X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=screenseqscommand.cpp;h=8c334cb217b00d709f7bfa35220dce6a8d6c5dad;hb=a4f05f09c250323132a4b9427f41a3ededaec634;hp=8a7a108eb322978377a9d09ffb880a2745c54caf;hpb=191ae1be0679d5cf4eda950b3b1bf26fb7dd503d;p=mothur.git diff --git a/screenseqscommand.cpp b/screenseqscommand.cpp index 8a7a108..8c334cb 100644 --- a/screenseqscommand.cpp +++ b/screenseqscommand.cpp @@ -26,8 +26,7 @@ vector ScreenSeqsCommand::getValidParameters(){ //********************************************************************************************************************** ScreenSeqsCommand::ScreenSeqsCommand(){ try { - abort = true; - //initialize outputTypes + abort = true; calledHelp = true; vector tempOutNames; outputTypes["fasta"] = tempOutNames; outputTypes["name"] = tempOutNames; @@ -67,10 +66,10 @@ vector ScreenSeqsCommand::getRequiredFiles(){ ScreenSeqsCommand::ScreenSeqsCommand(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 { //valid paramters for this command @@ -145,7 +144,7 @@ ScreenSeqsCommand::ScreenSeqsCommand(string option) { else if (groupfile == "not found") { groupfile = ""; } namefile = validParameter.validFile(parameters, "name", true); - if (namefile == "not open") { abort = true; } + if (namefile == "not open") { namefile = ""; abort = true; } else if (namefile == "not found") { namefile = ""; } alignreport = validParameter.validFile(parameters, "alignreport", true); @@ -247,11 +246,15 @@ ScreenSeqsCommand::~ScreenSeqsCommand(){ /* do nothing */ } int ScreenSeqsCommand::execute(){ try{ - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } - //if the user want to optimize we need to no the 90% mark + //if the user want to optimize we need to know the 90% mark vector positions; - if (optimize.size() != 0) { getSummary(positions); } //get summary is paralellized so we need to divideFile, no need to do this step twice so I moved it here + 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 { positions = m->divideFile(fastafile, processors); for (int i = 0; i < (positions.size()-1); i++) { @@ -267,7 +270,7 @@ int ScreenSeqsCommand::execute(){ int start = time(NULL); #ifdef USE_MPI - int pid, end, numSeqsPerProcessor; + int pid, numSeqsPerProcessor; int tag = 2001; vector MPIPos; @@ -481,6 +484,23 @@ int ScreenSeqsCommand::execute(){ for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); } m->mothurOutEndLine(); m->mothurOutEndLine(); + + //set fasta file as new current fastafile + string current = ""; + itTypes = outputTypes.find("fasta"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setFastaFile(current); } + } + + itTypes = outputTypes.find("name"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setNameFile(current); } + } + + itTypes = outputTypes.find("group"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setGroupFile(current); } + } m->mothurOut("It took " + toString(time(NULL) - start) + " secs to screen " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); @@ -507,13 +527,13 @@ int ScreenSeqsCommand::screenNameGroupFile(set badSeqNames){ outputNames.push_back(goodNameFile); outputTypes["name"].push_back(goodNameFile); ofstream goodNameOut; m->openOutputFile(goodNameFile, goodNameOut); - + while(!inputNames.eof()){ if (m->control_pressed) { goodNameOut.close(); inputNames.close(); remove(goodNameFile.c_str()); return 0; } inputNames >> seqName >> seqList; it = badSeqNames.find(seqName); - + if(it != badSeqNames.end()){ badSeqNames.erase(it); @@ -558,7 +578,7 @@ int ScreenSeqsCommand::screenNameGroupFile(set badSeqNames){ if (m->control_pressed) { goodGroupOut.close(); inputGroups.close(); remove(goodNameFile.c_str()); remove(goodGroupFile.c_str()); return 0; } inputGroups >> seqName >> group; - + it = badSeqGroups.find(seqName); if(it != badSeqGroups.end()){ @@ -580,7 +600,8 @@ int ScreenSeqsCommand::screenNameGroupFile(set badSeqNames){ } } } - + + return 0; } @@ -625,14 +646,15 @@ int ScreenSeqsCommand::getSummary(vector& positions){ sort(ambigBases.begin(), ambigBases.end()); sort(longHomoPolymer.begin(), longHomoPolymer.end()); - int criteriaPercentile = int(numSeqs * (criteria / (float) 100)); + //numSeqs is the number of unique seqs, startPosition.size() is the total number of seqs, we want to optimize using all seqs + int criteriaPercentile = int(startPosition.size() * (criteria / (float) 100)); for (int i = 0; i < optimize.size(); i++) { if (optimize[i] == "start") { startPos = startPosition[criteriaPercentile]; m->mothurOut("Optimizing start to " + toString(startPos) + "."); m->mothurOutEndLine(); } - else if (optimize[i] == "end") { int endcriteriaPercentile = int(numSeqs * ((100 - criteria) / (float) 100)); endPos = endPosition[endcriteriaPercentile]; m->mothurOut("Optimizing end to " + toString(endPos) + "."); m->mothurOutEndLine();} + else if (optimize[i] == "end") { int endcriteriaPercentile = int(endPosition.size() * ((100 - criteria) / (float) 100)); endPos = endPosition[endcriteriaPercentile]; m->mothurOut("Optimizing end to " + toString(endPos) + "."); m->mothurOutEndLine();} else if (optimize[i] == "maxambig") { maxAmbig = ambigBases[criteriaPercentile]; m->mothurOut("Optimizing maxambig to " + toString(maxAmbig) + "."); m->mothurOutEndLine(); } else if (optimize[i] == "maxhomop") { maxHomoP = longHomoPolymer[criteriaPercentile]; m->mothurOut("Optimizing maxhomop to " + toString(maxHomoP) + "."); m->mothurOutEndLine(); } - else if (optimize[i] == "minlength") { int mincriteriaPercentile = int(numSeqs * ((100 - criteria) / (float) 100)); minLength = seqLength[mincriteriaPercentile]; m->mothurOut("Optimizing minlength to " + toString(minLength) + "."); m->mothurOutEndLine(); } + else if (optimize[i] == "minlength") { int mincriteriaPercentile = int(seqLength.size() * ((100 - criteria) / (float) 100)); minLength = seqLength[mincriteriaPercentile]; m->mothurOut("Optimizing minlength to " + toString(minLength) + "."); m->mothurOutEndLine(); } else if (optimize[i] == "maxlength") { maxLength = seqLength[criteriaPercentile]; m->mothurOut("Optimizing maxlength to " + toString(maxLength) + "."); m->mothurOutEndLine(); } } @@ -662,11 +684,24 @@ int ScreenSeqsCommand::driverCreateSummary(vector& startPosition, vectorgobble(in); if (current.getName() != "") { - startPosition.push_back(current.getStartPos()); - endPosition.push_back(current.getEndPos()); - seqLength.push_back(current.getNumBases()); - ambigBases.push_back(current.getAmbigBases()); - longHomoPolymer.push_back(current.getLongHomoPolymer()); + int num = 1; + if (namefile != "") { + //make sure this sequence is in the namefile, else error + map::iterator it = nameMap.find(current.getName()); + + if (it == nameMap.end()) { m->mothurOut("[ERROR]: " + current.getName() + " is not in your namefile, please correct."); m->mothurOutEndLine(); m->control_pressed = true; } + else { num = it->second; } + } + + //for each sequence this sequence represents + for (int i = 0; i < num; i++) { + startPosition.push_back(current.getStartPos()); + endPosition.push_back(current.getEndPos()); + seqLength.push_back(current.getNumBases()); + ambigBases.push_back(current.getAmbigBases()); + longHomoPolymer.push_back(current.getLongHomoPolymer()); + } + count++; } @@ -712,6 +747,7 @@ int ScreenSeqsCommand::createProcessesCreateSummary(vector& startPosition, m->openOutputFile(tempFile, out); out << num << endl; + out << startPosition.size() << endl; for (int k = 0; k < startPosition.size(); k++) { out << startPosition[k] << '\t'; } out << endl; for (int k = 0; k < endPosition.size(); k++) { out << endPosition[k] << '\t'; } out << endl; for (int k = 0; k < seqLength.size(); k++) { out << seqLength[k] << '\t'; } out << endl; @@ -721,7 +757,11 @@ int ScreenSeqsCommand::createProcessesCreateSummary(vector& startPosition, out.close(); exit(0); - }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); } + }else { + m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine(); + for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); } + exit(0); + } } num = driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, lines[0]); @@ -740,6 +780,7 @@ int ScreenSeqsCommand::createProcessesCreateSummary(vector& startPosition, int temp, tempNum; in >> tempNum; m->gobble(in); num += tempNum; + in >> tempNum; m->gobble(in); for (int k = 0; k < tempNum; k++) { in >> temp; startPosition.push_back(temp); } m->gobble(in); for (int k = 0; k < tempNum; k++) { in >> temp; endPosition.push_back(temp); } m->gobble(in); for (int k = 0; k < tempNum; k++) { in >> temp; seqLength.push_back(temp); } m->gobble(in); @@ -1007,6 +1048,9 @@ int ScreenSeqsCommand::driverMPI(int start, int num, MPI_File& inMPI, MPI_File& delete buf3; } } + + //report progress + if((i) % 100 == 0){ m->mothurOut("Processing sequence: " + toString(i)); m->mothurOutEndLine(); } } return 1; @@ -1043,7 +1087,11 @@ int ScreenSeqsCommand::createProcesses(string goodFileName, string badAccnos, st out.close(); exit(0); - }else { m->mothurOut("unable to spawn the necessary processes."); m->mothurOutEndLine(); exit(0); } + }else { + m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine(); + for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); } + exit(0); + } } //force parent to wait until all the processes are done