X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=makecontigscommand.cpp;h=691d706ab00efa8823412753b72f08e5198873b1;hb=cbaa068e77aeb15bb06f0695a36d8f757977ed64;hp=f672840eda46116a14a917eaf82a1aa9f43e4e56;hpb=e7ae6e6b27c45b5691c19f423ec56faae8e2f255;p=mothur.git diff --git a/makecontigscommand.cpp b/makecontigscommand.cpp index f672840..691d706 100644 --- a/makecontigscommand.cpp +++ b/makecontigscommand.cpp @@ -38,8 +38,8 @@ string MakeContigsCommand::getHelpString(){ string helpString = ""; helpString += "The make.contigs command reads a forward fastq file and a reverse fastq file and outputs new fasta and quality files.\n"; helpString += "The make.contigs command parameters are ffastq, rfastq, align, match, mismatch, gapopen, gapextend and processors.\n"; - helpString += "The ffastq and rfastq parameter is required.\n"; - helpString += "The align parameter allows you to specify the alignment method to use. Your options are: gotoh, needleman, blast and noalign. The default is needleman.\n"; + helpString += "The ffastq and rfastq parameters are required.\n"; + helpString += "The align parameter allows you to specify the alignment method to use. Your options are: gotoh and needleman. The default is needleman.\n"; helpString += "The match parameter allows you to specify the bonus for having the same base. The default is 1.0.\n"; helpString += "The mistmatch parameter allows you to specify the penalty for having different bases. The default is -1.0.\n"; helpString += "The gapopen parameter allows you to specify the penalty for opening a gap in an alignment. The default is -2.0.\n"; @@ -56,7 +56,28 @@ string MakeContigsCommand::getHelpString(){ exit(1); } } - +//********************************************************************************************************************** +string MakeContigsCommand::getOutputFileNameTag(string type, string inputName=""){ + try { + string outputFileName = ""; + map >::iterator it; + + //is this a type this command creates + it = outputTypes.find(type); + if (it == outputTypes.end()) { m->mothurOut("[ERROR]: this command doesn't create a " + type + " output file.\n"); } + else { + if (type == "fasta") { outputFileName = "contigs.fasta"; } + else if (type == "qfile") { outputFileName = "contigs.qual"; } + else if (type == "mismatch") { outputFileName = "contigs.mismatch"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output file tag.\n"); m->control_pressed = true; } + } + return outputFileName; + } + catch(exception& e) { + m->errorOut(e, "MakeContigsCommand", "getOutputFileNameTag"); + exit(1); + } +} //********************************************************************************************************************** MakeContigsCommand::MakeContigsCommand(){ try { @@ -189,9 +210,9 @@ int MakeContigsCommand::execute(){ if (m->control_pressed) { return 0; } - string outFastaFile = outputDir + m->getRootName(m->getSimpleName(ffastqfile)) + "contigs.fasta"; - string outQualFile = outputDir + m->getRootName(m->getSimpleName(ffastqfile)) + "contigs.qual"; - string outMisMatchFile = outputDir + m->getRootName(m->getSimpleName(ffastqfile)) + "contigs.mismatches"; + string outFastaFile = outputDir + m->getRootName(m->getSimpleName(ffastqfile)) + getOutputFileNameTag("fasta"); + string outQualFile = outputDir + m->getRootName(m->getSimpleName(ffastqfile)) + getOutputFileNameTag("qfile"); + string outMisMatchFile = outputDir + m->getRootName(m->getSimpleName(ffastqfile)) + getOutputFileNameTag("mismatch"); outputNames.push_back(outFastaFile); outputTypes["fasta"].push_back(outFastaFile); outputNames.push_back(outQualFile); outputTypes["qfile"].push_back(outQualFile); outputNames.push_back(outMisMatchFile); outputTypes["mismatch"].push_back(outMisMatchFile); @@ -392,10 +413,32 @@ int MakeContigsCommand::driver(vector files, string outputFasta, string int numMismatches = 0; string seq1 = fSeq.getAligned(); string seq2 = rSeq.getAligned(); - vector scores1 = fQual.getQualityScores(); vector scores2 = rQual.getQualityScores(); - for (int i = 0; i < length; i++) { + + // if (num < 5) { cout << fSeq.getStartPos() << '\t' << fSeq.getEndPos() << '\t' << rSeq.getStartPos() << '\t' << rSeq.getEndPos() << endl; } + int overlapStart = fSeq.getStartPos(); + int seq2Start = rSeq.getStartPos(); + //bigger of the 2 starting positions is the location of the overlapping start + if (overlapStart < seq2Start) { //seq2 starts later so take from 0 to seq2Start from seq1 + overlapStart = seq2Start; + for (int i = 0; i < overlapStart; i++) { + contig += seq1[i]; + contigScores.push_back(scores1[ABaseMap[i]]); + } + }else { //seq1 starts later so take from 0 to overlapStart from seq2 + for (int i = 0; i < overlapStart; i++) { + contig += seq2[i]; + contigScores.push_back(scores2[BBaseMap[i]]); + } + } + + int seq1End = fSeq.getEndPos(); + int seq2End = rSeq.getEndPos(); + int overlapEnd = seq1End; + if (seq2End < overlapEnd) { overlapEnd = seq2End; } //smallest end position is where overlapping ends + + for (int i = overlapStart; i < overlapEnd; i++) { if (seq1[i] == seq2[i]) { //match, add base and choose highest score contig += seq1[i]; contigScores.push_back(scores1[ABaseMap[i]]); @@ -423,6 +466,19 @@ int MakeContigsCommand::driver(vector files, string outputFasta, string } } + if (seq1End < seq2End) { //seq1 ends before seq2 so take from overlap to length from seq2 + for (int i = overlapEnd; i < length; i++) { + contig += seq2[i]; + contigScores.push_back(scores2[BBaseMap[i]]); + } + }else { //seq2 ends before seq1 so take from overlap to length from seq1 + for (int i = overlapEnd; i < length; i++) { + contig += seq1[i]; + contigScores.push_back(scores1[ABaseMap[i]]); + } + + } + //if (num < 5) { cout << overlapStart << '\t' << overlapEnd << endl << seq1 << endl << seq2 << endl<< contig << endl; } //output outFasta << ">" << fSeq.getName() << endl << contig << endl; outQual << ">" << fSeq.getName() << endl;