X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=seqsummarycommand.cpp;h=6d8336cb40217a9547791e69765230dff0dabe62;hp=37d0fdfe10fdbe3eb11d3c7f74cbc84389acead9;hb=c48d91112209b841444923670dca5454da0e2a4d;hpb=5d00ef3e809832f08efdd691a9eace8ac20feb07 diff --git a/seqsummarycommand.cpp b/seqsummarycommand.cpp index 37d0fdf..6d8336c 100644 --- a/seqsummarycommand.cpp +++ b/seqsummarycommand.cpp @@ -8,16 +8,17 @@ */ #include "seqsummarycommand.h" - +#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 @@ -122,6 +147,13 @@ SeqSummaryCommand::SeqSummaryCommand(string option) { 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"){ @@ -133,11 +165,12 @@ SeqSummaryCommand::SeqSummaryCommand(string option) { m->setProcessors(temp); m->mothurConvert(temp, processors); - if (namefile == "") { - vector files; files.push_back(fastafile); - parser.getNameFile(files); - } - + if (countfile == "") { + if (namefile == "") { + vector files; files.push_back(fastafile); + parser.getNameFile(files); + } + } } } catch(exception& e) { @@ -155,7 +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; @@ -166,6 +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; } @@ -286,11 +326,12 @@ int SeqSummaryCommand::execute(){ MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case #else vector positions; - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + #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; @@ -323,7 +364,7 @@ int SeqSummaryCommand::execute(){ int size = startPosition.size(); //find means - float meanStartPosition, meanEndPosition, meanSeqLength, meanAmbigBases, meanLongHomoPolymer; + 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]; @@ -332,6 +373,7 @@ int SeqSummaryCommand::execute(){ 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; @@ -359,13 +401,13 @@ int SeqSummaryCommand::execute(){ 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 == "") { m->mothurOut("# of Seqs:\t" + toString(numSeqs)); 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) { 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(); @@ -373,6 +415,13 @@ int SeqSummaryCommand::execute(){ } #endif + //set fasta file as new current fastafile + string current = ""; + itTypes = outputTypes.find("summary"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSummaryFile(current); } + } + return 0; } catch(exception& e) { @@ -399,21 +448,25 @@ int SeqSummaryCommand::driverCreateSummary(vector& startPosition, vectorcontrol_pressed) { in.close(); outSummary.close(); return 1; } - + + if (m->debug) { m->mothurOut("[DEBUG]: count = " + toString(count) + "\n"); } + Sequence current(in); m->gobble(in); - + if (current.getName() != "") { + if (m->debug) { m->mothurOut("[DEBUG]: " + current.getName() + '\t' + toString(current.getNumBases()) + "\n"); } + 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; } } @@ -431,21 +484,18 @@ int SeqSummaryCommand::driverCreateSummary(vector& startPosition, vectordebug) { m->mothurOut("[DEBUG]: " + current.getName() + '\t' + toString(current.getNumBases()) + "\n"); } } - #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) + #if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) unsigned long long pos = in.tellg(); if ((pos == -1) || (pos >= filePos->end)) { break; } #else if (in.eof()) { break; } #endif - - //report progress - //if((count) % 100 == 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); } } - //report progress - //if((count) % 100 != 0){ m->mothurOut(toString(count)); m->mothurOutEndLine(); } - + in.close(); return count; @@ -484,11 +534,11 @@ 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; } } @@ -529,7 +579,7 @@ int SeqSummaryCommand::createProcessesCreateSummary(vector& startPosition, int num = 0; processIDS.clear(); -#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) +#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix) //loop through and create all the processes you want while (process != processors) { @@ -605,14 +655,17 @@ int SeqSummaryCommand::createProcessesCreateSummary(vector& startPosition, 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, namefile, nameMap); + seqSumData* tempSum = new seqSumData(filename, (sumFile+extension), m, lines[i]->start, lines[i]->end, hasNameMap, nameMap); pDataArray.push_back(tempSum); //MySeqSumThreadFunction is in header. It must be global or static to work with the threads. @@ -630,6 +683,14 @@ int SeqSummaryCommand::createProcessesCreateSummary(vector& startPosition, //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]; }