X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=shhhercommand.cpp;h=066a9c51f26b764f14fa3f9490e094748ad3e7b6;hb=0486bc2eed084ac387d2f59b6d23d13b2382daf7;hp=ffddb8cfc9de4dd92feb4966dd8d31477e1363fa;hpb=a3ab9317d12127ff88b039c9933e98eeb562eaa0;p=mothur.git diff --git a/shhhercommand.cpp b/shhhercommand.cpp index ffddb8c..066a9c5 100644 --- a/shhhercommand.cpp +++ b/shhhercommand.cpp @@ -32,7 +32,7 @@ vector ShhherCommand::getValidParameters(){ try { string Array[] = { - "file", "flow", "lookup", "cutoff", "sigma", "outputdir","inputdir", "processors" + "file", "flow", "lookup", "cutoff", "sigma", "outputdir","inputdir", "processors", "maxiter", "mindelta" }; vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); @@ -48,7 +48,7 @@ vector ShhherCommand::getValidParameters(){ ShhherCommand::ShhherCommand(){ try { - abort = true; + abort = true; calledHelp = true; //initialize outputTypes vector tempOutNames; @@ -101,17 +101,17 @@ ShhherCommand::ShhherCommand(string option) { #endif - abort = false; + abort = false; calledHelp = false; //allow user to run help - if(option == "help") { help(); abort = true; } + if(option == "help") { help(); abort = true; calledHelp = true; } else { //valid paramters for this command string AlignArray[] = { - "file", "flow", "lookup", "cutoff", "sigma", "outputdir","inputdir", "processors" + "file", "flow", "lookup", "cutoff", "sigma", "outputdir","inputdir", "processors", "maxiter", "mindelta" }; vector myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string))); @@ -171,7 +171,15 @@ ShhherCommand::ShhherCommand(string option) { m->mothurOutEndLine(); abort = true; } - else if (flowFileName == "not open" || flowFilesFileName == "not open") { abort = true; } + else if (flowFileName == "not open" || flowFilesFileName == "not open") { abort = true; } + + if(flowFileName != "not found"){ compositeFASTAFileName = ""; } + else{ + compositeFASTAFileName = flowFilesFileName.substr(0, flowFilesFileName.length()-10) + "pn.fasta"; + ofstream temp; + m->openOutputFile(compositeFASTAFileName, temp); + temp.close(); + } //if the user changes the output directory command factory will send this info to us in the output parameter outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ @@ -237,17 +245,11 @@ void ShhherCommand::help(){ #ifdef USE_MPI int ShhherCommand::execute(){ try { + if (abort == true) { if (calledHelp) { return 0; } return 2; } + int tag = 1976; MPI_Status status; - double begClock = clock(); - unsigned long int begTime = time(NULL); - - cout.setf(ios::fixed, ios::floatfield); - cout.setf(ios::showpoint); - cout << setprecision(2); - - if(pid == 0){ for(int i=1;imothurOut("\nGetting preliminary data...\n"); getSingleLookUp(); getJointLookUp(); @@ -283,15 +285,19 @@ int ShhherCommand::execute(){ } for(int i=0;i>>>>\tProcessing " << flowFileName << " (file " << i+1 << " of " << numFiles << ")\t<<<<<" << endl; - cout << "Reading flowgrams..." << endl; + + m->mothurOut("\n>>>>>\tProcessing " + flowFileName + " (file " + toString(i+1) + " of " + toString(numFiles) + ")\t<<<<<\n"); + m->mothurOut("Reading flowgrams...\n"); getFlowData(); - cout << "Identifying unique flowgrams..." << endl; + + m->mothurOut("Identifying unique flowgrams...\n"); getUniques(); - cout << "Calculating distances between flowgrams..." << endl; + m->mothurOut("Calculating distances between flowgrams...\n"); char fileName[1024]; strcpy(fileName, flowFileName.c_str()); @@ -322,11 +328,9 @@ int ShhherCommand::execute(){ string namesFileName = createNamesFile(); - cout << "\nClustering flowgrams..." << endl; + m->mothurOut("\nClustering flowgrams...\n"); string listFileName = cluster(distFileName, namesFileName); - // string listFileName = "PriestPot_C7.pn.list"; - // string listFileName = "test.mock_rep3.v69.pn.list"; - + getOTUData(listFileName); initPyroCluster(); @@ -343,9 +347,8 @@ int ShhherCommand::execute(){ int numOTUsOnCPU = numOTUs / ncpus; int numSeqsOnCPU = numSeqs / ncpus; - - cout << "\nDenoising flowgrams..." << endl; - cout << "iter\tmaxDelta\tnLL\t\tcycletime" << endl; + 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)){ @@ -365,7 +368,7 @@ int ShhherCommand::execute(){ MPI_Send(&nSeqsPerOTU[0], numOTUs, MPI_INT, i, tag, MPI_COMM_WORLD); MPI_Send(&cumNumSeqs[0], numOTUs, MPI_INT, i, tag, MPI_COMM_WORLD); } - + calcCentroidsDriver(0, numOTUsOnCPU); for(int i=1;imothurOut(toString(iter) + '\t' + toString(maxDelta) + '\t' + toString(nLL) + '\t' + toString(time(NULL) - cycTime) + '\t' + toString((clock() - cycClock)/(double)CLOCKS_PER_SEC) + '\n'); if((maxIters == 0 && maxDelta > minDelta) || iter < MIN_ITER || (maxDelta > minDelta && iter < maxIters)){ int live = 1; @@ -463,7 +466,7 @@ int ShhherCommand::execute(){ } - cout << "\nFinalizing..." << endl; + m->mothurOut("\nFinalizing...\n"); fill(); setOTUs(); vector otuCounts(numOTUs, 0); @@ -478,14 +481,12 @@ int ShhherCommand::execute(){ remove(distFileName.c_str()); remove(namesFileName.c_str()); remove(listFileName.c_str()); - - cout << "Total time to process " << flowFileName << ":\t" << time(NULL) - begTime << '\t' << setprecision(6) << (clock() - begClock)/(double)CLOCKS_PER_SEC << endl; + + m->mothurOut("Total time to process " + toString(flowFileName) + ":\t" + toString(time(NULL) - begTime) + '\t' + toString((clock() - begClock)/(double)CLOCKS_PER_SEC) + '\n'); } - } else{ int abort = 1; - bool live = 1; MPI_Recv(&abort, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status); if(abort){ return 0; } @@ -495,6 +496,8 @@ int ShhherCommand::execute(){ for(int i=0;imothurOut(toString(i) + '\t' + toString(time(NULL) - begTime) + '\t' + toString((clock()-begClock)/CLOCKS_PER_SEC) + '\n'); } } - cout << stopSeq << "\t" << (time(NULL) - begTime) << "\t" << (clock()-begClock)/CLOCKS_PER_SEC << endl; + + m->mothurOut(toString(stopSeq) + '\t' + toString(time(NULL) - begTime) + '\t' + toString((clock()-begClock)/CLOCKS_PER_SEC) + '\n'); string fDistFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".pn.dist"; if(pid != 0){ fDistFileName += ".temp." + toString(pid); } @@ -651,9 +654,6 @@ int ShhherCommand::execute(){ try { if (abort == true) { return 0; } - cout.setf(ios::fixed, ios::floatfield); - cout.setf(ios::showpoint); - getSingleLookUp(); getJointLookUp(); @@ -678,18 +678,19 @@ int ShhherCommand::execute(){ for(int i=0;i>>>>\tProcessing " << flowFileName << " (file " << i+1 << " of " << numFiles << ")\t<<<<<" << endl; - cout << "Reading flowgrams..." << endl; + m->mothurOut("\n>>>>>\tProcessing " + flowFileName + " (file " + toString(i+1) + " of " + toString(numFiles) + ")\t<<<<<\n"); + m->mothurOut("Reading flowgrams...\n"); getFlowData(); - cout << "Identifying unique flowgrams..." << endl; + + m->mothurOut("Identifying unique flowgrams...\n"); getUniques(); - cout << "Calculating distances between flowgrams..." << endl; + m->mothurOut("Calculating distances between flowgrams...\n"); string distFileName = createDistFile(processors); string namesFileName = createNamesFile(); - - cout << "\nClustering flowgrams..." << endl; + + m->mothurOut("\nClustering flowgrams...\n"); string listFileName = cluster(distFileName, namesFileName); getOTUData(listFileName); @@ -701,8 +702,8 @@ int ShhherCommand::execute(){ double begClock = clock(); unsigned long int begTime = time(NULL); - cout << "\nDenoising flowgrams..." << endl; - cout << "iter\tmaxDelta\tnLL\t\tcycletime" << endl; + 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)){ @@ -720,10 +721,11 @@ int ShhherCommand::execute(){ iter++; - cout << iter << '\t' << maxDelta << '\t' << setprecision(2) << nLL << '\t' << time(NULL) - cycTime << '\t' << setprecision(6) << (clock() - cycClock)/(double)CLOCKS_PER_SEC << endl; + m->mothurOut(toString(iter) + '\t' + toString(maxDelta) + '\t' + toString(nLL) + '\t' + toString(time(NULL) - cycTime) + '\t' + toString((clock() - cycClock)/(double)CLOCKS_PER_SEC) + '\n'); + } - cout << "\nFinalizing..." << endl; + m->mothurOut("\nFinalizing...\n"); fill(); setOTUs(); @@ -741,7 +743,7 @@ int ShhherCommand::execute(){ remove(namesFileName.c_str()); remove(listFileName.c_str()); - cout << "Total time to process " << flowFileName << ":\t" << time(NULL) - begTime << '\t' << setprecision(6) << (clock() - begClock)/(double)CLOCKS_PER_SEC << endl; + m->mothurOut("Total time to process " + flowFileName + ":\t" + toString(time(NULL) - begTime) + '\t' + toString((clock() - begClock)/(double)CLOCKS_PER_SEC) + '\n'); } return 0; } @@ -759,6 +761,11 @@ void ShhherCommand::getFlowData(){ m->openInputFile(flowFileName, flowFile); string seqName; + seqNameVector.clear(); + lengths.clear(); + flowDataIntI.clear(); + nameMap.clear(); + int currentNumFlowCells; @@ -895,13 +902,19 @@ void ShhherCommand::getUniques(){ int index = 0; vector current(numFlowCells); - for(int j=0;j uniqueLengths[j]) { uniqueLengths[j] = lengths[i]; } break; } index++; @@ -934,7 +948,7 @@ void ShhherCommand::getUniques(){ uniqueFlowDataIntI.resize(numFlowCells * numUniques); uniqueLengths.resize(numUniques); - flowDataPrI.assign(numSeqs * numFlowCells, 0); + flowDataPrI.resize(numSeqs * numFlowCells, 0); for(int i=0;imothurOutEndLine(); - cout << "Total time: " << (time(NULL) - begTime) << "\t" << (clock() - begClock)/CLOCKS_PER_SEC << endl;; + m->mothurOut("Total time: " + toString(time(NULL) - begTime) + '\t' + toString((clock() - begClock)/CLOCKS_PER_SEC) + '\n'); + return fDistFileName; } @@ -1133,9 +1148,6 @@ string ShhherCommand::createNamesFile(){ string ShhherCommand::cluster(string distFileName, string namesFileName){ try { - SparseMatrix* matrix; - ListVector* list; - RAbundVector* rabund; globaldata->setNameFile(namesFileName); globaldata->setColumnFile(distFileName); @@ -1148,13 +1160,13 @@ string ShhherCommand::cluster(string distFileName, string namesFileName){ clusterNameMap->readMap(); read->read(clusterNameMap); - list = read->getListVector(); - matrix = read->getMatrix(); + ListVector* list = read->getListVector(); + SparseMatrix* matrix = read->getMatrix(); delete read; delete clusterNameMap; - rabund = new RAbundVector(list->getRAbundVector()); + RAbundVector* rabund = new RAbundVector(list->getRAbundVector()); Cluster* cluster = new CompleteLinkage(rabund, list, matrix, cutoff, "furthest"); string tag = cluster->getTag(); @@ -1196,7 +1208,11 @@ void ShhherCommand::getOTUData(string listFileName){ otuData.assign(numSeqs, 0); cumNumSeqs.assign(numOTUs, 0); nSeqsPerOTU.assign(numOTUs, 0); - aaP.resize(numOTUs); + aaP.clear();aaP.resize(numOTUs); + + seqNumber.clear(); + aaI.clear(); + seqIndex.clear(); string singleOTU = ""; @@ -1248,6 +1264,8 @@ void ShhherCommand::getOTUData(string listFileName){ for(int j=nSeqsPerOTU[i];jerrorOut(e, "ShhherCommand", "getOTUData"); @@ -1375,6 +1394,7 @@ void ShhherCommand::calcCentroidsDriver(int start, int finish){ try{ + for(int i=start;i 0 && count > MIN_COUNT){ vector adF(nSeqsPerOTU[i]); vector anL(nSeqsPerOTU[i]); @@ -1666,9 +1686,9 @@ void ShhherCommand::calcNewDistancesChildMPI(int startSeq, int stopSeq, vector newTau(numOTUs,0); vector norms(numSeqs, 0); - otuIndex.resize(0); - seqIndex.resize(0); - singleTau.resize(0); + otuIndex.clear(); + seqIndex.clear(); + singleTau.clear(); @@ -1990,7 +2010,6 @@ void ShhherCommand::writeQualities(vector otuCounts){ void ShhherCommand::writeSequences(vector otuCounts){ try { - string bases = "TACG"; string fastaFileName = flowFileName.substr(0,flowFileName.find_last_of('.')) + ".pn.fasta"; @@ -2016,6 +2035,10 @@ void ShhherCommand::writeSequences(vector otuCounts){ } } fastaFile.close(); + + if(compositeFASTAFileName != ""){ + m->appendFiles(fastaFileName, compositeFASTAFileName); + } } catch(exception& e) { m->errorOut(e, "ShhherCommand", "writeSequences");