X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=screenseqscommand.cpp;h=cedb74a9b6c7f0a4162cdc96d9b774105e6cc1d0;hb=3094cb29c613d9687e861e1d0cf9104b7141d24e;hp=7318f9060cdbf5fa2aa5d23205163c1dfecc048a;hpb=86c838c428a9e7d26f902f5492738241fa72c4e7;p=mothur.git diff --git a/screenseqscommand.cpp b/screenseqscommand.cpp index 7318f90..cedb74a 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"); @@ -94,6 +97,7 @@ ScreenSeqsCommand::ScreenSeqsCommand(string option) { //allow user to run help if(option == "help") { help(); abort = true; calledHelp = true; } + else if(option == "citation") { citation(); abort = true; calledHelp = true;} else { vector myArray = setParameters(); @@ -117,6 +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); @@ -163,6 +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 @@ -172,23 +184,31 @@ ScreenSeqsCommand::ScreenSeqsCommand(string option) { if (fastafile != "") { m->mothurOut("Using " + fastafile + " as input file for the fasta parameter."); m->mothurOutEndLine(); } else { m->mothurOut("You have no current fastafile and the fasta parameter is required."); m->mothurOutEndLine(); abort = true; } } - else if (fastafile == "not open") { abort = true; } + else if (fastafile == "not open") { abort = true; } + else { m->setFastaFile(fastafile); } groupfile = validParameter.validFile(parameters, "group", true); if (groupfile == "not open") { abort = true; } else if (groupfile == "not found") { groupfile = ""; } + else { m->setGroupFile(groupfile); } qualfile = validParameter.validFile(parameters, "qfile", true); if (qualfile == "not open") { abort = true; } else if (qualfile == "not found") { qualfile = ""; } + else { m->setQualFile(qualfile); } namefile = validParameter.validFile(parameters, "name", true); if (namefile == "not open") { namefile = ""; abort = true; } else if (namefile == "not found") { namefile = ""; } - + else { m->setNameFile(namefile); } + 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"){ @@ -256,7 +276,7 @@ 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); } @@ -279,7 +299,7 @@ int ScreenSeqsCommand::execute(){ #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 @@ -389,7 +409,7 @@ int ScreenSeqsCommand::execute(){ if(processors == 1){ numFastaSeqs = driver(lines[0], goodSeqFile, badAccnosFile, fastafile, badSeqNames); - if (m->control_pressed) { remove(goodSeqFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(goodSeqFile); return 0; } }else{ processIDS.resize(0); @@ -402,13 +422,13 @@ int ScreenSeqsCommand::execute(){ //append alignment and report files for(int i=1;iappendFiles((goodSeqFile + toString(processIDS[i]) + ".temp"), goodSeqFile); - remove((goodSeqFile + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((goodSeqFile + toString(processIDS[i]) + ".temp")); m->appendFiles((badAccnosFile + toString(processIDS[i]) + ".temp"), badAccnosFile); - remove((badAccnosFile + toString(processIDS[i]) + ".temp").c_str()); + m->mothurRemove((badAccnosFile + toString(processIDS[i]) + ".temp")); } - if (m->control_pressed) { remove(goodSeqFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(goodSeqFile); return 0; } //read badSeqs in because root process doesnt know what other "bad" seqs the children found ifstream inBad; @@ -427,7 +447,7 @@ int ScreenSeqsCommand::execute(){ #else numFastaSeqs = driver(lines[0], goodSeqFile, badAccnosFile, fastafile, badSeqNames); - if (m->control_pressed) { remove(goodSeqFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(goodSeqFile); return 0; } #endif @@ -468,18 +488,19 @@ int ScreenSeqsCommand::execute(){ if(namefile != "" && groupfile != "") { screenNameGroupFile(badSeqNames); - if (m->control_pressed) { remove(goodSeqFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(goodSeqFile); return 0; } }else if(namefile != "") { screenNameGroupFile(badSeqNames); - if (m->control_pressed) { remove(goodSeqFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(goodSeqFile); return 0; } }else if(groupfile != "") { screenGroupFile(badSeqNames); } // this screens just the group - if (m->control_pressed) { remove(goodSeqFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(goodSeqFile); return 0; } if(alignreport != "") { screenAlignReport(badSeqNames); } if(qualfile != "") { screenQual(badSeqNames); } + if(taxonomy != "") { screenTaxonomy(badSeqNames); } - if (m->control_pressed) { remove(goodSeqFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(goodSeqFile); return 0; } #ifdef USE_MPI } @@ -514,6 +535,11 @@ 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(); @@ -542,7 +568,7 @@ int ScreenSeqsCommand::screenNameGroupFile(set badSeqNames){ ofstream goodNameOut; m->openOutputFile(goodNameFile, goodNameOut); while(!inputNames.eof()){ - if (m->control_pressed) { goodNameOut.close(); inputNames.close(); remove(goodNameFile.c_str()); return 0; } + if (m->control_pressed) { goodNameOut.close(); inputNames.close(); m->mothurRemove(goodNameFile); return 0; } inputNames >> seqName >> seqList; it = badSeqNames.find(seqName); @@ -588,7 +614,7 @@ int ScreenSeqsCommand::screenNameGroupFile(set badSeqNames){ ofstream goodGroupOut; m->openOutputFile(goodGroupFile, goodGroupOut); while(!inputGroups.eof()){ - if (m->control_pressed) { goodGroupOut.close(); inputGroups.close(); remove(goodNameFile.c_str()); remove(goodGroupFile.c_str()); return 0; } + if (m->control_pressed) { goodGroupOut.close(); inputGroups.close(); m->mothurRemove(goodNameFile); m->mothurRemove(goodGroupFile); return 0; } inputGroups >> seqName >> group; @@ -624,7 +650,7 @@ int ScreenSeqsCommand::screenNameGroupFile(set badSeqNames){ } } //*************************************************************************************************************** -int ScreenSeqsCommand::getSummary(vector& positions){ +int ScreenSeqsCommand::getSummary(vector& positions){ try { vector startPosition; @@ -633,12 +659,22 @@ int ScreenSeqsCommand::getSummary(vector& positions){ vector ambigBases; vector longHomoPolymer; - vector positions = m->divideFile(fastafile, processors); + 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)])); } + +#ifdef USE_MPI + MPI_Comm_rank(MPI_COMM_WORLD, &pid); + + if (pid == 0) { //only one process should fix files + driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, lines[0]); + } + + MPI_Barrier(MPI_COMM_WORLD); //make everyone wait +#else int numSeqs = 0; #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) if(processors == 1){ @@ -652,7 +688,7 @@ 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()); @@ -719,7 +755,7 @@ int ScreenSeqsCommand::driverCreateSummary(vector& startPosition, vector= filePos->end)) { break; } #else if (in.eof()) { break; } @@ -801,7 +837,7 @@ int ScreenSeqsCommand::createProcessesCreateSummary(vector& startPosition, for (int k = 0; k < tempNum; k++) { in >> temp; longHomoPolymer.push_back(temp); } m->gobble(in); in.close(); - remove(tempFilename.c_str()); + m->mothurRemove(tempFilename); } return num; @@ -827,7 +863,7 @@ int ScreenSeqsCommand::screenGroupFile(set badSeqNames){ ofstream goodGroupOut; m->openOutputFile(goodGroupFile, goodGroupOut); while(!inputGroups.eof()){ - if (m->control_pressed) { goodGroupOut.close(); inputGroups.close(); remove(goodGroupFile.c_str()); return 0; } + if (m->control_pressed) { goodGroupOut.close(); inputGroups.close(); m->mothurRemove(goodGroupFile); return 0; } inputGroups >> seqName >> group; it = badSeqNames.find(seqName); @@ -841,7 +877,7 @@ int ScreenSeqsCommand::screenGroupFile(set badSeqNames){ m->gobble(inputGroups); } - if (m->control_pressed) { goodGroupOut.close(); inputGroups.close(); remove(goodGroupFile.c_str()); return 0; } + if (m->control_pressed) { goodGroupOut.close(); inputGroups.close(); m->mothurRemove(goodGroupFile); return 0; } //we were unable to remove some of the bad sequences if (badSeqNames.size() != 0) { @@ -854,7 +890,7 @@ int ScreenSeqsCommand::screenGroupFile(set badSeqNames){ inputGroups.close(); goodGroupOut.close(); - if (m->control_pressed) { remove(goodGroupFile.c_str()); } + if (m->control_pressed) { m->mothurRemove(goodGroupFile); } return 0; @@ -885,7 +921,7 @@ int ScreenSeqsCommand::screenAlignReport(set badSeqNames){ } while(!inputAlignReport.eof()){ - if (m->control_pressed) { goodAlignReportOut.close(); inputAlignReport.close(); remove(goodAlignReportFile.c_str()); return 0; } + if (m->control_pressed) { goodAlignReportOut.close(); inputAlignReport.close(); m->mothurRemove(goodAlignReportFile); return 0; } inputAlignReport >> seqName; it = badSeqNames.find(seqName); @@ -905,7 +941,7 @@ int ScreenSeqsCommand::screenAlignReport(set badSeqNames){ m->gobble(inputAlignReport); } - if (m->control_pressed) { goodAlignReportOut.close(); inputAlignReport.close(); remove(goodAlignReportFile.c_str()); return 0; } + if (m->control_pressed) { goodAlignReportOut.close(); inputAlignReport.close(); m->mothurRemove(goodAlignReportFile); return 0; } //we were unable to remove some of the bad sequences if (badSeqNames.size() != 0) { @@ -918,7 +954,7 @@ int ScreenSeqsCommand::screenAlignReport(set badSeqNames){ inputAlignReport.close(); goodAlignReportOut.close(); - if (m->control_pressed) { remove(goodAlignReportFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(goodAlignReportFile); return 0; } return 0; @@ -931,6 +967,56 @@ 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; @@ -943,7 +1029,7 @@ int ScreenSeqsCommand::screenQual(set badSeqNames){ while(!in.eof()){ - if (m->control_pressed) { goodQual.close(); in.close(); remove(goodQualFile.c_str()); return 0; } + if (m->control_pressed) { goodQual.close(); in.close(); m->mothurRemove(goodQualFile); return 0; } string saveName = ""; string name = ""; @@ -991,7 +1077,7 @@ int ScreenSeqsCommand::screenQual(set badSeqNames){ } } - if (m->control_pressed) { remove(goodQualFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(goodQualFile); return 0; } return 0; @@ -1045,7 +1131,7 @@ 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; } @@ -1071,7 +1157,7 @@ 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; @@ -1191,7 +1277,7 @@ int ScreenSeqsCommand::createProcesses(string goodFileName, string badAccnos, st string tempFile = filename + toString(processIDS[i]) + ".num.temp"; m->openInputFile(tempFile, in); if (!in.eof()) { int tempNum = 0; in >> tempNum; num += tempNum; } - in.close(); remove(tempFile.c_str()); + in.close(); m->mothurRemove(tempFile); } return num;