X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=pairwiseseqscommand.cpp;h=1e7a3388ce4780d369142385cc0b7bf15e96fae9;hb=a4f05f09c250323132a4b9427f41a3ededaec634;hp=560d54dc98fc198369e832fad43ee9414a646cae;hpb=8f909cf6870cc2e4d8ed8cff69a203f6f83a160b;p=mothur.git diff --git a/pairwiseseqscommand.cpp b/pairwiseseqscommand.cpp index 560d54d..1e7a338 100644 --- a/pairwiseseqscommand.cpp +++ b/pairwiseseqscommand.cpp @@ -14,7 +14,6 @@ #include "needlemanoverlap.hpp" #include "blastalign.hpp" #include "noalign.hpp" -#include "nast.hpp" #include "ignoregaps.h" #include "eachgapdist.h" @@ -61,7 +60,7 @@ vector PairwiseSeqsCommand::getRequiredFiles(){ //********************************************************************************************************************** PairwiseSeqsCommand::PairwiseSeqsCommand(){ try { - //initialize outputTypes + abort = true; calledHelp = true; vector tempOutNames; outputTypes["phylip"] = tempOutNames; outputTypes["column"] = tempOutNames; @@ -74,10 +73,10 @@ PairwiseSeqsCommand::PairwiseSeqsCommand(){ //********************************************************************************************************************** PairwiseSeqsCommand::PairwiseSeqsCommand(string option) { try { - 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 { @@ -134,7 +133,9 @@ PairwiseSeqsCommand::PairwiseSeqsCommand(string option) { if (m->getDefaultPath() != "") { //default path is set string tryPath = m->getDefaultPath() + m->getSimpleName(fastaFileNames[i]); m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying default " + tryPath); m->mothurOutEndLine(); - ableToOpen = m->openInputFile(tryPath, in, "noerror"); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); fastaFileNames[i] = tryPath; } } @@ -144,7 +145,9 @@ PairwiseSeqsCommand::PairwiseSeqsCommand(string option) { if (m->getOutputDir() != "") { //default path is set string tryPath = m->getOutputDir() + m->getSimpleName(fastaFileNames[i]); m->mothurOut("Unable to open " + fastaFileNames[i] + ". Trying output directory " + tryPath); m->mothurOutEndLine(); - ableToOpen = m->openInputFile(tryPath, in, "noerror"); + ifstream in2; + ableToOpen = m->openInputFile(tryPath, in2, "noerror"); + in2.close(); fastaFileNames[i] = tryPath; } } @@ -263,7 +266,7 @@ void PairwiseSeqsCommand::help(){ int PairwiseSeqsCommand::execute(){ try { - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } int longestBase = 2000; //will need to update this in driver if we find sequences with more bases. hardcoded so we don't have the pre-read user fasta file. @@ -437,27 +440,18 @@ int PairwiseSeqsCommand::execute(){ }else{ //you have multiple processors for (int i = 0; i < processors; i++) { - lines.push_back(new linePair()); + distlinePair tempLine; + lines.push_back(tempLine); if (output != "square") { - lines[i]->start = int (sqrt(float(i)/float(processors)) * numSeqs); - lines[i]->end = int (sqrt(float(i+1)/float(processors)) * numSeqs); + lines[i].start = int (sqrt(float(i)/float(processors)) * numSeqs); + lines[i].end = int (sqrt(float(i+1)/float(processors)) * numSeqs); }else{ - lines[i]->start = int ((float(i)/float(processors)) * numSeqs); - lines[i]->end = int ((float(i+1)/float(processors)) * numSeqs); + lines[i].start = int ((float(i)/float(processors)) * numSeqs); + lines[i].end = int ((float(i+1)/float(processors)) * numSeqs); } } - - createProcesses(outputFile); - - map::iterator it = processIDS.begin(); - rename((outputFile + toString(it->second) + ".temp").c_str(), outputFile.c_str()); - it++; - //append and remove temp files - for (; it != processIDS.end(); it++) { - m->appendFiles((outputFile + toString(it->second) + ".temp"), outputFile); - remove((outputFile + toString(it->second) + ".temp").c_str()); - } + createProcesses(outputFile); } #else //ifstream inFASTA; @@ -499,6 +493,19 @@ int PairwiseSeqsCommand::execute(){ delete distCalculator; + //set phylip file as new current phylipfile + string current = ""; + itTypes = outputTypes.find("phylip"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setPhylipFile(current); } + } + + //set column file as new current columnfile + itTypes = outputTypes.find("column"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setColumnFile(current); } + } + m->mothurOutEndLine(); m->mothurOut("Output File Name: "); m->mothurOutEndLine(); for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } @@ -517,7 +524,7 @@ int PairwiseSeqsCommand::execute(){ void PairwiseSeqsCommand::createProcesses(string filename) { try { #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - int process = 0; + int process = 1; processIDS.clear(); //loop through and create all the processes you want @@ -525,20 +532,35 @@ void PairwiseSeqsCommand::createProcesses(string filename) { int pid = fork(); if (pid > 0) { - processIDS[lines[process]->end] = pid; //create map from line number to pid so you can append files in correct order later + processIDS.push_back(pid); process++; }else if (pid == 0){ - if (output != "square") { driver(lines[process]->start, lines[process]->end, filename + toString(getpid()) + ".temp", cutoff); } - else { driver(lines[process]->start, lines[process]->end, filename + toString(getpid()) + ".temp", "square"); } + if (output != "square") { driver(lines[process].start, lines[process].end, filename + toString(getpid()) + ".temp", cutoff); } + else { driver(lines[process].start, lines[process].end, filename + toString(getpid()) + ".temp", "square"); } + exit(0); + }else { + m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine(); + for (int i=0;imothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); } + } } - + + //parent do my part + if (output != "square") { driver(lines[0].start, lines[0].end, filename, cutoff); } + else { driver(lines[0].start, lines[0].end, filename, "square"); } + + //force parent to wait until all the processes are done - for (map::iterator it = processIDS.begin(); it != processIDS.end(); it++) { - int temp = it->second; + for (int i=0;iappendFiles((filename + toString(processIDS[i]) + ".temp"), filename); + remove((filename + toString(processIDS[i]) + ".temp").c_str()); + } #endif } catch(exception& e) { @@ -582,16 +604,16 @@ int PairwiseSeqsCommand::driver(int startLine, int endLine, string dFileName, fl alignment->resize(alignDB.get(j).getUnaligned().length()+1); } - Sequence* seqI = new Sequence(alignDB.get(i).getName(), alignDB.get(i).getAligned()); - Sequence* seqJ = new Sequence(alignDB.get(j).getName(), alignDB.get(j).getAligned()); + Sequence seqI(alignDB.get(i).getName(), alignDB.get(i).getAligned()); + Sequence seqJ(alignDB.get(j).getName(), alignDB.get(j).getAligned()); - Nast(alignment, seqI, seqJ); + alignment->align(seqI.getUnaligned(), seqJ.getUnaligned()); + seqI.setAligned(alignment->getSeqAAln()); + seqJ.setAligned(alignment->getSeqBAln()); - distCalculator->calcDist(*seqI, *seqJ); + distCalculator->calcDist(seqI, seqJ); double dist = distCalculator->getDist(); - delete seqI; delete seqJ; - if(dist <= cutoff){ if (output == "column") { outFile << alignDB.get(i).getName() << ' ' << alignDB.get(j).getName() << ' ' << dist << endl; } } @@ -650,15 +672,15 @@ int PairwiseSeqsCommand::driver(int startLine, int endLine, string dFileName, st alignment->resize(alignDB.get(j).getUnaligned().length()+1); } - Sequence* seqI = new Sequence(alignDB.get(i).getName(), alignDB.get(i).getAligned()); - Sequence* seqJ = new Sequence(alignDB.get(j).getName(), alignDB.get(j).getAligned()); + Sequence seqI(alignDB.get(i).getName(), alignDB.get(i).getAligned()); + Sequence seqJ(alignDB.get(j).getName(), alignDB.get(j).getAligned()); - Nast(alignment, seqI, seqJ); + alignment->align(seqI.getUnaligned(), seqJ.getUnaligned()); + seqI.setAligned(alignment->getSeqAAln()); + seqJ.setAligned(alignment->getSeqBAln()); - distCalculator->calcDist(*seqI, *seqJ); + distCalculator->calcDist(seqI, seqJ); double dist = distCalculator->getDist(); - - delete seqI; delete seqJ; outFile << dist << '\t'; } @@ -705,16 +727,16 @@ int PairwiseSeqsCommand::driverMPI(int startLine, int endLine, MPI_File& outMPI, alignment->resize(alignDB.get(j).getUnaligned().length()+1); } - Sequence* seqI = new Sequence(alignDB.get(i).getName(), alignDB.get(i).getAligned()); - Sequence* seqJ = new Sequence(alignDB.get(j).getName(), alignDB.get(j).getAligned()); + Sequence seqI(alignDB.get(i).getName(), alignDB.get(i).getAligned()); + Sequence seqJ(alignDB.get(j).getName(), alignDB.get(j).getAligned()); - Nast(alignment, seqI, seqJ); + alignment->align(seqI.getUnaligned(), seqJ.getUnaligned()); + seqI.setAligned(alignment->getSeqAAln()); + seqJ.setAligned(alignment->getSeqBAln()); - distCalculator->calcDist(*seqI, *seqJ); + distCalculator->calcDist(seqI, seqJ); double dist = distCalculator->getDist(); - delete seqI; delete seqJ; - if(dist <= cutoff){ outputString += (alignDB.get(i).getName() + ' ' + alignDB.get(j).getName() + ' ' + toString(dist) + '\n'); } @@ -759,7 +781,7 @@ int PairwiseSeqsCommand::driverMPI(int startLine, int endLine, string file, unsi MPI_File_open(MPI_COMM_SELF, filename, amode, MPI_INFO_NULL, &outMPI); - int startTime = time(NULL); + string outputString = ""; size = 0; @@ -786,16 +808,16 @@ int PairwiseSeqsCommand::driverMPI(int startLine, int endLine, string file, unsi alignment->resize(alignDB.get(j).getUnaligned().length()+1); } - Sequence* seqI = new Sequence(alignDB.get(i).getName(), alignDB.get(i).getAligned()); - Sequence* seqJ = new Sequence(alignDB.get(j).getName(), alignDB.get(j).getAligned()); + Sequence seqI(alignDB.get(i).getName(), alignDB.get(i).getAligned()); + Sequence seqJ(alignDB.get(j).getName(), alignDB.get(j).getAligned()); - Nast(alignment, seqI, seqJ); + alignment->align(seqI.getUnaligned(), seqJ.getUnaligned()); + seqI.setAligned(alignment->getSeqAAln()); + seqJ.setAligned(alignment->getSeqBAln()); - distCalculator->calcDist(*seqI, *seqJ); + distCalculator->calcDist(seqI, seqJ); double dist = distCalculator->getDist(); - delete seqI; delete seqJ; - outputString += toString(dist) + "\t"; } @@ -835,7 +857,7 @@ int PairwiseSeqsCommand::driverMPI(int startLine, int endLine, string file, unsi MPI_File_open(MPI_COMM_SELF, filename, amode, MPI_INFO_NULL, &outMPI); - int startTime = time(NULL); + string outputString = ""; size = 0; @@ -862,16 +884,16 @@ int PairwiseSeqsCommand::driverMPI(int startLine, int endLine, string file, unsi alignment->resize(alignDB.get(j).getUnaligned().length()+1); } - Sequence* seqI = new Sequence(alignDB.get(i).getName(), alignDB.get(i).getAligned()); - Sequence* seqJ = new Sequence(alignDB.get(j).getName(), alignDB.get(j).getAligned()); + Sequence seqI(alignDB.get(i).getName(), alignDB.get(i).getAligned()); + Sequence seqJ(alignDB.get(j).getName(), alignDB.get(j).getAligned()); - Nast(alignment, seqI, seqJ); + alignment->align(seqI.getUnaligned(), seqJ.getUnaligned()); + seqI.setAligned(alignment->getSeqAAln()); + seqJ.setAligned(alignment->getSeqBAln()); - distCalculator->calcDist(*seqI, *seqJ); + distCalculator->calcDist(seqI, seqJ); double dist = distCalculator->getDist(); - delete seqI; delete seqJ; - outputString += toString(dist) + "\t"; }