From: westcott Date: Tue, 19 Oct 2010 16:51:27 +0000 (+0000) Subject: fixed bug with aligner and longest base. added deunique.seqs command. X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=commitdiff_plain;h=f09bacac01a842a74c1c1b69070852cc969ff506 fixed bug with aligner and longest base. added deunique.seqs command. --- diff --git a/Mothur.xcodeproj/project.pbxproj b/Mothur.xcodeproj/project.pbxproj index 5e84c7e..18269cf 100644 --- a/Mothur.xcodeproj/project.pbxproj +++ b/Mothur.xcodeproj/project.pbxproj @@ -56,6 +56,8 @@ A74A9C03124B72DB000D5D25 /* clusterfragmentscommand.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = clusterfragmentscommand.h; sourceTree = ""; }; A74A9C04124B72DB000D5D25 /* clusterfragmentscommand.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; path = clusterfragmentscommand.cpp; sourceTree = ""; }; A7639F8D1175DF35008F5578 /* makefile */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.make; path = makefile; sourceTree = ""; }; + A76714DD126DE45A003F359A /* deuniqueseqscommand.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = deuniqueseqscommand.h; sourceTree = ""; }; + A76714DE126DE45A003F359A /* deuniqueseqscommand.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; path = deuniqueseqscommand.cpp; sourceTree = ""; }; A768D95D1248FEAA008AB1D0 /* mothur */ = {isa = PBXFileReference; lastKnownFileType = "compiled.mach-o.executable"; path = mothur; sourceTree = ""; }; A76AAD02117F322B003D8DA1 /* phylosummary.h */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.c.h; path = phylosummary.h; sourceTree = ""; }; A76AAD03117F322B003D8DA1 /* phylosummary.cpp */ = {isa = PBXFileReference; fileEncoding = 4; lastKnownFileType = sourcecode.cpp.cpp; path = phylosummary.cpp; sourceTree = ""; }; @@ -774,6 +776,8 @@ A7DA203A113FECD400BF472F /* deconvolutecommand.cpp */, A7DA203F113FECD400BF472F /* distancecommand.h */, A7DA203E113FECD400BF472F /* distancecommand.cpp */, + A76714DD126DE45A003F359A /* deuniqueseqscommand.h */, + A76714DE126DE45A003F359A /* deuniqueseqscommand.cpp */, A7DA2050113FECD400BF472F /* filterseqscommand.h */, A7DA204F113FECD400BF472F /* filterseqscommand.cpp */, A7DA205B113FECD400BF472F /* getgroupcommand.h */, diff --git a/aligncommand.cpp b/aligncommand.cpp index b4af28a..63a117d 100644 --- a/aligncommand.cpp +++ b/aligncommand.cpp @@ -562,10 +562,11 @@ int AlignCommand::driver(linePair* filePos, string alignFName, string reportFNam Sequence temp = templateDB->findClosestSequence(candidateSeq); Sequence* templateSeq = &temp; - + float searchScore = templateDB->getSearchScore(); Nast* nast = new Nast(alignment, candidateSeq, templateSeq); + Sequence* copy; Nast* nast2; @@ -580,6 +581,7 @@ int AlignCommand::driver(linePair* filePos, string alignFName, string reportFNam string wasBetter = ""; //if the user wants you to try the reverse if (flip) { + //get reverse compliment copy = new Sequence(candidateSeq->getName(), originalUnaligned); copy->reverseComplement(); diff --git a/alignmentdb.cpp b/alignmentdb.cpp index e2b1dc1..5a30272 100644 --- a/alignmentdb.cpp +++ b/alignmentdb.cpp @@ -74,7 +74,7 @@ AlignmentDB::AlignmentDB(string fastaFileName, string s, int kmerSize, float gap if (temp.getName() != "") { templateSequences.push_back(temp); //save longest base - if (temp.getUnaligned().length() > longest) { longest = temp.getUnaligned().length()+1; } + if (temp.getUnaligned().length() >= longest) { longest = temp.getUnaligned().length()+1; } } } @@ -94,11 +94,10 @@ AlignmentDB::AlignmentDB(string fastaFileName, string s, int kmerSize, float gap if (temp.getName() != "") { templateSequences.push_back(temp); //save longest base - if (temp.getUnaligned().length() > longest) { longest = temp.getUnaligned().length()+1; } + if (temp.getUnaligned().length() >= longest) { longest = (temp.getUnaligned().length()+1); } } } fastaFile.close(); - #endif numSeqs = templateSequences.size(); diff --git a/commandfactory.cpp b/commandfactory.cpp index 917ab96..a4d6506 100644 --- a/commandfactory.cpp +++ b/commandfactory.cpp @@ -93,6 +93,8 @@ #include "removelineagecommand.h" #include "parsefastaqcommand.h" #include "pipelinepdscommand.h" +#include "deuniqueseqscommand.h" + /*******************************************************/ @@ -191,6 +193,7 @@ CommandFactory::CommandFactory(){ commands["get.lineage"] = "get.lineage"; commands["remove.lineage"] = "remove.lineage"; commands["fastq.info"] = "fastq.info"; + commands["deunique.seqs"] = "deunique.seqs"; commands["pipeline.pds"] = "MPIEnabled"; commands["classify.seqs"] = "MPIEnabled"; commands["dist.seqs"] = "MPIEnabled"; @@ -334,6 +337,7 @@ Command* CommandFactory::getCommand(string commandName, string optionString){ else if(commandName == "remove.lineage") { command = new RemoveLineageCommand(optionString); } else if(commandName == "fastq.info") { command = new ParseFastaQCommand(optionString); } else if(commandName == "pipeline.pds") { command = new PipelineCommand(optionString); } + else if(commandName == "deunique.seqs") { command = new DeUniqueSeqsCommand(optionString); } else { command = new NoCommand(optionString); } return command; @@ -445,6 +449,7 @@ Command* CommandFactory::getCommand(string commandName, string optionString, str else if(commandName == "get.lineage") { pipecommand = new GetLineageCommand(optionString); } else if(commandName == "remove.lineage") { pipecommand = new RemoveLineageCommand(optionString); } else if(commandName == "fastq.info") { pipecommand = new ParseFastaQCommand(optionString); } + else if(commandName == "deunique.seqs") { pipecommand = new DeUniqueSeqsCommand(optionString); } else { pipecommand = new NoCommand(optionString); } return pipecommand; @@ -544,6 +549,7 @@ Command* CommandFactory::getCommand(string commandName){ else if(commandName == "get.lineage") { shellcommand = new GetLineageCommand(); } else if(commandName == "remove.lineage") { shellcommand = new RemoveLineageCommand(); } else if(commandName == "fastq.info") { shellcommand = new ParseFastaQCommand(); } + else if(commandName == "deunique.seqs") { shellcommand = new DeUniqueSeqsCommand(); } else { shellcommand = new NoCommand(); } return shellcommand; diff --git a/deuniqueseqscommand.cpp b/deuniqueseqscommand.cpp new file mode 100644 index 0000000..a4395a7 --- /dev/null +++ b/deuniqueseqscommand.cpp @@ -0,0 +1,260 @@ +/* + * deuniqueseqscommand.cpp + * Mothur + * + * Created by westcott on 10/19/10. + * Copyright 2010 Schloss Lab. All rights reserved. + * + */ + +#include "deuniqueseqscommand.h" +#include "sequence.hpp" + +//********************************************************************************************************************** +vector DeUniqueSeqsCommand::getValidParameters(){ + try { + string Array[] = {"fasta", "name","outputdir","inputdir"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "DeUniqueSeqsCommand", "getValidParameters"); + exit(1); + } +} +//********************************************************************************************************************** +DeUniqueSeqsCommand::DeUniqueSeqsCommand(){ + try { + //initialize outputTypes + vector tempOutNames; + outputTypes["fasta"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "DeUniqueSeqsCommand", "DeconvoluteCommand"); + exit(1); + } +} +//********************************************************************************************************************** +vector DeUniqueSeqsCommand::getRequiredParameters(){ + try { + string Array[] = {"fasta","name"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "DeUniqueSeqsCommand", "getRequiredParameters"); + exit(1); + } +} +//********************************************************************************************************************** +vector DeUniqueSeqsCommand::getRequiredFiles(){ + try { + vector myArray; + return myArray; + } + catch(exception& e) { + m->errorOut(e, "DeUniqueSeqsCommand", "getRequiredFiles"); + exit(1); + } +} +/**************************************************************************************/ +DeUniqueSeqsCommand::DeUniqueSeqsCommand(string option) { + try { + abort = false; + + //allow user to run help + if(option == "help") { help(); abort = true; } + + else { + //valid paramters for this command + string Array[] = {"fasta", "name","outputdir","inputdir"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + + OptionParser parser(option); + map parameters = parser.getParameters(); + + ValidParameters validParameter; + map::iterator it; + + //check to make sure all parameters are valid for command + for (it = parameters.begin(); it != parameters.end(); it++) { + if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; } + } + + //initialize outputTypes + vector tempOutNames; + outputTypes["fasta"] = 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); + if (inputDir == "not found"){ inputDir = ""; } + else { + string path; + it = parameters.find("fasta"); + //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["fasta"] = inputDir + it->second; } + } + + it = parameters.find("name"); + //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["name"] = inputDir + it->second; } + } + } + + + //check for required parameters + fastaFile = validParameter.validFile(parameters, "fasta", true); + if (fastaFile == "not open") { abort = true; } + else if (fastaFile == "not found") { fastaFile = ""; m->mothurOut("fasta is a required parameter for the deunique.seqs command."); 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"){ + outputDir = ""; + outputDir += m->hasPath(fastaFile); //if user entered a file with a path then preserve it + } + + nameFile = validParameter.validFile(parameters, "name", true); + if (nameFile == "not open") { abort = true; } + else if (nameFile == "not found"){ nameFile = ""; m->mothurOut("name is a required parameter for the deunique.seqs command."); m->mothurOutEndLine(); abort = true; } + } + + } + catch(exception& e) { + m->errorOut(e, "DeUniqueSeqsCommand", "DeUniqueSeqsCommand"); + exit(1); + } +} +//********************************************************************************************************************** + +void DeUniqueSeqsCommand::help(){ + try { + m->mothurOut("The deunique.seqs command reads a fastafile and namefile, and creates a fastafile containing all the sequences.\n"); + m->mothurOut("The deunique.seqs command parameters are fasta and name, both are required.\n"); + m->mothurOut("The deunique.seqs command should be in the following format: \n"); + m->mothurOut("deunique.seqs(fasta=yourFastaFile, name=yourNameFile) \n"); + m->mothurOut("Example deunique.seqs(fasta=abrecovery.unique.fasta, name=abrecovery.names).\n"); + m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n"); + + } + catch(exception& e) { + m->errorOut(e, "DeUniqueSeqsCommand", "help"); + exit(1); + } +} + +/**************************************************************************************/ +int DeUniqueSeqsCommand::execute() { + try { + + if (abort == true) { return 0; } + + //prepare filenames and open files + ofstream out; + string outFastaFile = m->getRootName(m->getSimpleName(fastaFile)); + int pos = outFastaFile.find("unique"); + if (pos != string::npos) { + outFastaFile = outputDir + outFastaFile.substr(0, pos) + "redundant" + m->getExtension(fastaFile); + }else{ + outFastaFile = outputDir + m->getRootName(m->getSimpleName(fastaFile)) + "redundant" + m->getExtension(fastaFile); + } + m->openOutputFile(outFastaFile, out); + + readNamesFile(); + if (m->control_pressed) { out.close(); outputTypes.clear(); remove(outFastaFile.c_str()); return 0; } + + ifstream in; + m->openInputFile(fastaFile, in); + + while (!in.eof()) { + + if (m->control_pressed) { in.close(); out.close(); outputTypes.clear(); remove(outFastaFile.c_str()); return 0; } + + Sequence seq(in); m->gobble(in); + + if (seq.getName() != "") { + + //look for sequence name in nameMap + map::iterator it = nameMap.find(seq.getName()); + + if (it == nameMap.end()) { m->mothurOut("[ERROR]: Your namefile does not contain " + seq.getName() + ", aborting."); m->mothurOutEndLine(); m->control_pressed = true; } + else { + vector names; + m->splitAtComma(it->second, names); + + //output sequences + for (int i = 0; i < names.size(); i++) { + out << ">" << names[i] << endl; + out << seq.getAligned() << endl; + } + + //remove seq from name map so we can check for seqs in namefile not in fastafile later + nameMap.erase(it); + } + } + } + in.close(); + out.close(); + + if (nameMap.size() != 0) { //then there are names in the namefile not in the fastafile + for (map::iterator it = nameMap.begin(); it != nameMap.end(); it++) { + m->mothurOut(it->first + " is not in your fasta file, but is in your name file. Please correct."); m->mothurOutEndLine(); + } + } + + if (m->control_pressed) { outputTypes.clear(); remove(outFastaFile.c_str()); return 0; } + + m->mothurOutEndLine(); + m->mothurOut("Output File Names: "); m->mothurOutEndLine(); + m->mothurOut(outFastaFile); m->mothurOutEndLine(); + outputNames.push_back(outFastaFile); outputTypes["fasta"].push_back(outFastaFile); + m->mothurOutEndLine(); + + return 0; + } + catch(exception& e) { + m->errorOut(e, "DeUniqueSeqsCommand", "execute"); + exit(1); + } +} +//********************************************************************************************************************** +int DeUniqueSeqsCommand::readNamesFile() { + try { + + ifstream inNames; + m->openInputFile(nameFile, inNames); + + string name, names; + map::iterator it; + + while(inNames){ + + if(m->control_pressed) { break; } + + inNames >> name; m->gobble(inNames); + inNames >> names; + + it = nameMap.find(name); + + if (it == nameMap.end()) { nameMap[name] = names; } + else { m->mothurOut("[ERROR]: Your namefile already contains " + name + ", aborting."); m->mothurOutEndLine(); m->control_pressed = true; } + + m->gobble(inNames); + } + inNames.close(); + + return 0; + + } + catch(exception& e) { + m->errorOut(e, "DeUniqueSeqsCommand", "readNamesFile"); + exit(1); + } +} + +/**************************************************************************************/ diff --git a/deuniqueseqscommand.h b/deuniqueseqscommand.h new file mode 100644 index 0000000..13176a8 --- /dev/null +++ b/deuniqueseqscommand.h @@ -0,0 +1,43 @@ +#ifndef DEUNIQUESEQSCOMMAND_H +#define DEUNIQUESEQSCOMMAND_H +/* + * deuniqueseqscommand.h + * Mothur + * + * Created by westcott on 10/19/10. + * Copyright 2010 Schloss Lab. All rights reserved. + * + */ + +#include "command.hpp" + +/* This command is the reverse of unique.seqs */ + +class DeUniqueSeqsCommand : public Command { + +public: + DeUniqueSeqsCommand(string); + DeUniqueSeqsCommand(); + ~DeUniqueSeqsCommand() {} + vector getRequiredParameters(); + vector getValidParameters(); + vector getRequiredFiles(); + map > getOutputFiles() { return outputTypes; } + int execute(); + void help(); + +private: + + string fastaFile, nameFile, outputDir; + vector outputNames; + map > outputTypes; + bool abort; + + map nameMap; + + int readNamesFile(); + +}; + +#endif + diff --git a/mothur b/mothur index 4499d7e..6480cd4 100755 Binary files a/mothur and b/mothur differ diff --git a/nast.cpp b/nast.cpp index b288752..28cc659 100644 --- a/nast.cpp +++ b/nast.cpp @@ -25,8 +25,10 @@ Nast::Nast(Alignment* method, Sequence* cand, Sequence* temp) : alignment(method try { m = MothurOut::getInstance(); maxInsertLength = 0; + pairwiseAlignSeqs(); // This is part A in Fig. 2 of DeSantis et al. regapSequences(); // This is parts B-F in Fig. 2 of DeSantis et al. + } catch(exception& e) { m->errorOut(e, "Nast", "Nast"); diff --git a/needlemanoverlap.cpp b/needlemanoverlap.cpp index d13925e..1239beb 100644 --- a/needlemanoverlap.cpp +++ b/needlemanoverlap.cpp @@ -51,18 +51,21 @@ NeedlemanOverlap::~NeedlemanOverlap(){ /* do nothing */ } void NeedlemanOverlap::align(string A, string B){ try { + seqA = ' ' + A; lA = seqA.length(); // algorithm requires a dummy space at the beginning of each string seqB = ' ' + B; lB = seqB.length(); // algorithm requires a dummy space at the beginning of each string if (lA > nRows) { m->mothurOut("One of your candidate sequences is longer than you longest template sequence. Your longest template sequence is " + toString(nRows) + ". Your candidate is " + toString(lA) + "."); m->mothurOutEndLine(); } for(int i=1;igetRootName(m->getSimpleName(sffFile)) + "fasta"; string qualFile = m->getRootName(m->getSimpleName(sffFile)) + "qual"; - thisCommand = "trim.seqs(processors=" + toString(processors) + ", fasta=" + fastaFile + ", allfiles=T, maxambig=0, maxhomop=8, flip=T, bdiffs=1, pdiffs=2, qwindowaverage=35, qwindowsize=50, oligos=" + oligosFile + ", qfile=" + qualFile + ")"; - commands.push_back(thisCommand); + //thisCommand = "trim.seqs(processors=" + toString(processors) + ", fasta=" + fastaFile + ", allfiles=T, maxambig=0, maxhomop=8, flip=T, bdiffs=1, pdiffs=2, qwindowaverage=35, qwindowsize=50, oligos=" + oligosFile + ", qfile=" + qualFile + ")"; + //commands.push_back(thisCommand); //unique.seqs string groupFile = m->getRootName(m->getSimpleName(fastaFile)) + "groups";