X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=preclustercommand.cpp;h=14409f5a0a6aa86a166f42ff6f99423777237759;hb=6973be461c47c057531f447de22003a30cabc193;hp=479a862935430545757ef48c6da660913782cfa2;hpb=b5a791c81d432082bf38755a08b33863f255341d;p=mothur.git diff --git a/preclustercommand.cpp b/preclustercommand.cpp index 479a862..14409f5 100644 --- a/preclustercommand.cpp +++ b/preclustercommand.cpp @@ -12,34 +12,116 @@ //********************************************************************************************************************** inline bool comparePriority(seqPNode first, seqPNode second) { return (first.numIdentical > second.numIdentical); } //********************************************************************************************************************** +vector PreClusterCommand::getValidParameters(){ + try { + string Array[] = {"fasta", "name", "diffs", "outputdir","inputdir"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "PreClusterCommand", "getValidParameters"); + exit(1); + } +} +//********************************************************************************************************************** +PreClusterCommand::PreClusterCommand(){ + try { + abort = true; calledHelp = true; + vector tempOutNames; + outputTypes["fasta"] = tempOutNames; + outputTypes["name"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "PreClusterCommand", "PreClusterCommand"); + exit(1); + } +} +//********************************************************************************************************************** +vector PreClusterCommand::getRequiredParameters(){ + try { + string Array[] = {"fasta"}; + vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + return myArray; + } + catch(exception& e) { + m->errorOut(e, "PreClusterCommand", "getRequiredParameters"); + exit(1); + } +} +//********************************************************************************************************************** +vector PreClusterCommand::getRequiredFiles(){ + try { + vector myArray; + return myArray; + } + catch(exception& e) { + m->errorOut(e, "PreClusterCommand", "getRequiredFiles"); + exit(1); + } +} +//********************************************************************************************************************** -PreClusterCommand::PreClusterCommand(string option){ +PreClusterCommand::PreClusterCommand(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 - string Array[] = {"fasta", "name", "diffs"}; + string Array[] = {"fasta", "name", "diffs", "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 (map::iterator it2 = parameters.begin(); it2 != parameters.end(); it2++) { if (validParameter.isValidParameter(it2->first, myArray, it2->second) != true) { abort = true; } } + //initialize outputTypes + vector tempOutNames; + outputTypes["fasta"] = tempOutNames; + outputTypes["name"] = 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 found") { mothurOut("fasta is a required parameter for the pre.cluster command."); mothurOutEndLine(); abort = true; } + if (fastafile == "not found") { m->mothurOut("fasta is a required parameter for the pre.cluster command."); m->mothurOutEndLine(); abort = true; } else if (fastafile == "not open") { 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 + } + //check for optional parameter and set defaults // ...at some point should added some additional type checking... namefile = validParameter.validFile(parameters, "name", true); @@ -53,7 +135,7 @@ PreClusterCommand::PreClusterCommand(string option){ } catch(exception& e) { - errorOut(e, "PreClusterCommand", "PreClusterCommand"); + m->errorOut(e, "PreClusterCommand", "PreClusterCommand"); exit(1); } } @@ -64,18 +146,18 @@ PreClusterCommand::~PreClusterCommand(){} void PreClusterCommand::help(){ try { - mothurOut("The pre.cluster command groups sequences that are within a given number of base mismatches.\n"); - mothurOut("The pre.cluster command outputs a new fasta and name file.\n"); - mothurOut("The pre.cluster command parameters are fasta, names and diffs. The fasta parameter is required. \n"); - mothurOut("The names parameter allows you to give a list of seqs that are identical. This file is 2 columns, first column is name or representative sequence, second column is a list of its identical sequences separated by commas.\n"); - mothurOut("The diffs parameter allows you to specify maximum number of mismatched bases allowed between sequences in a grouping. The default is 1.\n"); - mothurOut("The pre.cluster command should be in the following format: \n"); - mothurOut("pre.cluster(fasta=yourFastaFile, names=yourNamesFile, diffs=yourMaxDiffs) \n"); - mothurOut("Example pre.cluster(fasta=amazon.fasta, diffs=2).\n"); - mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n"); + m->mothurOut("The pre.cluster command groups sequences that are within a given number of base mismatches.\n"); + m->mothurOut("The pre.cluster command outputs a new fasta and name file.\n"); + m->mothurOut("The pre.cluster command parameters are fasta, names and diffs. The fasta parameter is required. \n"); + m->mothurOut("The names parameter allows you to give a list of seqs that are identical. This file is 2 columns, first column is name or representative sequence, second column is a list of its identical sequences separated by commas.\n"); + m->mothurOut("The diffs parameter allows you to specify maximum number of mismatched bases allowed between sequences in a grouping. The default is 1.\n"); + m->mothurOut("The pre.cluster command should be in the following format: \n"); + m->mothurOut("pre.cluster(fasta=yourFastaFile, names=yourNamesFile, diffs=yourMaxDiffs) \n"); + m->mothurOut("Example pre.cluster(fasta=amazon.fasta, diffs=2).\n"); + m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n\n"); } catch(exception& e) { - errorOut(e, "PreClusterCommand", "help"); + m->errorOut(e, "PreClusterCommand", "help"); exit(1); } } @@ -84,139 +166,156 @@ void PreClusterCommand::help(){ int PreClusterCommand::execute(){ try { - if (abort == true) { return 0; } + if (abort == true) { if (calledHelp) { return 0; } return 2; } + + int start = time(NULL); //reads fasta file and return number of seqs - int numSeqs = readSeqs(); //fills alignSeqs and makes all seqs active + int numSeqs = readFASTA(); //fills alignSeqs and makes all seqs active + + if (m->control_pressed) { return 0; } - if (numSeqs == 0) { mothurOut("Error reading fasta file...please correct."); mothurOutEndLine(); return 0; } - if (diffs > length) { mothurOut("Error: diffs is greater than your sequence length."); mothurOutEndLine(); return 0; } + if (numSeqs == 0) { m->mothurOut("Error reading fasta file...please correct."); m->mothurOutEndLine(); return 0; } + if (diffs > length) { m->mothurOut("Error: diffs is greater than your sequence length."); m->mothurOutEndLine(); return 0; } //clear sizes since you only needed this info to build the alignSeqs seqPNode structs - sizes.clear(); +// sizes.clear(); //sort seqs by number of identical seqs sort(alignSeqs.begin(), alignSeqs.end(), comparePriority); - //go through active list and cluster everthing you can, until all nodes are inactive - //taking advantage of the fact that maps are already sorted - map::iterator itActive; - map::iterator it2Active; int count = 0; - - for (int i = 0; i < alignSeqs.size(); i++) { - - //are you active - itActive = active.find(alignSeqs[i].seq.getName()); + + //think about running through twice... + for (int i = 0; i < numSeqs; i++) { - if (itActive != active.end()) { //this sequence has not been merged yet + //are you active + // itActive = active.find(alignSeqs[i].seq.getName()); + if (alignSeqs[i].active) { //this sequence has not been merged yet + //try to merge it with all smaller seqs - for (int j = i; j < alignSeqs.size(); j++) { + for (int j = i+1; j < numSeqs; j++) { - if (i != j) { - //are you active - it2Active = active.find(alignSeqs[j].seq.getName()); - if (it2Active != active.end()) { //this sequence has not been merged yet - //are you within "diff" bases - int mismatch = calcMisMatches(alignSeqs[i].seq.getAligned(), alignSeqs[j].seq.getAligned()); - - if (mismatch <= diffs) { - //merge - names[alignSeqs[i].seq.getName()] += "," + names[alignSeqs[j].seq.getName()]; - - //remove from active list - active.erase(it2Active); - - //set numIdentical to 0, so you only print the representative seqs in the fasta file - alignSeqs[j].numIdentical = 0; - count++; - } - }//end if j active - }//end if i != j - }//end for loop + if (m->control_pressed) { return 0; } + + if (alignSeqs[j].active) { //this sequence has not been merged yet + //are you within "diff" bases + int mismatch = calcMisMatches(alignSeqs[i].seq.getAligned(), alignSeqs[j].seq.getAligned()); + + if (mismatch <= diffs) { + //merge + alignSeqs[i].names += ',' + alignSeqs[j].names; + alignSeqs[i].numIdentical += alignSeqs[j].numIdentical; + + alignSeqs[j].active = 0; + alignSeqs[j].numIdentical = 0; + count++; + } + }//end if j active + }//end if i != j + + //remove from active list + alignSeqs[i].active = 0; - //remove from active list - active.erase(itActive); }//end if active i + if(i % 100 == 0) { m->mothurOut(toString(i) + "\t" + toString(numSeqs - count) + "\t" + toString(count)); m->mothurOutEndLine(); } } + + if(numSeqs % 100 != 0) { m->mothurOut(toString(numSeqs) + "\t" + toString(numSeqs - count) + "\t" + toString(count)); m->mothurOutEndLine(); } - string newFastaFile = getRootName(fastafile) + "precluster" + getExtension(fastafile); - string newNamesFile = getRootName(fastafile) + "precluster.names"; + string fileroot = outputDir + m->getRootName(m->getSimpleName(fastafile)); + + string newFastaFile = fileroot + "precluster" + m->getExtension(fastafile); + string newNamesFile = fileroot + "precluster.names"; + + if (m->control_pressed) { return 0; } + + m->mothurOut("Total number of sequences before precluster was " + toString(alignSeqs.size()) + "."); m->mothurOutEndLine(); + m->mothurOut("pre.cluster removed " + toString(count) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine(); printData(newFastaFile, newNamesFile); - mothurOut("Total number of sequences before precluster was " + toString(alignSeqs.size()) + "."); mothurOutEndLine(); - mothurOut("pre.cluster removed " + toString(count) + " sequences."); mothurOutEndLine(); + m->mothurOut("It took " + toString(time(NULL) - start) + " secs to cluster " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); + if (m->control_pressed) { remove(newFastaFile.c_str()); remove(newNamesFile.c_str()); return 0; } + + m->mothurOutEndLine(); + m->mothurOut("Output File Names: "); m->mothurOutEndLine(); + m->mothurOut(newFastaFile); m->mothurOutEndLine(); outputNames.push_back(newFastaFile); outputTypes["fasta"].push_back(newFastaFile); + m->mothurOut(newNamesFile); m->mothurOutEndLine(); outputNames.push_back(newNamesFile); outputTypes["name"].push_back(newNamesFile); + m->mothurOutEndLine(); + return 0; } catch(exception& e) { - errorOut(e, "PreClusterCommand", "execute"); + m->errorOut(e, "PreClusterCommand", "execute"); exit(1); } } + /**************************************************************************************************/ -int PreClusterCommand::readSeqs(){ + +//this requires the names and fasta file to be in the same order + +int PreClusterCommand::readFASTA(){ try { + //ifstream inNames; ifstream inFasta; - openInputFile(fastafile, inFasta); + + //m->openInputFile(namefile, inNames); + m->openInputFile(fastafile, inFasta); + + //string firstCol, secondCol, nameString; length = 0; - map::iterator it; - + while (!inFasta.eof()) { - Sequence temp(inFasta); //read seq - if (temp.getName() != "") { + if (m->control_pressed) { inFasta.close(); return 0; } + + //inNames >> firstCol >> secondCol; + //nameString = secondCol; + + //m->gobble(inNames); + //int size = 1; + //while (secondCol.find_first_of(',') != -1) { + // size++; + // secondCol = secondCol.substr(secondCol.find_first_of(',')+1, secondCol.length()); + //} + + Sequence seq(inFasta); m->gobble(inFasta); + + if (seq.getName() != "") { //can get "" if commented line is at end of fasta file if (namefile != "") { - //make sure fasta and name files match - it = names.find(temp.getName()); - if (it == names.end()) { mothurOut(temp.getName() + " is not in your names file, please correct."); mothurOutEndLine(); exit(1); } - }else { sizes[temp.getName()] = 1; } - - seqPNode tempNode(sizes[temp.getName()], temp); - alignSeqs.push_back(tempNode); - active[temp.getName()] = true; + itSize = sizes.find(seq.getName()); + + if (itSize == sizes.end()) { m->mothurOut(seq.getName() + " is not in your names file, please correct."); m->mothurOutEndLine(); exit(1); } + else{ + seqPNode tempNode(itSize->second, seq, names[seq.getName()]); + alignSeqs.push_back(tempNode); + if (seq.getAligned().length() > length) { length = seq.getAligned().length(); } + } + }else { //no names file, you are identical to yourself + seqPNode tempNode(1, seq, seq.getName()); + alignSeqs.push_back(tempNode); + if (seq.getAligned().length() > length) { length = seq.getAligned().length(); } + } } - gobble(inFasta); } inFasta.close(); - - if (alignSeqs.size() != 0) { length = alignSeqs[0].seq.getAligned().length(); } - + //inNames.close(); return alignSeqs.size(); } + catch(exception& e) { - errorOut(e, "PreClusterCommand", "readSeqs"); - exit(1); - } -} -/**************************************************************************************************/ -void PreClusterCommand::readNameFile(){ - try { - ifstream in; - openInputFile(namefile, in); - string firstCol, secondCol; - - while (!in.eof()) { - in >> firstCol >> secondCol; gobble(in); - names[firstCol] = secondCol; - int size = 1; - while (secondCol.find_first_of(',') != -1) { - size++; - secondCol = secondCol.substr(secondCol.find_first_of(',')+1, secondCol.length()); - } - sizes[firstCol] = size; - } - in.close(); - } - catch(exception& e) { - errorOut(e, "PreClusterCommand", "readNameFile"); + m->errorOut(e, "PreClusterCommand", "readFASTA"); exit(1); } } + /**************************************************************************************************/ + int PreClusterCommand::calcMisMatches(string seq1, string seq2){ try { int numBad = 0; @@ -230,27 +329,26 @@ int PreClusterCommand::calcMisMatches(string seq1, string seq2){ return numBad; } catch(exception& e) { - errorOut(e, "PreClusterCommand", "calcMisMatches"); + m->errorOut(e, "PreClusterCommand", "calcMisMatches"); exit(1); } } + /**************************************************************************************************/ + void PreClusterCommand::printData(string newfasta, string newname){ try { ofstream outFasta; ofstream outNames; - openOutputFile(newfasta, outFasta); - openOutputFile(newname, outNames); + + m->openOutputFile(newfasta, outFasta); + m->openOutputFile(newname, outNames); - map::iterator itNames; for (int i = 0; i < alignSeqs.size(); i++) { if (alignSeqs[i].numIdentical != 0) { alignSeqs[i].seq.printSequence(outFasta); - - itNames = names.find(alignSeqs[i].seq.getName()); - - outNames << itNames->first << '\t' << itNames->second << endl; + outNames << alignSeqs[i].seq.getName() << '\t' << alignSeqs[i].names << endl; } } @@ -259,7 +357,32 @@ void PreClusterCommand::printData(string newfasta, string newname){ } catch(exception& e) { - errorOut(e, "PreClusterCommand", "printData"); + m->errorOut(e, "PreClusterCommand", "printData"); + exit(1); + } +} +/**************************************************************************************************/ + +void PreClusterCommand::readNameFile(){ + try { + ifstream in; + m->openInputFile(namefile, in); + string firstCol, secondCol; + + while (!in.eof()) { + in >> firstCol >> secondCol; m->gobble(in); + names[firstCol] = secondCol; + int size = 1; + + for(int i=0;ierrorOut(e, "PreClusterCommand", "readNameFile"); exit(1); } }