X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=preclustercommand.cpp;h=0275648066b99c55533e4aaacdcde5bb21edad5f;hb=51d85e945d26324e50d6406829ce006ccc9eb1c0;hp=1697ff7a0d57534b81e9b1f135cf494ae55c2d34;hpb=6777912b79eb6d011163da47f3ada09d1a9293c1;p=mothur.git diff --git a/preclustercommand.cpp b/preclustercommand.cpp index 1697ff7..0275648 100644 --- a/preclustercommand.cpp +++ b/preclustercommand.cpp @@ -13,7 +13,7 @@ inline bool comparePriority(seqPNode first, seqPNode second) { return (first.numIdentical > second.numIdentical); } //********************************************************************************************************************** -PreClusterCommand::PreClusterCommand(string option){ +PreClusterCommand::PreClusterCommand(string option) { try { abort = false; @@ -60,7 +60,7 @@ PreClusterCommand::PreClusterCommand(string option){ //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 @@ -74,7 +74,7 @@ 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 { readNameFile(); } string temp = validParameter.validFile(parameters, "diffs", false); if(temp == "not found"){ temp = "1"; } convert(temp, diffs); @@ -82,7 +82,7 @@ PreClusterCommand::PreClusterCommand(string option){ } catch(exception& e) { - errorOut(e, "PreClusterCommand", "PreClusterCommand"); + m->errorOut(e, "PreClusterCommand", "PreClusterCommand"); exit(1); } } @@ -93,18 +93,18 @@ 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); } } @@ -116,144 +116,146 @@ int PreClusterCommand::execute(){ if (abort == true) { return 0; } //reads fasta file and return number of seqs - int numSeqs = readNamesFASTA(); //fills alignSeqs and makes all seqs active - - 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; } + int numSeqs = readFASTA(); //fills alignSeqs and makes all seqs active - //clear sizes since you only needed this info to build the alignSeqs seqPNode structs -// sizes.clear(); + 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 newNamesFile = fileroot + "precluster.names"; + ofstream outFasta; + ofstream outNames; + + openOutputFile(newFastaFile, outFasta); + openOutputFile(newNamesFile, outNames); + //sort seqs by number of identical seqs - sort(alignSeqs.begin(), alignSeqs.end(), comparePriority); + alignSeqs.sort(comparePriority); int count = 0; + int i = 0; //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 (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()); + list::iterator itList; + list::iterator itList2; + for (itList = alignSeqs.begin(); itList != alignSeqs.end(); itList++) { + //try to merge it with all smaller seqs + for (itList2 = alignSeqs.begin(); itList2 != alignSeqs.end(); itList2++) { + + if (m->control_pressed) { outFasta.close(); outNames.close(); remove(newFastaFile.c_str()); remove(newNamesFile.c_str()); return 0; } + + 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 - alignSeqs[i].names += ',' + alignSeqs[j].names; - alignSeqs[i].numIdentical += alignSeqs[j].numIdentical; + if (mismatch <= diffs) { + //merge + (*itList).names += ',' + (*itList2).names; + (*itList).numIdentical += (*itList2).numIdentical; - alignSeqs[j].active = 0; - alignSeqs[j].numIdentical = 0; - count++; - } - }//end if j active - }//end if i != j + alignSeqs.erase(itList2); itList2--; + count++; + } + } + } + + //ouptut this sequence + printData(outFasta, outNames, (*itList)); + + //remove sequence + alignSeqs.erase(itList); itList--; + + i++; - //remove from active list - alignSeqs[i].active = 0; - }//end if active i - if(i % 100 == 0) { cout << i << '\t' << numSeqs - count << '\t' << count << endl; } + 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(); } - string fileroot = outputDir + getRootName(getSimpleName(fastafile)); + outFasta.close(); + outNames.close(); - string newFastaFile = fileroot + "precluster" + getExtension(fastafile); - string newNamesFile = fileroot + "precluster.names"; + if (m->control_pressed) { remove(newFastaFile.c_str()); remove(newNamesFile.c_str()); return 0; } + + m->mothurOut("Total number of sequences before precluster was " + toString(numSeqs) + "."); m->mothurOutEndLine(); + m->mothurOut("pre.cluster removed " + toString(count) + " sequences."); m->mothurOutEndLine(); - mothurOut("Total number of sequences before precluster was " + toString(alignSeqs.size()) + "."); mothurOutEndLine(); - mothurOut("pre.cluster removed " + toString(count) + " sequences."); mothurOutEndLine(); - printData(newFastaFile, newNamesFile); + 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(); + m->mothurOut(newNamesFile); m->mothurOutEndLine(); + m->mothurOutEndLine(); return 0; } catch(exception& e) { - errorOut(e, "PreClusterCommand", "execute"); + m->errorOut(e, "PreClusterCommand", "execute"); exit(1); } } + /**************************************************************************************************/ + +//this requires the names and fasta file to be in the same order + int PreClusterCommand::readFASTA(){ try { -// ifstream inFasta; -// openInputFile(fastafile, inFasta); -// length = 0; -//// map::iterator it; -// -// while (!inFasta.eof()) { -// Sequence temp(inFasta); //read seq -// -// if (temp.getName() != "") { -// 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; -// } -// gobble(inFasta); -// } -// inFasta.close(); -// -// if (alignSeqs.size() != 0) { length = alignSeqs[0].seq.getAligned().length(); } -// - return alignSeqs.size(); - } - catch(exception& e) { - errorOut(e, "PreClusterCommand", "readFASTA"); - exit(1); - } -} -/**************************************************************************************************/ -//this seems to require the names and fasta file to be in the same order??? -int PreClusterCommand::readNamesFASTA(){ - try { - ifstream inNames; + //ifstream inNames; ifstream inFasta; - openInputFile(namefile, inNames); + //openInputFile(namefile, inNames); openInputFile(fastafile, inFasta); - string firstCol, secondCol, nameString; + //string firstCol, secondCol, nameString; length = 0; - while (inFasta && inNames) { - - inNames >> firstCol >> secondCol; - nameString = secondCol; + while (!inFasta.eof()) { - gobble(inNames); - int size = 1; - while (secondCol.find_first_of(',') != -1) { - size++; - secondCol = secondCol.substr(secondCol.find_first_of(',')+1, secondCol.length()); - } + if (m->control_pressed) { inFasta.close(); return 0; } + + //inNames >> firstCol >> secondCol; + //nameString = secondCol; + + //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); - if (seq.getName() != firstCol) { mothurOut(seq.getName() + " is not in your names file, please correct."); mothurOutEndLine(); exit(1); } - else{ - seqPNode tempNode(size, seq, nameString); - alignSeqs.push_back(tempNode); - if (seq.getAligned().length() > length) { length = alignSeqs[0].seq.getAligned().length(); } - } + + if (seq.getName() != "") { //can get "" if commented line is at end of fasta file + if (namefile != "") { + 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(); } + } + } } inFasta.close(); - inNames.close(); + //inNames.close(); return alignSeqs.size(); } catch(exception& e) { - errorOut(e, "PreClusterCommand", "readNamesFASTA"); + m->errorOut(e, "PreClusterCommand", "readFASTA"); exit(1); } } @@ -273,35 +275,45 @@ 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){ +void PreClusterCommand::printData(ofstream& outFasta, ofstream& outNames, seqPNode thisSeq){ try { - ofstream outFasta; - ofstream outNames; - - openOutputFile(newfasta, outFasta); - openOutputFile(newname, outNames); + thisSeq.seq.printSequence(outFasta); + outNames << thisSeq.seq.getName() << '\t' << thisSeq.names << endl; + } + catch(exception& e) { + m->errorOut(e, "PreClusterCommand", "printData"); + exit(1); + } +} +/**************************************************************************************************/ + +void PreClusterCommand::readNameFile(){ + try { + ifstream in; + openInputFile(namefile, in); + string firstCol, secondCol; - - 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; + 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; } - - outFasta.close(); - outNames.close(); - + in.close(); } catch(exception& e) { - errorOut(e, "PreClusterCommand", "printData"); + m->errorOut(e, "PreClusterCommand", "readNameFile"); exit(1); } }