X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=makebiomcommand.cpp;h=7cf8295fb48e02b0dce58b32432d39c736695259;hp=68e70ee3e48e58a89f75ef7547d7409620947cb2;hb=d1c97b8c04bb75faca1e76ffad60b37a4d789d3d;hpb=90708fe9701e3827e477c82fb3652539c3bf2a0d diff --git a/makebiomcommand.cpp b/makebiomcommand.cpp index 68e70ee..7cf8295 100644 --- a/makebiomcommand.cpp +++ b/makebiomcommand.cpp @@ -9,6 +9,7 @@ #include "makebiomcommand.h" #include "sharedrabundvector.h" #include "inputdata.h" +#include "sharedutilities.h" //taken from http://biom-format.org/documentation/biom_format.html /* Minimal Sparse @@ -91,13 +92,16 @@ //********************************************************************************************************************** vector MakeBiomCommand::setParameters(){ try { - CommandParameter pshared("shared", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pshared); - CommandParameter pcontaxonomy("contaxonomy", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pcontaxonomy); - CommandParameter pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups); - CommandParameter plabel("label", "String", "", "", "", "", "",false,false); parameters.push_back(plabel); - CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir); - CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir); - CommandParameter pmatrixtype("matrixtype", "Multiple", "sparse-dense", "sparse", "", "", "",false,false); parameters.push_back(pmatrixtype); + CommandParameter pshared("shared", "InputTypes", "", "", "none", "none", "none","biom",false,true,true); parameters.push_back(pshared); + CommandParameter pcontaxonomy("constaxonomy", "InputTypes", "", "", "none", "none", "none","",false,false); parameters.push_back(pcontaxonomy); + //CommandParameter preference("referencetax", "InputTypes", "", "", "none", "none", "none","",false,false); parameters.push_back(preference); + CommandParameter pmetadata("metadata", "InputTypes", "", "", "none", "none", "none","",false,false); parameters.push_back(pmetadata); + CommandParameter pgroups("groups", "String", "", "", "", "", "","",false,false); parameters.push_back(pgroups); + CommandParameter plabel("label", "String", "", "", "", "", "","",false,false); parameters.push_back(plabel); + //CommandParameter ppicrust("picrust", "Boolean", "", "F", "", "", "","",false,false); parameters.push_back(ppicrust); + CommandParameter pinputdir("inputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(pinputdir); + CommandParameter poutputdir("outputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(poutputdir); + CommandParameter pmatrixtype("matrixtype", "Multiple", "sparse-dense", "sparse", "", "", "","",false,false); parameters.push_back(pmatrixtype); vector myArray; for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); } @@ -112,11 +116,14 @@ vector MakeBiomCommand::setParameters(){ string MakeBiomCommand::getHelpString(){ try { string helpString = ""; - helpString += "The make.biom command parameters are shared, contaxonomy, groups, matrixtype and label. shared is required, unless you have a valid current file.\n"; + helpString += "The make.biom command parameters are shared, contaxonomy, metadata, groups, matrixtype and label. shared is required, unless you have a valid current file.\n"; //, picrust and referencetax helpString += "The groups parameter allows you to specify which of the groups in your groupfile you would like included. The group names are separated by dashes.\n"; helpString += "The label parameter allows you to select what distance levels you would like, and are also separated by dashes.\n"; helpString += "The matrixtype parameter allows you to select what type you would like to make. Choices are sparse and dense, default is sparse.\n"; - helpString += "The contaxonomy file is the taxonomy file outputted by classify.otu(list=yourListfile, taxonomy=yourTaxonomyFile). Be SURE that the you are the constaxonomy file distance matches the shared file distance. ie, for *.0.03.cons.taxonomy set label=0.03. Mothur is smart enough to handle shared files that have been subsampled.\n"; + helpString += "The contaxonomy file is the taxonomy file outputted by classify.otu(list=yourListfile, taxonomy=yourTaxonomyFile). Be SURE that the you are the constaxonomy file distance matches the shared file distance. ie, for *.0.03.cons.taxonomy set label=0.03. Mothur is smart enough to handle shared files that have been subsampled. It is used to assign taxonomy information to the metadata of rows.\n"; + helpString += "The metadata parameter is used to provide experimental parameters to the columns. Things like 'sample1 gut human_gut'. \n"; + //helpString += "The picrust parameter is used to indicate the biom file is for input to picrust. NOTE: Picrust requires a greengenes taxonomy. \n"; + //helpString += "The referencetax parameter is used with the picrust parameter. Picrust requires the name of the reference taxonomy sequence to be in the biom file. \n"; helpString += "The make.biom command should be in the following format: make.biom(shared=yourShared, groups=yourGroups, label=yourLabels).\n"; helpString += "Example make.biom(shared=abrecovery.an.shared, groups=A-B-C).\n"; helpString += "The default value for groups is all the groups in your groupfile, and all labels in your inputfile will be used.\n"; @@ -130,24 +137,19 @@ string MakeBiomCommand::getHelpString(){ } } //********************************************************************************************************************** -string MakeBiomCommand::getOutputFileNameTag(string type, string inputName=""){ - try { - string outputFileName = ""; - map >::iterator it; +string MakeBiomCommand::getOutputPattern(string type) { + try { + string pattern = ""; - //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 == "biom") { outputFileName = "biom"; } - 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, "MakeBiomCommand", "getOutputFileNameTag"); - exit(1); - } + if (type == "biom") { pattern = "[filename],[distance],biom"; } + else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; } + + return pattern; + } + catch(exception& e) { + m->errorOut(e, "MakeBiomCommand", "getOutputPattern"); + exit(1); + } } //********************************************************************************************************************** @@ -204,12 +206,28 @@ MakeBiomCommand::MakeBiomCommand(string option) { if (path == "") { parameters["shared"] = inputDir + it->second; } } - it = parameters.find("contaxonomy"); + it = parameters.find("constaxonomy"); //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["contaxonomy"] = inputDir + it->second; } + if (path == "") { parameters["constaxonomy"] = inputDir + it->second; } + } + + it = parameters.find("referencetax"); + //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["referencetax"] = inputDir + it->second; } + } + + it = parameters.find("metadata"); + //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["metadata"] = inputDir + it->second; } } } @@ -227,10 +245,17 @@ MakeBiomCommand::MakeBiomCommand(string option) { //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 = m->hasPath(sharedfile); } - contaxonomyfile = validParameter.validFile(parameters, "contaxonomy", true); + contaxonomyfile = validParameter.validFile(parameters, "constaxonomy", true); if (contaxonomyfile == "not found") { contaxonomyfile = ""; } else if (contaxonomyfile == "not open") { contaxonomyfile = ""; abort = true; } + + //referenceTax = validParameter.validFile(parameters, "referencetax", true); + //if (referenceTax == "not found") { referenceTax = ""; } + //else if (referenceTax == "not open") { referenceTax = ""; abort = true; } + metadatafile = validParameter.validFile(parameters, "metadata", true); + if (metadatafile == "not found") { metadatafile = ""; } + else if (metadatafile == "not open") { metadatafile = ""; abort = true; } //check for optional parameter and set defaults // ...at some point should added some additional type checking... @@ -240,6 +265,13 @@ MakeBiomCommand::MakeBiomCommand(string option) { if(label != "all") { m->splitAtDash(label, labels); allLines = 0; } else { allLines = 1; } } + + //string temp = validParameter.validFile(parameters, "picrust", false); if (temp == "not found"){ temp = "f"; } + //picrust = m->isTrue(temp); + //if (picrust && ((contaxonomyfile == "") || (referenceTax == ""))) { + //m->mothurOut("[ERROR]: the picrust parameter requires a consensus taxonomy with greengenes taxonomy the reference."); m->mothurOutEndLine(); abort = true; + //} + picrust=false; groups = validParameter.validFile(parameters, "groups", false); if (groups == "not found") { groups = ""; } @@ -280,6 +312,8 @@ int MakeBiomCommand::execute(){ labels.insert(lastLabel); } + getSampleMetaData(lookup); + //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label. set processedLabels; set userLabels = labels; @@ -372,8 +406,10 @@ int MakeBiomCommand::execute(){ //********************************************************************************************************************** int MakeBiomCommand::getBiom(vector& lookup){ try { - - string outputFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + lookup[0]->getLabel() + "." + getOutputFileNameTag("biom"); + map variables; + variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(sharedfile)); + variables["[distance]"] = lookup[0]->getLabel(); + string outputFileName = getOutputFileName("biom",variables); ofstream out; m->openOutputFile(outputFileName, out); outputNames.push_back(outputFileName); outputTypes["biom"].push_back(outputFileName); @@ -392,7 +428,9 @@ int MakeBiomCommand::getBiom(vector& lookup){ out << "{\n" + spaces + "\"id\":\"" + sharedfile + "-" + lookup[0]->getLabel() + "\",\n" + spaces + "\"format\": \"Biological Observation Matrix 0.9.1\",\n" + spaces + "\"format_url\": \"http://biom-format.org\",\n"; out << spaces + "\"type\": \"OTU table\",\n" + spaces + "\"generated_by\": \"" << mothurString << "\",\n" + spaces + "\"date\": \"" << dateString << "\",\n"; - vector metadata = getMetaData(lookup); + int numBins = lookup[0]->getNumBins(); + vector picrustLabels; + vector metadata = getMetaData(lookup, picrustLabels); if (m->control_pressed) { out.close(); return 0; } @@ -407,12 +445,13 @@ int MakeBiomCommand::getBiom(vector& lookup){ out << spaces + "\"rows\":[\n"; string rowFront = spaces + spaces + "{\"id\":\""; string rowBack = "\", \"metadata\":"; - for (int i = 0; i < m->currentBinLabels.size()-1; i++) { + for (int i = 0; i < numBins-1; i++) { if (m->control_pressed) { out.close(); return 0; } - out << rowFront << m->currentBinLabels[i] << rowBack << metadata[i] << "},\n"; + if (!picrust) { out << rowFront << m->currentBinLabels[i] << rowBack << metadata[i] << "},\n"; } + else { out << rowFront << picrustLabels[i] << rowBack << metadata[i] << "},\n"; } } - out << rowFront << m->currentBinLabels[(m->currentBinLabels.size()-1)] << rowBack << metadata[(m->currentBinLabels.size()-1)] << "}\n" + spaces + "],\n"; - + if (!picrust) { out << rowFront << m->currentBinLabels[(numBins-1)] << rowBack << metadata[(numBins-1)] << "}\n" + spaces + "],\n"; } + else { out << rowFront << picrustLabels[(numBins-1)] << rowBack << metadata[(numBins-1)] << "}\n" + spaces + "],\n"; } //get column info /*"columns": [ {"id":"Sample1", "metadata":null}, @@ -423,16 +462,16 @@ int MakeBiomCommand::getBiom(vector& lookup){ {"id":"Sample6", "metadata":null} ],*/ - string colBack = "\", \"metadata\":null}"; + string colBack = "\", \"metadata\":"; out << spaces + "\"columns\":[\n"; for (int i = 0; i < lookup.size()-1; i++) { if (m->control_pressed) { out.close(); return 0; } - out << rowFront << lookup[i]->getGroup() << colBack << ",\n"; + out << rowFront << lookup[i]->getGroup() << colBack << sampleMetadata[i] << "},\n"; } - out << rowFront << lookup[(lookup.size()-1)]->getGroup() << colBack << "\n" + spaces + "],\n"; + out << rowFront << lookup[(lookup.size()-1)]->getGroup() << colBack << sampleMetadata[lookup.size()-1] << "}\n" + spaces + "],\n"; out << spaces + "\"matrix_type\": \"" << format << "\",\n" + spaces + "\"matrix_element_type\": \"int\",\n"; - out << spaces + "\"shape\": [" << m->currentBinLabels.size() << "," << lookup.size() << "],\n"; + out << spaces + "\"shape\": [" << numBins << "," << lookup.size() << "],\n"; out << spaces + "\"data\": ["; vector dataRows; @@ -504,7 +543,7 @@ int MakeBiomCommand::getBiom(vector& lookup){ } } //********************************************************************************************************************** -vector MakeBiomCommand::getMetaData(vector& lookup){ +vector MakeBiomCommand::getMetaData(vector& lookup, vector& picrustLabels){ try { vector metadata; @@ -567,7 +606,7 @@ vector MakeBiomCommand::getMetaData(vector& lookup) //traverse the binLabels forming the metadata strings and saving them //make sure to sanity check map::iterator it; - for (int i = 0; i < m->currentBinLabels.size(); i++) { + for (int i = 0; i < lookup[0]->getNumBins(); i++) { if (m->control_pressed) { return metadata; } @@ -575,6 +614,10 @@ vector MakeBiomCommand::getMetaData(vector& lookup) if (it == labelTaxMap.end()) { m->mothurOut("[ERROR]: can't find taxonomy information for " + m->currentBinLabels[i] + ".\n"); m->control_pressed = true; } else { + if (picrust) { + string temp = it->second; m->removeConfidences(temp); + picrustLabels.push_back(temp); + } vector bootstrapValues; string data = "{\"taxonomy\":["; @@ -608,6 +651,81 @@ vector MakeBiomCommand::getMetaData(vector& lookup) } } +//********************************************************************************************************************** +int MakeBiomCommand::getSampleMetaData(vector& lookup){ + try { + sampleMetadata.clear(); + if (metadatafile == "") { for (int i = 0; i < lookup.size(); i++) { sampleMetadata.push_back("null"); } } + else { + ifstream in; + m->openInputFile(metadatafile, in); + + vector groupNames, metadataLabels; + map > lines; + + string headerLine = m->getline(in); m->gobble(in); + vector pieces = m->splitWhiteSpace(headerLine); + + //save names of columns you are reading + for (int i = 1; i < pieces.size(); i++) { + metadataLabels.push_back(pieces[i]); + } + int count = metadataLabels.size(); + + vector groups = m->getGroups(); + + //read rest of file + while (!in.eof()) { + + if (m->control_pressed) { in.close(); return 0; } + + string group = ""; + in >> group; m->gobble(in); + groupNames.push_back(group); + + string line = m->getline(in); m->gobble(in); + vector thisPieces = m->splitWhiteSpaceWithQuotes(line); + + if (thisPieces.size() != count) { m->mothurOut("[ERROR]: expected " + toString(count) + " items of data for sample " + group + " read " + toString(thisPieces.size()) + ", quitting.\n"); } + else { if (m->inUsersGroups(group, groups)) { lines[group] = thisPieces; } } + + m->gobble(in); + } + in.close(); + + map >::iterator it; + for (int i = 0; i < lookup.size(); i++) { + + if (m->control_pressed) { return 0; } + + it = lines.find(lookup[i]->getGroup()); + + if (it == lines.end()) { m->mothurOut("[ERROR]: can't find metadata information for " + lookup[i]->getGroup() + ", quitting.\n"); m->control_pressed = true; } + else { + vector values = it->second; + + string data = "{"; + for (int j = 0; j < metadataLabels.size()-1; j++) { + values[j] = m->removeQuotes(values[j]); + data += "\"" + metadataLabels[j] + "\":\"" + values[j] + "\", "; + } + values[metadataLabels.size()-1] = m->removeQuotes(values[metadataLabels.size()-1]); + data += "\"" + metadataLabels[metadataLabels.size()-1] + "\":\"" + values[metadataLabels.size()-1] + "\"}"; + sampleMetadata.push_back(data); + } + } + } + + return 0; + + } + catch(exception& e) { + m->errorOut(e, "MakeBiomCommand", "getSampleMetaData"); + exit(1); + } + +} + /**************************************************************************************************/ //returns {Bacteria, Bacteroidetes, ..} and scores is filled with {100, 98, ...} or {null, null, null} vector MakeBiomCommand::parseTax(string tax, vector& scores) {