X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=preclustercommand.cpp;h=67b2f31bebe6185fef4137d9c5e839a35b23f0b8;hb=ae57e166b2ed7b475ec3f466106bd76fabadd063;hp=c5b66a81eaa7ec7381af79c4170f1efaf41e91d1;hpb=b45e6f3d505a212cebad5ca22e2611b1466bca48;p=mothur.git diff --git a/preclustercommand.cpp b/preclustercommand.cpp index c5b66a8..67b2f31 100644 --- a/preclustercommand.cpp +++ b/preclustercommand.cpp @@ -8,22 +8,81 @@ */ #include "preclustercommand.h" +#include "sequenceparser.h" +#include "deconvolutecommand.h" //********************************************************************************************************************** inline bool comparePriority(seqPNode first, seqPNode second) { return (first.numIdentical > second.numIdentical); } + +//********************************************************************************************************************** +vector PreClusterCommand::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 pgroup("group", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pgroup); + CommandParameter pdiffs("diffs", "Number", "", "0", "", "", "",false,false); parameters.push_back(pdiffs); + CommandParameter pbygroup("bygroup", "Boolean", "", "T", "", "", "",false,false); parameters.push_back(pbygroup); + 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); } + return myArray; + } + catch(exception& e) { + m->errorOut(e, "PreClusterCommand", "setParameters"); + exit(1); + } +} +//********************************************************************************************************************** +string PreClusterCommand::getHelpString(){ + try { + string helpString = ""; + helpString += "The pre.cluster command groups sequences that are within a given number of base mismatches.\n"; + helpString += "The pre.cluster command outputs a new fasta and name file.\n"; + helpString += "The pre.cluster command parameters are fasta, names and diffs. The fasta parameter is required. \n"; + helpString += "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"; + helpString += "The group parameter allows you to provide a group file. \n"; + helpString += "The diffs parameter allows you to specify maximum number of mismatched bases allowed between sequences in a grouping. The default is 1.\n"; + helpString += "The bygroup parameter allows you to indicate you whether you want to cluster sequences only within each group, default=T, when a groupfile is given. \n"; + helpString += "The pre.cluster command should be in the following format: \n"; + helpString += "pre.cluster(fasta=yourFastaFile, names=yourNamesFile, diffs=yourMaxDiffs) \n"; + helpString += "Example pre.cluster(fasta=amazon.fasta, diffs=2).\n"; + helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n"; + return helpString; + } + catch(exception& e) { + m->errorOut(e, "PreClusterCommand", "getHelpString"); + exit(1); + } +} + +//********************************************************************************************************************** +PreClusterCommand::PreClusterCommand(){ + try { + abort = true; calledHelp = true; + setParameters(); + vector tempOutNames; + outputTypes["fasta"] = tempOutNames; + outputTypes["name"] = tempOutNames; + } + catch(exception& e) { + m->errorOut(e, "PreClusterCommand", "PreClusterCommand"); + exit(1); + } +} //********************************************************************************************************************** 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 if(option == "citation") { citation(); abort = true; calledHelp = true;} else { - //valid paramters for this command - string Array[] = {"fasta", "name", "diffs", "outputdir","inputdir"}; - vector myArray (Array, Array+(sizeof(Array)/sizeof(string))); + vector myArray = setParameters(); OptionParser parser(option); map parameters = parser.getParameters(); @@ -36,6 +95,11 @@ PreClusterCommand::PreClusterCommand(string option) { 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 = ""; } @@ -44,7 +108,7 @@ PreClusterCommand::PreClusterCommand(string option) { it = parameters.find("fasta"); //user has given a template file if(it != parameters.end()){ - path = hasPath(it->second); + 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; } } @@ -52,21 +116,34 @@ PreClusterCommand::PreClusterCommand(string option) { it = parameters.find("name"); //user has given a template file if(it != parameters.end()){ - path = hasPath(it->second); + 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; } } + + it = parameters.find("group"); + //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["group"] = inputDir + it->second; } + } } //check for required parameters fastafile = validParameter.validFile(parameters, "fasta", true); - if (fastafile == "not found") { m->mothurOut("fasta is a required parameter for the pre.cluster command."); m->mothurOutEndLine(); abort = true; } + if (fastafile == "not found") { + 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 if (fastafile == "not open") { abort = true; } + else { m->setFastaFile(fastafile); } //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 += hasPath(fastafile); //if user entered a file with a path then preserve it + outputDir += m->hasPath(fastafile); //if user entered a file with a path then preserve it } //check for optional parameter and set defaults @@ -74,10 +151,25 @@ PreClusterCommand::PreClusterCommand(string option) { namefile = validParameter.validFile(parameters, "name", true); if (namefile == "not found") { namefile = ""; } else if (namefile == "not open") { abort = true; } - else { readNameFile(); } + else { m->setNameFile(namefile); } + + groupfile = validParameter.validFile(parameters, "group", true); + if (groupfile == "not found") { groupfile = ""; } + else if (groupfile == "not open") { abort = true; groupfile = ""; } + else { m->setGroupFile(groupfile); } - string temp = validParameter.validFile(parameters, "diffs", false); if(temp == "not found"){ temp = "1"; } + string temp = validParameter.validFile(parameters, "diffs", false); if(temp == "not found"){ temp = "1"; } convert(temp, diffs); + + temp = validParameter.validFile(parameters, "bygroup", false); + if (temp == "not found") { + if (groupfile != "") { temp = "T"; } + else { temp = "F"; } + } + bygroup = m->isTrue(temp); + + if (bygroup && (groupfile == "")) { m->mothurOut("You cannot set bygroup=T, unless you provide a groupfile."); m->mothurOutEndLine(); abort=true; } + } } @@ -86,117 +178,124 @@ PreClusterCommand::PreClusterCommand(string option) { exit(1); } } - -//********************************************************************************************************************** -PreClusterCommand::~PreClusterCommand(){} -//********************************************************************************************************************** - -void PreClusterCommand::help(){ - try { - 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) { - m->errorOut(e, "PreClusterCommand", "help"); - exit(1); - } -} //********************************************************************************************************************** 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 = readFASTA(); //fills alignSeqs and makes all seqs active - - if (m->control_pressed) { 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; } - - string fileroot = outputDir + getRootName(getSimpleName(fastafile)); - string newFastaFile = fileroot + "precluster" + getExtension(fastafile); + string fileroot = outputDir + m->getRootName(m->getSimpleName(fastafile)); + string newFastaFile = fileroot + "precluster" + m->getExtension(fastafile); string newNamesFile = fileroot + "precluster.names"; - ofstream outFasta; - ofstream outNames; - openOutputFile(newFastaFile, outFasta); - openOutputFile(newNamesFile, outNames); - - //sort seqs by number of identical seqs - alignSeqs.sort(comparePriority); - - int count = 0; - int i = 0; - //think about running through twice... - list::iterator itList; - list::iterator itList2; - for (itList = alignSeqs.begin(); itList != alignSeqs.end();) { + if (bygroup) { + //clear out old files + ofstream outFasta; m->openOutputFile(newFastaFile, outFasta); outFasta.close(); + ofstream outNames; m->openOutputFile(newNamesFile, outNames); outNames.close(); + + //parse fasta and name file by group + SequenceParser* parser; + if (namefile != "") { parser = new SequenceParser(groupfile, fastafile, namefile); } + else { parser = new SequenceParser(groupfile, fastafile); } - //try to merge it with all smaller seqs - for (itList2 = alignSeqs.begin(); itList2 != alignSeqs.end();) { + vector groups = parser->getNamesOfGroups(); + + //precluster each group + for (int i = 0; i < groups.size(); i++) { - if (m->control_pressed) { outFasta.close(); outNames.close(); remove(newFastaFile.c_str()); remove(newNamesFile.c_str()); return 0; } + start = time(NULL); - - if (itList->seq.getName() != itList2->seq.getName()) { //you don't want to merge with yourself - //are you within "diff" bases - - int mismatch = calcMisMatches((*itList).seq.getAligned(), (*itList2).seq.getAligned()); - - if (mismatch <= diffs) { - //merge - (*itList).names += ',' + (*itList2).names; - (*itList).numIdentical += (*itList2).numIdentical; - - itList2 = alignSeqs.erase(itList2); //itList2--; - count++; - }else{ itList2++; } - }else{ itList2++; } + if (m->control_pressed) { delete parser; m->mothurRemove(newFastaFile); m->mothurRemove(newNamesFile); return 0; } + + m->mothurOutEndLine(); m->mothurOut("Processing group " + groups[i] + ":"); m->mothurOutEndLine(); + + map thisNameMap; + if (namefile != "") { thisNameMap = parser->getNameMap(groups[i]); } + vector thisSeqs = parser->getSeqs(groups[i]); + + //fill alignSeqs with this groups info. + int numSeqs = loadSeqs(thisNameMap, thisSeqs); + + if (m->control_pressed) { delete parser; m->mothurRemove(newFastaFile); m->mothurRemove(newNamesFile); return 0; } + + if (diffs > length) { m->mothurOut("Error: diffs is greater than your sequence length."); m->mothurOutEndLine(); return 0; } + + int count = process(); + + if (m->control_pressed) { delete parser; m->mothurRemove(newFastaFile); m->mothurRemove(newNamesFile); 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); + + m->mothurOut("It took " + toString(time(NULL) - start) + " secs to cluster " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); + } - - //ouptut this sequence - printData(outFasta, outNames, (*itList)); - //remove sequence - itList = alignSeqs.erase(itList); + //run unique.seqs for deconvolute results + string inputString = "fasta=" + newFastaFile; + if (namefile != "") { inputString += ", name=" + newNamesFile; } + m->mothurOutEndLine(); + m->mothurOut("/******************************************/"); m->mothurOutEndLine(); + m->mothurOut("Running command: unique.seqs(" + inputString + ")"); m->mothurOutEndLine(); - i++; + Command* uniqueCommand = new DeconvoluteCommand(inputString); + uniqueCommand->execute(); - if(i % 100 == 0) { m->mothurOut(toString(i) + "\t" + toString(numSeqs - count) + "\t" + toString(count)); m->mothurOutEndLine(); } - } - - if(i % 100 != 0) { m->mothurOut(toString(i) + "\t" + toString(numSeqs - count) + "\t" + toString(count)); m->mothurOutEndLine(); } + map > filenames = uniqueCommand->getOutputFiles(); + + delete uniqueCommand; + + m->mothurOut("/******************************************/"); m->mothurOutEndLine(); + + newNamesFile = filenames["name"][0]; + newFastaFile = filenames["fasta"][0]; + + }else { + if (namefile != "") { readNameFile(); } - outFasta.close(); - outNames.close(); + //reads fasta file and return number of seqs + int numSeqs = readFASTA(); //fills alignSeqs and makes all seqs active - if (m->control_pressed) { remove(newFastaFile.c_str()); remove(newNamesFile.c_str()); return 0; } - - m->mothurOut("It took " + toString(time(NULL) - start) + " secs to cluster " + toString(numSeqs) + " sequences."); - m->mothurOut("Total number of sequences before precluster was " + toString(numSeqs) + "."); m->mothurOutEndLine(); - m->mothurOut("pre.cluster removed " + toString(count) + " sequences."); m->mothurOutEndLine(); + if (m->control_pressed) { 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; } + + int count = process(); + + 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); + + 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; } + if (m->control_pressed) { m->mothurRemove(newFastaFile); m->mothurRemove(newNamesFile); return 0; } m->mothurOutEndLine(); m->mothurOut("Output File Names: "); m->mothurOutEndLine(); - m->mothurOut(newFastaFile); m->mothurOutEndLine(); - m->mothurOut(newNamesFile); 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(); - + + //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); } + } + return 0; } @@ -205,18 +304,70 @@ int PreClusterCommand::execute(){ exit(1); } } - /**************************************************************************************************/ - -//this requires the names and fasta file to be in the same order - +int PreClusterCommand::process(){ + try { + + //sort seqs by number of identical seqs + sort(alignSeqs.begin(), alignSeqs.end(), comparePriority); + + int count = 0; + int numSeqs = alignSeqs.size(); + + //think about running through twice... + for (int i = 0; i < numSeqs; i++) { + + //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+1; j < numSeqs; j++) { + + 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; + + }//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(); } + + return count; + + } + catch(exception& e) { + m->errorOut(e, "PreClusterCommand", "process"); + exit(1); + } +} +/**************************************************************************************************/ int PreClusterCommand::readFASTA(){ try { //ifstream inNames; ifstream inFasta; - //openInputFile(namefile, inNames); - openInputFile(fastafile, inFasta); + //m->openInputFile(namefile, inNames); + m->openInputFile(fastafile, inFasta); //string firstCol, secondCol, nameString; length = 0; @@ -228,14 +379,14 @@ int PreClusterCommand::readFASTA(){ //inNames >> firstCol >> secondCol; //nameString = secondCol; - //gobble(inNames); + //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); gobble(inFasta); + Sequence seq(inFasta); m->gobble(inFasta); if (seq.getName() != "") { //can get "" if commented line is at end of fasta file if (namefile != "") { @@ -264,7 +415,55 @@ int PreClusterCommand::readFASTA(){ exit(1); } } - +/**************************************************************************************************/ +int PreClusterCommand::loadSeqs(map& thisName, vector& thisSeqs){ + try { + length = 0; + alignSeqs.clear(); + map::iterator it; + bool error = false; + + for (int i = 0; i < thisSeqs.size(); i++) { + + if (m->control_pressed) { return 0; } + + if (namefile != "") { + it = thisName.find(thisSeqs[i].getName()); + + //should never be true since parser checks for this + if (it == thisName.end()) { m->mothurOut(thisSeqs[i].getName() + " is not in your names file, please correct."); m->mothurOutEndLine(); error = true; } + else{ + //get number of reps + int numReps = 0; + for(int j=0;j<(it->second).length();j++){ + if((it->second)[j] == ','){ numReps++; } + } + + seqPNode tempNode(numReps, thisSeqs[i], it->second); + alignSeqs.push_back(tempNode); + if (thisSeqs[i].getAligned().length() > length) { length = thisSeqs[i].getAligned().length(); } + } + }else { //no names file, you are identical to yourself + seqPNode tempNode(1, thisSeqs[i], thisSeqs[i].getName()); + alignSeqs.push_back(tempNode); + if (thisSeqs[i].getAligned().length() > length) { length = thisSeqs[i].getAligned().length(); } + } + } + + //sanity check + if (error) { m->control_pressed = true; } + + thisSeqs.clear(); + + return alignSeqs.size(); + } + + catch(exception& e) { + m->errorOut(e, "PreClusterCommand", "loadSeqs"); + exit(1); + } +} + /**************************************************************************************************/ int PreClusterCommand::calcMisMatches(string seq1, string seq2){ @@ -287,10 +486,29 @@ int PreClusterCommand::calcMisMatches(string seq1, string seq2){ /**************************************************************************************************/ -void PreClusterCommand::printData(ofstream& outFasta, ofstream& outNames, seqPNode thisSeq){ +void PreClusterCommand::printData(string newfasta, string newname){ try { - thisSeq.seq.printSequence(outFasta); - outNames << thisSeq.seq.getName() << '\t' << thisSeq.names << endl; + ofstream outFasta; + ofstream outNames; + + if (bygroup) { + m->openOutputFileAppend(newfasta, outFasta); + m->openOutputFileAppend(newname, outNames); + }else { + m->openOutputFile(newfasta, outFasta); + m->openOutputFile(newname, outNames); + } + + for (int i = 0; i < alignSeqs.size(); i++) { + if (alignSeqs[i].numIdentical != 0) { + alignSeqs[i].seq.printSequence(outFasta); + outNames << alignSeqs[i].seq.getName() << '\t' << alignSeqs[i].names << endl; + } + } + + outFasta.close(); + outNames.close(); + } catch(exception& e) { m->errorOut(e, "PreClusterCommand", "printData"); @@ -302,11 +520,11 @@ void PreClusterCommand::printData(ofstream& outFasta, ofstream& outNames, seqPNo void PreClusterCommand::readNameFile(){ try { ifstream in; - openInputFile(namefile, in); + m->openInputFile(namefile, in); string firstCol, secondCol; while (!in.eof()) { - in >> firstCol >> secondCol; gobble(in); + in >> firstCol >> secondCol; m->gobble(in); names[firstCol] = secondCol; int size = 1;