#include "makebiomcommand.h"
#include "sharedrabundvector.h"
#include "inputdata.h"
+#include "sharedutilities.h"
//taken from http://biom-format.org/documentation/biom_format.html
/* Minimal Sparse
//**********************************************************************************************************************
vector<string> 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<string> myArray;
for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
string MakeBiomCommand::getHelpString(){
try {
string helpString = "";
- helpString += "The make.biom command parameters are shared, 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. 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 += "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 after the classify.otu command was run on the list and taxonomy file.\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";
helpString += "The make.biom command outputs a .biom file.\n";
exit(1);
}
}
+//**********************************************************************************************************************
+string MakeBiomCommand::getOutputPattern(string type) {
+ try {
+ string pattern = "";
+
+ 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);
+ }
+}
+
//**********************************************************************************************************************
MakeBiomCommand::MakeBiomCommand(){
try {
for (it = parameters.begin(); it != parameters.end(); it++) {
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
-
+
//initialize outputTypes
vector<string> tempOutNames;
- outputTypes["relabund"] = tempOutNames;
+ outputTypes["biom"] = 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 (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["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["contaxonomy"] = inputDir + it->second; }
+ if (path == "") { parameters["metadata"] = inputDir + it->second; }
}
}
//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...
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 = ""; }
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<string> processedLabels;
set<string> userLabels = labels;
if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; }
+ //set sabund file as new current sabundfile
+ string current = "";
+ itTypes = outputTypes.find("biom");
+ if (itTypes != outputTypes.end()) {
+ if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setBiomFile(current); }
+ }
+
+
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
//**********************************************************************************************************************
int MakeBiomCommand::getBiom(vector<SharedRAbundVector*>& lookup){
try {
-
- string outputFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + lookup[0]->getLabel() + ".biom";
+ map<string, string> 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);
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<string> metadata = getMetaData(lookup);
+ int numBins = lookup[0]->getNumBins();
+ vector<string> picrustLabels;
+ vector<string> metadata = getMetaData(lookup, picrustLabels);
if (m->control_pressed) { out.close(); return 0; }
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->currentSharedBinLabels[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->currentSharedBinLabels[(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},
{"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<string> dataRows;
}
}
//**********************************************************************************************************************
-vector<string> MakeBiomCommand::getMetaData(vector<SharedRAbundVector*>& lookup){
+vector<string> MakeBiomCommand::getMetaData(vector<SharedRAbundVector*>& lookup, vector<string>& picrustLabels){
try {
vector<string> metadata;
in.close();
//should the labels be Otu001 or PhyloType001
- string firstBin = m->currentBinLabels[0];
+ string firstBin = m->currentSharedBinLabels[0];
string binTag = "Otu";
if ((firstBin.find("Otu")) == string::npos) { binTag = "PhyloType"; }
if (m->control_pressed) { return metadata; }
//if there is a bin label use it otherwise make one
- string binLabel = binTag;
- string sbinNumber = otuLabels[i];
- if (sbinNumber.length() < snumBins.length()) {
- int diff = snumBins.length() - sbinNumber.length();
- for (int h = 0; h < diff; h++) { binLabel += "0"; }
- }
- binLabel += sbinNumber;
-
- labelTaxMap[binLabel] = taxs[i];
+ if (m->isContainingOnlyDigits(otuLabels[i])) {
+ string binLabel = binTag;
+ string sbinNumber = otuLabels[i];
+ if (sbinNumber.length() < snumBins.length()) {
+ int diff = snumBins.length() - sbinNumber.length();
+ for (int h = 0; h < diff; h++) { binLabel += "0"; }
+ }
+ binLabel += sbinNumber;
+ binLabel = m->getSimpleLabel(binLabel);
+ labelTaxMap[binLabel] = taxs[i];
+ }else { labelTaxMap[m->getSimpleLabel(otuLabels[i])] = taxs[i]; }
}
//traverse the binLabels forming the metadata strings and saving them
//make sure to sanity check
map<string, string>::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; }
- it = labelTaxMap.find(m->currentBinLabels[i]);
+ it = labelTaxMap.find(m->getSimpleLabel(m->currentSharedBinLabels[i]));
- if (it == labelTaxMap.end()) { m->mothurOut("[ERROR]: can't find taxonomy information for " + m->currentBinLabels[i] + ".\n"); m->control_pressed = true; }
+ if (it == labelTaxMap.end()) { m->mothurOut("[ERROR]: can't find taxonomy information for " + m->currentSharedBinLabels[i] + ".\n"); m->control_pressed = true; }
else {
+ if (picrust) {
+ string temp = it->second; m->removeConfidences(temp);
+ picrustLabels.push_back(temp);
+ }
vector<string> bootstrapValues;
string data = "{\"taxonomy\":[";
}
}
+//**********************************************************************************************************************
+int MakeBiomCommand::getSampleMetaData(vector<SharedRAbundVector*>& 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<string> groupNames, metadataLabels;
+ map<string, vector<string> > lines;
+
+ string headerLine = m->getline(in); m->gobble(in);
+ vector<string> 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<string> 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<string> 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<string, vector<string> >::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<string> 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<string> MakeBiomCommand::parseTax(string tax, vector<string>& scores) {
scores.push_back(confidenceScore);
}else{ scores.push_back("null"); }
}
- }
+ }else{ scores.push_back("null"); }
//strip "" if they are there
pos = taxon.find("\"");