X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=seqsummarycommand.cpp;h=fdf95ee6fc5f87ed7629de936f4b0543dc89906a;hp=620de95c333cdee1360157f7aa7068f3ce1f1f62;hb=1a20e24ee786195ab0e1cccd4f5aede7a88f3f4e;hpb=a150db3c2bfd7b76420048a1e94ebe397f2c6045 diff --git a/seqsummarycommand.cpp b/seqsummarycommand.cpp index 620de95..fdf95ee 100644 --- a/seqsummarycommand.cpp +++ b/seqsummarycommand.cpp @@ -8,16 +8,17 @@ */ #include "seqsummarycommand.h" -#include "sequence.hpp" +#include "counttable.h" //********************************************************************************************************************** vector SeqSummaryCommand::setParameters(){ try { - CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pfasta); - CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname); - CommandParameter pprocessors("processors", "Number", "", "1", "", "", "",false,false); parameters.push_back(pprocessors); - CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir); - CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir); + CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none","summary",false,true,true); parameters.push_back(pfasta); + CommandParameter pname("name", "InputTypes", "", "", "namecount", "none", "none","",false,false,true); parameters.push_back(pname); + CommandParameter pcount("count", "InputTypes", "", "", "namecount", "none", "none","",false,false,true); parameters.push_back(pcount); + CommandParameter pprocessors("processors", "Number", "", "1", "", "", "","",false,false,true); parameters.push_back(pprocessors); + CommandParameter pinputdir("inputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(pinputdir); + CommandParameter poutputdir("outputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(poutputdir); vector myArray; for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); } @@ -33,8 +34,9 @@ string SeqSummaryCommand::getHelpString(){ try { string helpString = ""; helpString += "The summary.seqs command reads a fastafile and summarizes the sequences.\n"; - helpString += "The summary.seqs command parameters are fasta, name and processors, fasta is required, unless you have a valid current fasta file.\n"; + helpString += "The summary.seqs command parameters are fasta, name, count and processors, fasta is required, unless you have a valid current fasta file.\n"; helpString += "The name parameter allows you to enter a name file associated with your fasta file. \n"; + helpString += "The count parameter allows you to enter a count file associated with your fasta file. \n"; helpString += "The summary.seqs command should be in the following format: \n"; helpString += "summary.seqs(fasta=yourFastaFile, processors=2) \n"; helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n"; @@ -45,6 +47,21 @@ string SeqSummaryCommand::getHelpString(){ exit(1); } } +//********************************************************************************************************************** +string SeqSummaryCommand::getOutputPattern(string type) { + try { + string pattern = ""; + + if (type == "summary") { pattern = "[filename],summary"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "SeqSummaryCommand", "getOutputPattern"); + exit(1); + } +} //********************************************************************************************************************** SeqSummaryCommand::SeqSummaryCommand(){ @@ -103,6 +120,14 @@ SeqSummaryCommand::SeqSummaryCommand(string option) { //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { parameters["name"] = inputDir + it->second; } } + + it = parameters.find("count"); + //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["count"] = inputDir + it->second; } + } } //initialize outputTypes @@ -116,11 +141,19 @@ SeqSummaryCommand::SeqSummaryCommand(string option) { fastafile = m->getFastaFile(); 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 { m->setFastaFile(fastafile); } namefile = validParameter.validFile(parameters, "name", true); if (namefile == "not open") { namefile = ""; abort = true; } else if (namefile == "not found") { namefile = ""; } + else { m->setNameFile(namefile); } + + countfile = validParameter.validFile(parameters, "count", true); + if (countfile == "not open") { abort = true; countfile = ""; } + else if (countfile == "not found") { countfile = ""; } + else { m->setCountTableFile(countfile); } + + if ((countfile != "") && (namefile != "")) { m->mothurOut("You must enter ONLY ONE of the following: count or name."); m->mothurOutEndLine(); abort = true; } //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"){ @@ -130,9 +163,14 @@ SeqSummaryCommand::SeqSummaryCommand(string option) { string temp = validParameter.validFile(parameters, "processors", false); if (temp == "not found"){ temp = m->getProcessors(); } m->setProcessors(temp); - convert(temp, processors); - - + m->mothurConvert(temp, processors); + + if (countfile == "") { + if (namefile == "") { + vector files; files.push_back(fastafile); + parser.getNameFile(files); + } + } } } catch(exception& e) { @@ -150,7 +188,9 @@ int SeqSummaryCommand::execute(){ //set current fasta to fastafile m->setFastaFile(fastafile); - string summaryFile = outputDir + m->getSimpleName(fastafile) + ".summary"; + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastafile)); + string summaryFile = getOutputFileName("summary",variables); int numSeqs = 0; @@ -161,6 +201,11 @@ int SeqSummaryCommand::execute(){ vector longHomoPolymer; if (namefile != "") { nameMap = m->readNames(namefile); } + else if (countfile != "") { + CountTable ct; + ct.readTable(countfile); + nameMap = ct.getNameMap(); + } if (m->control_pressed) { return 0; } @@ -169,7 +214,7 @@ int SeqSummaryCommand::execute(){ int tag = 2001; int startTag = 1; int endTag = 2; int lengthTag = 3; int baseTag = 4; int lhomoTag = 5; int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY; - vector MPIPos; + vector MPIPos; MPI_Status status; MPI_Status statusOut; @@ -280,31 +325,31 @@ int SeqSummaryCommand::execute(){ MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case #else - 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)])); - } - - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - if(processors == 1){ - numSeqs = driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile, lines[0]); - }else{ - numSeqs = createProcessesCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile); - - rename((summaryFile + toString(processIDS[0]) + ".temp").c_str(), summaryFile.c_str()); - //append files - for(int i=1;iappendFiles((summaryFile + toString(processIDS[i]) + ".temp"), summaryFile); - remove((summaryFile + toString(processIDS[i]) + ".temp").c_str()); - } + vector positions; + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) + 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 = m->setFilePosFasta(fastafile, numSeqs); + if (positions.size() < processors) { processors = positions.size(); } + + //figure out how many sequences you have to process + int numSeqsPerProcessor = numSeqs / processors; + for (int i = 0; i < processors; i++) { + int startIndex = i * numSeqsPerProcessor; + if(i == (processors - 1)){ numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; } + lines.push_back(new linePair(positions[startIndex], numSeqsPerProcessor)); } - - if (m->control_pressed) { return 0; } - #else + #endif + + + if(processors == 1){ numSeqs = driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile, lines[0]); - if (m->control_pressed) { return 0; } - #endif + }else{ + numSeqs = createProcessesCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile); + } + + if (m->control_pressed) { return 0; } #endif #ifdef USE_MPI @@ -317,6 +362,20 @@ int SeqSummaryCommand::execute(){ sort(ambigBases.begin(), ambigBases.end()); sort(longHomoPolymer.begin(), longHomoPolymer.end()); int size = startPosition.size(); + + //find means + double meanStartPosition, meanEndPosition, meanSeqLength, meanAmbigBases, meanLongHomoPolymer; + meanStartPosition = 0; meanEndPosition = 0; meanSeqLength = 0; meanAmbigBases = 0; meanLongHomoPolymer = 0; + for (int i = 0; i < size; i++) { + meanStartPosition += startPosition[i]; + meanEndPosition += endPosition[i]; + meanSeqLength += seqLength[i]; + meanAmbigBases += ambigBases[i]; + meanLongHomoPolymer += longHomoPolymer[i]; + } + + //this is an int divide so the remainder is lost + meanStartPosition /= (float) size; meanEndPosition /= (float) size; meanLongHomoPolymer /= (float) size; meanSeqLength /= (float) size; meanAmbigBases /= (float) size; int ptile0_25 = int(size * 0.025); int ptile25 = int(size * 0.250); @@ -329,24 +388,26 @@ int SeqSummaryCommand::execute(){ if (startPosition[0] == -1) { startPosition[0] = 0; } if (endPosition[0] == -1) { endPosition[0] = 0; } - if (m->control_pressed) { remove(summaryFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; } m->mothurOutEndLine(); - m->mothurOut("\t\tStart\tEnd\tNBases\tAmbigs\tPolymer"); m->mothurOutEndLine(); - m->mothurOut("Minimum:\t" + toString(startPosition[0]) + "\t" + toString(endPosition[0]) + "\t" + toString(seqLength[0]) + "\t" + toString(ambigBases[0]) + "\t" + toString(longHomoPolymer[0])); m->mothurOutEndLine(); - m->mothurOut("2.5%-tile:\t" + toString(startPosition[ptile0_25]) + "\t" + toString(endPosition[ptile0_25]) + "\t" + toString(seqLength[ptile0_25]) + "\t" + toString(ambigBases[ptile0_25]) + "\t"+ toString(longHomoPolymer[ptile0_25])); m->mothurOutEndLine(); - m->mothurOut("25%-tile:\t" + toString(startPosition[ptile25]) + "\t" + toString(endPosition[ptile25]) + "\t" + toString(seqLength[ptile25]) + "\t" + toString(ambigBases[ptile25]) + "\t" + toString(longHomoPolymer[ptile25])); m->mothurOutEndLine(); - m->mothurOut("Median: \t" + toString(startPosition[ptile50]) + "\t" + toString(endPosition[ptile50]) + "\t" + toString(seqLength[ptile50]) + "\t" + toString(ambigBases[ptile50]) + "\t" + toString(longHomoPolymer[ptile50])); m->mothurOutEndLine(); - m->mothurOut("75%-tile:\t" + toString(startPosition[ptile75]) + "\t" + toString(endPosition[ptile75]) + "\t" + toString(seqLength[ptile75]) + "\t" + toString(ambigBases[ptile75]) + "\t" + toString(longHomoPolymer[ptile75])); m->mothurOutEndLine(); - m->mothurOut("97.5%-tile:\t" + toString(startPosition[ptile97_5]) + "\t" + toString(endPosition[ptile97_5]) + "\t" + toString(seqLength[ptile97_5]) + "\t" + toString(ambigBases[ptile97_5]) + "\t" + toString(longHomoPolymer[ptile97_5])); m->mothurOutEndLine(); - m->mothurOut("Maximum:\t" + toString(startPosition[ptile100]) + "\t" + toString(endPosition[ptile100]) + "\t" + toString(seqLength[ptile100]) + "\t" + toString(ambigBases[ptile100]) + "\t" + toString(longHomoPolymer[ptile100])); m->mothurOutEndLine(); - if (namefile == "") { m->mothurOut("# of Seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); } + m->mothurOut("\t\tStart\tEnd\tNBases\tAmbigs\tPolymer\tNumSeqs"); m->mothurOutEndLine(); + m->mothurOut("Minimum:\t" + toString(startPosition[0]) + "\t" + toString(endPosition[0]) + "\t" + toString(seqLength[0]) + "\t" + toString(ambigBases[0]) + "\t" + toString(longHomoPolymer[0]) + "\t" + toString(1)); m->mothurOutEndLine(); + m->mothurOut("2.5%-tile:\t" + toString(startPosition[ptile0_25]) + "\t" + toString(endPosition[ptile0_25]) + "\t" + toString(seqLength[ptile0_25]) + "\t" + toString(ambigBases[ptile0_25]) + "\t"+ toString(longHomoPolymer[ptile0_25]) + "\t" + toString(ptile0_25+1)); m->mothurOutEndLine(); + m->mothurOut("25%-tile:\t" + toString(startPosition[ptile25]) + "\t" + toString(endPosition[ptile25]) + "\t" + toString(seqLength[ptile25]) + "\t" + toString(ambigBases[ptile25]) + "\t" + toString(longHomoPolymer[ptile25]) + "\t" + toString(ptile25+1)); m->mothurOutEndLine(); + m->mothurOut("Median: \t" + toString(startPosition[ptile50]) + "\t" + toString(endPosition[ptile50]) + "\t" + toString(seqLength[ptile50]) + "\t" + toString(ambigBases[ptile50]) + "\t" + toString(longHomoPolymer[ptile50]) + "\t" + toString(ptile50+1)); m->mothurOutEndLine(); + m->mothurOut("75%-tile:\t" + toString(startPosition[ptile75]) + "\t" + toString(endPosition[ptile75]) + "\t" + toString(seqLength[ptile75]) + "\t" + toString(ambigBases[ptile75]) + "\t" + toString(longHomoPolymer[ptile75]) + "\t" + toString(ptile75+1)); m->mothurOutEndLine(); + m->mothurOut("97.5%-tile:\t" + toString(startPosition[ptile97_5]) + "\t" + toString(endPosition[ptile97_5]) + "\t" + toString(seqLength[ptile97_5]) + "\t" + toString(ambigBases[ptile97_5]) + "\t" + toString(longHomoPolymer[ptile97_5]) + "\t" + toString(ptile97_5+1)); m->mothurOutEndLine(); + m->mothurOut("Maximum:\t" + toString(startPosition[ptile100]) + "\t" + toString(endPosition[ptile100]) + "\t" + toString(seqLength[ptile100]) + "\t" + toString(ambigBases[ptile100]) + "\t" + toString(longHomoPolymer[ptile100]) + "\t" + toString(ptile100+1)); m->mothurOutEndLine(); + m->mothurOut("Mean:\t" + toString(meanStartPosition) + "\t" + toString(meanEndPosition) + "\t" + toString(meanSeqLength) + "\t" + toString(meanAmbigBases) + "\t" + toString(meanLongHomoPolymer)); m->mothurOutEndLine(); + + if ((namefile == "") && (countfile == "")) { m->mothurOut("# of Seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); } else { m->mothurOut("# of unique seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); m->mothurOut("total # of seqs:\t" + toString(startPosition.size())); m->mothurOutEndLine(); } - if (m->control_pressed) { remove(summaryFile.c_str()); return 0; } + if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; } m->mothurOutEndLine(); - m->mothurOut("Output File Name: "); m->mothurOutEndLine(); + m->mothurOut("Output File Names: "); m->mothurOutEndLine(); m->mothurOut(summaryFile); m->mothurOutEndLine(); outputNames.push_back(summaryFile); outputTypes["summary"].push_back(summaryFile); m->mothurOutEndLine(); @@ -380,21 +441,21 @@ int SeqSummaryCommand::driverCreateSummary(vector& startPosition, vectorcontrol_pressed) { in.close(); outSummary.close(); return 1; } Sequence current(in); m->gobble(in); - + if (current.getName() != "") { int num = 1; - if (namefile != "") { + if ((namefile != "") || (countfile != "")) { //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; } + if (it == nameMap.end()) { m->mothurOut("[ERROR]: '" + current.getName() + "' is not in your name or count file, please correct."); m->mothurOutEndLine(); m->control_pressed = true; } else { num = it->second; } } @@ -414,8 +475,8 @@ int SeqSummaryCommand::driverCreateSummary(vector& startPosition, vector= filePos->end)) { break; } #else if (in.eof()) { break; } @@ -438,7 +499,7 @@ int SeqSummaryCommand::driverCreateSummary(vector& startPosition, vector& startPosition, vector& endPosition, vector& seqLength, vector& ambigBases, vector& longHomoPolymer, MPI_File& inMPI, MPI_File& outMPI, vector& MPIPos) { +int SeqSummaryCommand::MPICreateSummary(int start, int num, vector& startPosition, vector& endPosition, vector& seqLength, vector& ambigBases, vector& longHomoPolymer, MPI_File& inMPI, MPI_File& outMPI, vector& MPIPos) { try { int pid; @@ -465,11 +526,11 @@ int SeqSummaryCommand::MPICreateSummary(int start, int num, vector& startPo if (current.getName() != "") { int num = 1; - if (namefile != "") { + if ((namefile != "") || (countfile != "")) { //make sure this sequence is in the namefile, else error map::iterator it = nameMap.find(current.getName()); - if (it == nameMap.end()) { cout << "[ERROR]: " << current.getName() << " is not in your namefile, please correct." << endl; m->control_pressed = true; } + if (it == nameMap.end()) { cout << "[ERROR]: " << current.getName() << " is not in your name or count file, please correct." << endl; m->control_pressed = true; } else { num = it->second; } } @@ -506,11 +567,12 @@ int SeqSummaryCommand::MPICreateSummary(int start, int num, vector& startPo /**************************************************************************************************/ int SeqSummaryCommand::createProcessesCreateSummary(vector& startPosition, vector& endPosition, vector& seqLength, vector& ambigBases, vector& longHomoPolymer, string filename, string sumFile) { try { -#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) - int process = 0; + int process = 1; int num = 0; processIDS.clear(); +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) + //loop through and create all the processes you want while (process != processors) { int pid = fork(); @@ -544,8 +606,11 @@ int SeqSummaryCommand::createProcessesCreateSummary(vector& startPosition, } } + //do your part + num = driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, sumFile, lines[0]); + //force parent to wait until all the processes are done - for (int i=0;i& 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); + + m->appendFiles((sumFile + toString(processIDS[i]) + ".temp"), sumFile); + m->mothurRemove((sumFile + toString(processIDS[i]) + ".temp")); } - return num; +#else + ////////////////////////////////////////////////////////////////////////////////////////////////////// + //Windows version shared memory, so be careful when passing variables through the seqSumData struct. + //Above fork() will clone, so memory is separate, but that's not the case with windows, + //Taking advantage of shared memory to allow both threads to add info to vectors. + ////////////////////////////////////////////////////////////////////////////////////////////////////// + + vector pDataArray; + DWORD dwThreadIdArray[processors-1]; + HANDLE hThreadArray[processors-1]; + + bool hasNameMap = false; + if ((namefile !="") || (countfile != "")) { hasNameMap = true; } + + //Create processor worker threads. + for( int i=0; istart, lines[i]->end, hasNameMap, nameMap); + pDataArray.push_back(tempSum); + + //MySeqSumThreadFunction is in header. It must be global or static to work with the threads. + //default security attributes, thread function name, argument to thread function, use default creation flags, returns the thread identifier + hThreadArray[i] = CreateThread(NULL, 0, MySeqSumThreadFunction, pDataArray[i], 0, &dwThreadIdArray[i]); + } + + //do your part + num = driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, (sumFile+toString(processors-1)+".temp"), lines[processors-1]); + processIDS.push_back(processors-1); + + //Wait until all threads have terminated. + WaitForMultipleObjects(processors-1, hThreadArray, TRUE, INFINITE); + + //Close all thread handles and free memory allocations. + for(int i=0; i < pDataArray.size(); i++){ + num += pDataArray[i]->count; + if (pDataArray[i]->count != pDataArray[i]->end) { + m->mothurOut("[ERROR]: process " + toString(i) + " only processed " + toString(pDataArray[i]->count) + " of " + toString(pDataArray[i]->end) + " sequences assigned to it, quitting. \n"); m->control_pressed = true; + } + for (int k = 0; k < pDataArray[i]->startPosition.size(); k++) { startPosition.push_back(pDataArray[i]->startPosition[k]); } + for (int k = 0; k < pDataArray[i]->endPosition.size(); k++) { endPosition.push_back(pDataArray[i]->endPosition[k]); } + for (int k = 0; k < pDataArray[i]->seqLength.size(); k++) { seqLength.push_back(pDataArray[i]->seqLength[k]); } + for (int k = 0; k < pDataArray[i]->ambigBases.size(); k++) { ambigBases.push_back(pDataArray[i]->ambigBases[k]); } + for (int k = 0; k < pDataArray[i]->longHomoPolymer.size(); k++) { longHomoPolymer.push_back(pDataArray[i]->longHomoPolymer[k]); } + CloseHandle(hThreadArray[i]); + delete pDataArray[i]; + } + + //append files + for(int i=0;iappendFiles((sumFile + toString(processIDS[i]) + ".temp"), sumFile); + m->mothurRemove((sumFile + toString(processIDS[i]) + ".temp")); + } #endif + return num; } catch(exception& e) { m->errorOut(e, "SeqSummaryCommand", "createProcessesCreateSummary");