X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=screenseqscommand.cpp;h=ec2d8e003a31a712de8c32d255d164b6b13cd46c;hb=7f0cae4f4853cc3f12bc751ee06ea31c7c97496e;hp=f751525b6b1fa6211ab6017063ba6c7090f6e484;hpb=19fcbbdba99658f5eca244803280f9ee7f9f6607;p=mothur.git diff --git a/screenseqscommand.cpp b/screenseqscommand.cpp index f751525..ec2d8e0 100644 --- a/screenseqscommand.cpp +++ b/screenseqscommand.cpp @@ -18,6 +18,7 @@ vector ScreenSeqsCommand::setParameters(){ CommandParameter pgroup("group", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pgroup); CommandParameter pqfile("qfile", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pqfile); CommandParameter palignreport("alignreport", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(palignreport); + CommandParameter ptax("taxonomy", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(ptax); CommandParameter pstart("start", "Number", "", "-1", "", "", "",false,false); parameters.push_back(pstart); CommandParameter pend("end", "Number", "", "-1", "", "", "",false,false); parameters.push_back(pend); CommandParameter pmaxambig("maxambig", "Number", "", "-1", "", "", "",false,false); parameters.push_back(pmaxambig); @@ -44,8 +45,9 @@ string ScreenSeqsCommand::getHelpString(){ try { string helpString = ""; helpString += "The screen.seqs command reads a fastafile and creates .....\n"; - helpString += "The screen.seqs command parameters are fasta, start, end, maxambig, maxhomop, minlength, maxlength, name, group, qfile, optimize, criteria and processors.\n"; + helpString += "The screen.seqs command parameters are fasta, start, end, maxambig, maxhomop, minlength, maxlength, name, group, qfile, alignreport, taxonomy, optimize, criteria and processors.\n"; helpString += "The fasta parameter is required.\n"; + helpString += "The alignreport and taxonomy parameters allow you to remove bad seqs from taxonomy and alignreport files.\n"; helpString += "The start parameter .... The default is -1.\n"; helpString += "The end parameter .... The default is -1.\n"; helpString += "The maxambig parameter allows you to set the maximum number of ambigious bases allowed. The default is -1.\n"; @@ -80,6 +82,7 @@ ScreenSeqsCommand::ScreenSeqsCommand(){ outputTypes["alignreport"] = tempOutNames; outputTypes["accnos"] = tempOutNames; outputTypes["qfile"] = tempOutNames; + outputTypes["taxonomy"] = tempOutNames; } catch(exception& e) { m->errorOut(e, "ScreenSeqsCommand", "ScreenSeqsCommand"); @@ -118,6 +121,7 @@ ScreenSeqsCommand::ScreenSeqsCommand(string option) { outputTypes["alignreport"] = tempOutNames; outputTypes["accnos"] = tempOutNames; outputTypes["qfile"] = tempOutNames; + outputTypes["taxonomy"] = tempOutNames; //if the user changes the input directory command factory will send this info to us in the output parameter string inputDir = validParameter.validFile(parameters, "inputdir", false); @@ -164,6 +168,13 @@ ScreenSeqsCommand::ScreenSeqsCommand(string option) { if (path == "") { parameters["qfile"] = inputDir + it->second; } } + it = parameters.find("taxonomy"); + //user has given a template file + if(it != parameters.end()){ + path = m->hasPath(it->second); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { parameters["taxonomy"] = inputDir + it->second; } + } } //check for required parameters @@ -193,7 +204,11 @@ ScreenSeqsCommand::ScreenSeqsCommand(string option) { alignreport = validParameter.validFile(parameters, "alignreport", true); if (alignreport == "not open") { abort = true; } - else if (alignreport == "not found") { alignreport = ""; } + else if (alignreport == "not found") { alignreport = ""; } + + taxonomy = validParameter.validFile(parameters, "taxonomy", true); + if (taxonomy == "not open") { abort = true; } + else if (taxonomy == "not found") { taxonomy = ""; } //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"){ @@ -261,17 +276,22 @@ int ScreenSeqsCommand::execute(){ if (abort == true) { if (calledHelp) { return 0; } return 2; } //if the user want to optimize we need to know the 90% mark - vector positions; + vector positions; if (optimize.size() != 0) { //get summary is paralellized so we need to divideFile, no need to do this step twice so I moved it here //use the namefile to optimize correctly if (namefile != "") { nameMap = m->readNames(namefile); } getSummary(positions); } else { + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) positions = m->divideFile(fastafile, processors); for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); - } + } + #else + positions.push_back(0); positions.push_back(1000); + lines.push_back(new linePair(0, 1000)); + #endif } string goodSeqFile = outputDir + m->getRootName(m->getSimpleName(fastafile)) + "good" + m->getExtension(fastafile); @@ -280,16 +300,16 @@ int ScreenSeqsCommand::execute(){ int numFastaSeqs = 0; set badSeqNames; int start = time(NULL); - + #ifdef USE_MPI int pid, numSeqsPerProcessor; int tag = 2001; - vector MPIPos; + vector MPIPos; MPI_Status status; MPI_Comm_rank(MPI_COMM_WORLD, &pid); //find out who we are MPI_Comm_size(MPI_COMM_WORLD, &processors); - + MPI_File inMPI; MPI_File outMPIGood; MPI_File outMPIBadAccnos; @@ -483,6 +503,7 @@ int ScreenSeqsCommand::execute(){ if(alignreport != "") { screenAlignReport(badSeqNames); } if(qualfile != "") { screenQual(badSeqNames); } + if(taxonomy != "") { screenTaxonomy(badSeqNames); } if (m->control_pressed) { m->mothurRemove(goodSeqFile); return 0; } @@ -519,6 +540,11 @@ int ScreenSeqsCommand::execute(){ if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setQualFile(current); } } + + itTypes = outputTypes.find("taxonomy"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setTaxonomyFile(current); } + } m->mothurOut("It took " + toString(time(NULL) - start) + " secs to screen " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); @@ -629,7 +655,7 @@ int ScreenSeqsCommand::screenNameGroupFile(set badSeqNames){ } } //*************************************************************************************************************** -int ScreenSeqsCommand::getSummary(vector& positions){ +int ScreenSeqsCommand::getSummary(vector& positions){ try { vector startPosition; @@ -638,12 +664,23 @@ int ScreenSeqsCommand::getSummary(vector& positions){ vector ambigBases; vector longHomoPolymer; - vector positions = m->divideFile(fastafile, processors); - +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + vector positions = m->divideFile(fastafile, processors); + for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); } +#else + lines.push_back(new linePair(0, 1000)); +#endif +#ifdef USE_MPI + int pid; + MPI_Comm_rank(MPI_COMM_WORLD, &pid); + + if (pid == 0) { + driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, lines[0]); +#else int numSeqs = 0; #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) if(processors == 1){ @@ -657,7 +694,7 @@ int ScreenSeqsCommand::getSummary(vector& positions){ numSeqs = driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, lines[0]); if (m->control_pressed) { return 0; } #endif - +#endif sort(startPosition.begin(), startPosition.end()); sort(endPosition.begin(), endPosition.end()); sort(seqLength.begin(), seqLength.end()); @@ -676,6 +713,33 @@ int ScreenSeqsCommand::getSummary(vector& positions){ else if (optimize[i] == "maxlength") { maxLength = seqLength[criteriaPercentile]; m->mothurOut("Optimizing maxlength to " + toString(maxLength) + "."); m->mothurOutEndLine(); } } +#ifdef USE_MPI + } + + MPI_Status status; + MPI_Comm_rank(MPI_COMM_WORLD, &pid); + MPI_Comm_size(MPI_COMM_WORLD, &processors); + + if (pid == 0) { + //send file positions to all processes + for(int i = 1; i < processors; i++) { + MPI_Send(&startPos, 1, MPI_INT, i, 2001, MPI_COMM_WORLD); + MPI_Send(&endPos, 1, MPI_INT, i, 2001, MPI_COMM_WORLD); + MPI_Send(&maxAmbig, 1, MPI_INT, i, 2001, MPI_COMM_WORLD); + MPI_Send(&maxHomoP, 1, MPI_INT, i, 2001, MPI_COMM_WORLD); + MPI_Send(&minLength, 1, MPI_INT, i, 2001, MPI_COMM_WORLD); + MPI_Send(&maxLength, 1, MPI_INT, i, 2001, MPI_COMM_WORLD); + } + }else { + MPI_Recv(&startPos, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status); + MPI_Recv(&endPos, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status); + MPI_Recv(&maxAmbig, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status); + MPI_Recv(&maxHomoP, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status); + MPI_Recv(&minLength, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status); + MPI_Recv(&maxLength, 1, MPI_INT, 0, 2001, MPI_COMM_WORLD, &status); + } + MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case +#endif return 0; } catch(exception& e) { @@ -722,9 +786,9 @@ int ScreenSeqsCommand::driverCreateSummary(vector& startPosition, vectormothurOut("Optimizing sequence: " + toString(count)); m->mothurOutEndLine(); } #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - unsigned long int pos = in.tellg(); + unsigned long long pos = in.tellg(); if ((pos == -1) || (pos >= filePos->end)) { break; } #else if (in.eof()) { break; } @@ -936,6 +1000,56 @@ int ScreenSeqsCommand::screenAlignReport(set badSeqNames){ } //*************************************************************************************************************** +int ScreenSeqsCommand::screenTaxonomy(set badSeqNames){ + try { + ifstream input; + m->openInputFile(taxonomy, input); + string seqName, tax; + set::iterator it; + + string goodTaxFile = outputDir + m->getRootName(m->getSimpleName(taxonomy)) + "good" + m->getExtension(taxonomy); + outputNames.push_back(goodTaxFile); outputTypes["taxonomy"].push_back(goodTaxFile); + ofstream goodTaxOut; m->openOutputFile(goodTaxFile, goodTaxOut); + + while(!input.eof()){ + if (m->control_pressed) { goodTaxOut.close(); input.close(); m->mothurRemove(goodTaxFile); return 0; } + + input >> seqName >> tax; + it = badSeqNames.find(seqName); + + if(it != badSeqNames.end()){ badSeqNames.erase(it); } + else{ + goodTaxOut << seqName << '\t' << tax << endl; + } + m->gobble(input); + } + + if (m->control_pressed) { goodTaxOut.close(); input.close(); m->mothurRemove(goodTaxFile); return 0; } + + //we were unable to remove some of the bad sequences + if (badSeqNames.size() != 0) { + for (it = badSeqNames.begin(); it != badSeqNames.end(); it++) { + m->mothurOut("Your taxonomy file does not include the sequence " + *it + " please correct."); + m->mothurOutEndLine(); + } + } + + input.close(); + goodTaxOut.close(); + + if (m->control_pressed) { m->mothurRemove(goodTaxFile); return 0; } + + return 0; + + } + catch(exception& e) { + m->errorOut(e, "ScreenSeqsCommand", "screenTaxonomy"); + exit(1); + } + +} +//*************************************************************************************************************** + int ScreenSeqsCommand::screenQual(set badSeqNames){ try { ifstream in; @@ -1050,7 +1164,7 @@ int ScreenSeqsCommand::driver(linePair* filePos, string goodFName, string badAcc } #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - unsigned long int pos = inFASTA.tellg(); + unsigned long long pos = inFASTA.tellg(); if ((pos == -1) || (pos >= filePos->end)) { break; } #else if (inFASTA.eof()) { break; } @@ -1076,7 +1190,7 @@ int ScreenSeqsCommand::driver(linePair* filePos, string goodFName, string badAcc } //********************************************************************************************************************** #ifdef USE_MPI -int ScreenSeqsCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& goodFile, MPI_File& badAccnosFile, vector& MPIPos, set& badSeqNames){ +int ScreenSeqsCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& goodFile, MPI_File& badAccnosFile, vector& MPIPos, set& badSeqNames){ try { string outputString = ""; MPI_Status statusGood;