m->mothurOut("\nFinalizing...\n"); fill(); + + if (m->control_pressed) { break; } + setOTUs(); + if (m->control_pressed) { break; } + vector otuCounts(numOTUs, 0); for(int i=0;icontrol_pressed) { break; } + writeQualities(otuCounts); if (m->control_pressed) { break; } + writeSequences(otuCounts); if (m->control_pressed) { break; } + writeNames(otuCounts); if (m->control_pressed) { break; } + writeClusters(otuCounts); if (m->control_pressed) { break; } + writeGroups(); if (m->control_pressed) { break; } m->mothurOut("Total time to process " + flowFileName + ":\t" + toString(time(NULL) - begTime) + '\t' + toString((clock() - begClock)/(double)CLOCKS_PER_SEC) + '\n'); } + if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } + + if(compositeFASTAFileName != ""){ outputNames.push_back(compositeFASTAFileName); outputNames.push_back(compositeNamesFileName); @@ -850,6 +920,9 @@ void ShhherCommand::getFlowData(){ flowFile >> numFlowCells; int index = 0;//pcluster while(!flowFile.eof()){ + + if (m->control_pressed) { break; } + flowFile >> seqName >> currentNumFlowCells; lengths.push_back(currentNumFlowCells); @@ -869,6 +942,9 @@ void ShhherCommand::getFlowData(){ numSeqs = seqNameVector.size(); for(int i=0;icontrol_pressed) { break; } + int iNumFlowCells = i * numFlowCells; for(int j=lengths[i];jopenInputFile(lookupFileName, lookUpFile); for(int i=0;icontrol_pressed) { break; } + float logFracFreq; lookUpFile >> logFracFreq; @@ -919,6 +998,9 @@ void ShhherCommand::getJointLookUp(){ jointLookUp.resize(NUMBINS * NUMBINS, 0); for(int i=0;icontrol_pressed) { break; } + for(int j=0;jcontrol_pressed) { break; } + float negLogProb = singleLookUp[i * NUMBINS + intIntensity]; if(negLogProb < minNegLogProb) { minNegLogProb = negLogProb; } } @@ -975,6 +1060,9 @@ void ShhherCommand::getUniques(){ vector uniqueFlowDataIntI(numFlowCells * numSeqs, -1); for(int i=0;icontrol_pressed) { break; } + int index = 0; vector current(numFlowCells); @@ -1025,7 +1113,7 @@ void ShhherCommand::getUniques(){ uniqueLengths.resize(numUniques); flowDataPrI.resize(numSeqs * numFlowCells, 0); - for(int i=0;icontrol_pressed) { break; } flowDataPrI[i] = getProbIntensity(flowDataIntI[i]); } } catch(exception& e) { m->errorOut(e, "ShhherCommand", "getUniques"); @@ -1046,6 +1134,9 @@ float ShhherCommand::calcPairwiseDist(int seqA, int seqB){ float dist = 0; for(int i=0;icontrol_pressed) { break; } + int flowAIntI = flowDataIntI[ANumFlowCells + i]; float flowAPrI = flowDataPrI[ANumFlowCells + i]; @@ -1078,6 +1169,9 @@ void ShhherCommand::flowDistParentFork(string distFileName, int startSeq, int st double begClock = clock(); for(int i=startSeq;icontrol_pressed) { break; } + for(int j=0;jmothurOutEndLine(); } } - m->mothurOut(toString(stopSeq-1) + "\t" + toString(time(NULL) - begTime)); - m->mothurOut("\t" + toString((clock()-begClock)/CLOCKS_PER_SEC)); - m->mothurOutEndLine(); ofstream distFile(distFileName.c_str()); distFile << outStream.str(); distFile.close(); + + if (m->control_pressed) {} + else { + m->mothurOut(toString(stopSeq-1) + "\t" + toString(time(NULL) - begTime)); + m->mothurOut("\t" + toString((clock()-begClock)/CLOCKS_PER_SEC)); + m->mothurOutEndLine(); + } } catch(exception& e) { m->errorOut(e, "ShhherCommand", "flowDistParentFork"); @@ -1112,31 +1210,37 @@ void ShhherCommand::flowDistParentFork(string distFileName, int startSeq, int st string ShhherCommand::createDistFile(int processors){ try{ +//////////////////////// until I figure out the shared memory issue ////////////////////// +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) +#else + processors=1; +#endif +//////////////////////// until I figure out the shared memory issue ////////////////////// + string fDistFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".shhh.dist"; - unsigned long int begTime = time(NULL); + unsigned long long begTime = time(NULL); double begClock = clock(); - - vector start; - vector end; -#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + if (numSeqs < processors){ processors = 1; } + if(processors == 1) { flowDistParentFork(fDistFileName, 0, numUniques); } + else{ //you have multiple processors - if (numSeqs < processors){ processors = 1; } - vector start(processors, 0); vector end(processors, 0); + int process = 1; + vector processIDs; + for (int i = 0; i < processors; i++) { start[i] = int(sqrt(float(i)/float(processors)) * numUniques); end[i] = int(sqrt(float(i+1)/float(processors)) * numUniques); } - int process = 1; - vector processIDs; - +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + //loop through and create all the processes you want while (process != processors) { int pid = fork(); @@ -1163,24 +1267,55 @@ string ShhherCommand::createDistFile(int processors){ int temp = processIDs[i]; wait(&temp); } +#else + ////////////////////////////////////////////////////////////////////////////////////////////////////// + //Windows version shared memory, so be careful when passing variables through the flowDistParentForkData struct. + //Above fork() will clone, so memory is separate, but that's not the case with windows, + ////////////////////////////////////////////////////////////////////////////////////////////////////// + + vector pDataArray; + DWORD dwThreadIdArray[processors-1]; + HANDLE hThreadArray[processors-1]; + + //Create processor worker threads. + for(int i = 0; i < processors-1; i++){ + // Allocate memory for thread data. + string extension = extension = toString(i) + ".temp"; + + flowDistParentForkData* tempdist = new flowDistParentForkData((fDistFileName + extension), mapUniqueToSeq, mapSeqToUnique, lengths, flowDataIntI, flowDataPrI, jointLookUp, m, start[i+1], end[i+1], numFlowCells, cutoff, i); + pDataArray.push_back(tempdist); + processIDs.push_back(i); + + //MySeqSumThreadFunction is in header. It must be global or static to work with the threads. + //default security attributes, thread function name, argument to thread function, use default creation flags, returns the thread identifier + hThreadArray[i] = CreateThread(NULL, 0, MyflowDistParentForkThreadFunction, pDataArray[i], 0, &dwThreadIdArray[i]); + } + + //parent does its part + flowDistParentFork(fDistFileName, start[0], end[0]); + + //Wait until all threads have terminated. + WaitForMultipleObjects(processors-1, hThreadArray, TRUE, INFINITE); + + //Close all thread handles and free memory allocations. + for(int i=0; i < pDataArray.size(); i++){ + CloseHandle(hThreadArray[i]); + delete pDataArray[i]; + } + +#endif //append and remove temp files for (int i=0;iappendFiles((fDistFileName + toString(processIDs[i]) + ".temp"), fDistFileName); - remove((fDistFileName + toString(processIDs[i]) + ".temp").c_str()); + m->mothurRemove((fDistFileName + toString(processIDs[i]) + ".temp")); } } -#else - flowDistParentFork(fDistFileName, 0, numUniques); -#endif - m->mothurOutEndLine(); - m->mothurOut("Total time: " + toString(time(NULL) - begTime) + '\t' + toString((clock() - begClock)/CLOCKS_PER_SEC) + '\n'); - return fDistFileName; } catch(exception& e) { @@ -1206,6 +1341,9 @@ string ShhherCommand::createNamesFile(){ m->openOutputFile(nameFileName, nameFile); for(int i=0;icontrol_pressed) { break; } + // nameFile << seqNameVector[mapUniqueToSeq[i]] << '\t' << duplicateNames[i].substr(0, duplicateNames[i].find_last_of(',')) << endl; nameFile << mapUniqueToSeq[i] << '\t' << duplicateNames[i].substr(0, duplicateNames[i].find_last_of(',')) << endl; } @@ -1244,6 +1382,9 @@ string ShhherCommand::cluster(string distFileName, string namesFileName){ double clusterCutoff = cutoff; while (matrix->getSmallDist() <= clusterCutoff && matrix->getNNodes() > 0){ + + if (m->control_pressed) { break; } + cluster->update(clusterCutoff); } @@ -1288,6 +1429,8 @@ void ShhherCommand::getOTUData(string listFileName){ string singleOTU = ""; for(int i=0;icontrol_pressed) { break; } listFile >> singleOTU; @@ -1358,6 +1501,8 @@ void ShhherCommand::getOTUData(string listFileName){ void ShhherCommand::initPyroCluster(){ try{ + if (numOTUs < processors) { processors = 1; } + dist.assign(numSeqs * numOTUs, 0); change.assign(numOTUs, 1); centroids.assign(numOTUs, -1); @@ -1387,6 +1532,9 @@ void ShhherCommand::fill(){ try { int index = 0; for(int i=0;icontrol_pressed) { break; } + cumNumSeqs[i] = index; for(int j=0;jcontrol_pressed) { break; } + double count = 0; int position = 0; int minFlowGram = 100000000; @@ -1545,7 +1694,7 @@ double ShhherCommand::getDistToCentroid(int cent, int flow, int length){ int flowBValue = flow * numFlowCells; double dist = 0; - + for(int i=0;icontrol_pressed) { break; } + double difference = weight[i]; weight[i] = 0; @@ -1606,6 +1757,9 @@ double ShhherCommand::getLikelihood(){ string hold; for(int i=0;icontrol_pressed) { break; } + for(int j=0;jcontrol_pressed) { break; } + if(unique[i] == 1){ for(int j=i+1;jcontrol_pressed) { break; } + double offset = 1e8; int indexOffset = i * numOTUs; @@ -1818,6 +1976,9 @@ void ShhherCommand::calcNewDistancesChild(int startSeq, int stopSeq, vector child_singleTau.resize(0); for(int i=startSeq;icontrol_pressed) { break; } + double offset = 1e8; int indexOffset = i * numOTUs; @@ -1869,22 +2030,26 @@ void ShhherCommand::calcNewDistancesParent(int startSeq, int stopSeq){ vector newTau(numOTUs,0); vector norms(numSeqs, 0); nSeqsPerOTU.assign(numOTUs, 0); - + for(int i=startSeq;icontrol_pressed) { break; } + + int indexOffset = i * numOTUs; + double offset = 1e8; for(int j=0;j MIN_WEIGHT && change[j] == 1){ dist[indexOffset + j] = getDistToCentroid(centroids[j], i, lengths[i]); } - + if(weight[j] > MIN_WEIGHT && dist[indexOffset + j] < offset){ offset = dist[indexOffset + j]; } } - + for(int j=0;j MIN_WEIGHT){ newTau[j] = exp(sigma * (-dist[indexOffset + j] + offset)) * weight[j]; @@ -1894,11 +2059,11 @@ void ShhherCommand::calcNewDistancesParent(int startSeq, int stopSeq){ newTau[j] = 0.0; } } - + for(int j=0;j MIN_TAU){ @@ -1917,7 +2082,9 @@ void ShhherCommand::calcNewDistancesParent(int startSeq, int stopSeq){ nSeqsPerOTU[j]++; } } + } + } catch(exception& e) { m->errorOut(e, "ShhherCommand", "calcNewDistancesParent"); @@ -1933,6 +2100,9 @@ void ShhherCommand::setOTUs(){ vector bigTauMatrix(numOTUs * numSeqs, 0.0000); for(int i=0;icontrol_pressed) { break; } + for(int j=0;j otuCounts){ try { - string qualityFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".shhh.qual"; + string thisOutputDir = outputDir; + if (outputDir == "") { thisOutputDir += m->hasPath(flowFileName); } + string qualityFileName = thisOutputDir + m->getRootName(m->getSimpleName(flowFileName)) + ".shhh.qual"; ofstream qualityFile; m->openOutputFile(qualityFileName, qualityFile); @@ -1994,6 +2166,9 @@ void ShhherCommand::writeQualities(vector otuCounts){ for(int i=0;icontrol_pressed) { break; } + int index = 0; int base = 0; @@ -2082,14 +2257,18 @@ void ShhherCommand::writeQualities(vector otuCounts){ void ShhherCommand::writeSequences(vector otuCounts){ try { - - string fastaFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".shhh.fasta"; + string thisOutputDir = outputDir; + if (outputDir == "") { thisOutputDir += m->hasPath(flowFileName); } + string fastaFileName = thisOutputDir + m->getRootName(m->getSimpleName(flowFileName)) + ".shhh.fasta"; ofstream fastaFile; m->openOutputFile(fastaFileName, fastaFile); vector names(numOTUs, ""); for(int i=0;icontrol_pressed) { break; } + int index = centroids[i]; if(otuCounts[i] > 0){ @@ -2126,11 +2305,16 @@ void ShhherCommand::writeSequences(vector otuCounts){ void ShhherCommand::writeNames(vector otuCounts){ try { - string nameFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".shhh.names"; + string thisOutputDir = outputDir; + if (outputDir == "") { thisOutputDir += m->hasPath(flowFileName); } + string nameFileName = thisOutputDir + m->getRootName(m->getSimpleName(flowFileName)) + ".shhh.names"; ofstream nameFile; m->openOutputFile(nameFileName, nameFile); for(int i=0;icontrol_pressed) { break; } + if(otuCounts[i] > 0){ nameFile << seqNameVector[aaI[i][0]] << '\t' << seqNameVector[aaI[i][0]]; @@ -2159,12 +2343,15 @@ void ShhherCommand::writeNames(vector otuCounts){ void ShhherCommand::writeGroups(){ try { - string fileRoot = flowFileName.substr(0,flowFileName.find_last_of('.')); + string thisOutputDir = outputDir; + if (outputDir == "") { thisOutputDir += m->hasPath(flowFileName); } + string fileRoot = thisOutputDir + m->getRootName(m->getSimpleName(flowFileName)); string groupFileName = fileRoot + ".shhh.groups"; ofstream groupFile; m->openOutputFile(groupFileName, groupFile); for(int i=0;icontrol_pressed) { break; } groupFile << seqNameVector[i] << '\t' << fileRoot << endl; } groupFile.close(); @@ -2181,13 +2368,19 @@ void ShhherCommand::writeGroups(){ void ShhherCommand::writeClusters(vector otuCounts){ try { - string otuCountsFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".shhh.counts"; + string thisOutputDir = outputDir; + if (outputDir == "") { thisOutputDir += m->hasPath(flowFileName); } + string otuCountsFileName = thisOutputDir + m->getRootName(m->getSimpleName(flowFileName)) + ".shhh.counts"; ofstream otuCountsFile; m->openOutputFile(otuCountsFileName, otuCountsFile); string bases = flowOrder; for(int i=0;icontrol_pressed) { + break; + } //output the translated version of the centroid sequence for the otu if(otuCounts[i] > 0){ int index = centroids[i];