try {
ValidCalculators validCalculator;
string helpString = "";
- helpString += "The rarefaction.single command parameters are list, sabund, rabund, shared, label, iters, freq, calc, processors and abund. list, sabund, rabund or shared is required unless you have a valid current file. \n";
+ helpString += "The rarefaction.single command parameters are list, sabund, rabund, shared, label, iters, freq, calc, processors, groupmode and abund. list, sabund, rabund or shared is required unless you have a valid current file. \n";
helpString += "The freq parameter is used indicate when to output your data, by default it is set to 100. But you can set it to a percentage of the number of sequence. For example freq=0.10, means 10%. \n";
helpString += "The processors parameter allows you to specify the number of processors to use. The default is 1.\n";
helpString += "The rarefaction.single command should be in the following format: \n";
exit(1);
}
}
+//**********************************************************************************************************************
+string RareFactCommand::getOutputFileNameTag(string type, string inputName=""){
+ try {
+ string outputFileName = "";
+ map<string, vector<string> >::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 == "rarefaction") { outputFileName = "rarefaction"; }
+ else if (type == "r_chao") { outputFileName = "r_chao"; }
+ else if (type == "r_ace") { outputFileName = "r_ace"; }
+ else if (type == "r_jack") { outputFileName = "r_jack"; }
+ else if (type == "r_shannon") { outputFileName = "r_shannon"; }
+ else if (type == "r_shannoneven") { outputFileName = "r_shannoneven"; }
+ else if (type == "r_smithwilson") { outputFileName = "r_smithwilson"; }
+ else if (type == "r_npshannon") { outputFileName = "r_npshannon"; }
+ else if (type == "r_simpson") { outputFileName = "r_simpson"; }
+ else if (type == "r_simpsoneven") { outputFileName = "r_simpsoneven"; }
+ else if (type == "r_invsimpson") { outputFileName = "r_invsimpson"; }
+ else if (type == "r_bootstrap") { outputFileName = "r_bootstrap"; }
+ else if (type == "r_coverage") { outputFileName = "r_coverage"; }
+ else if (type == "r_nseqs") { outputFileName = "r_nseqs"; }
+ 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, "RareFactCommand", "getOutputFileNameTag");
+ exit(1);
+ }
+}
//**********************************************************************************************************************
RareFactCommand::RareFactCommand(){
if (abort == true) { if (calledHelp) { return 0; } return 2; }
+ map<string, set<int> > labelToEnds;
if ((format != "sharedfile")) { inputFileNames.push_back(inputfile); }
- else { inputFileNames = parseSharedFile(sharedfile); format = "rabund"; }
-
- if (m->control_pressed) { return 0; }
+ else { inputFileNames = parseSharedFile(sharedfile, labelToEnds); format = "rabund"; }
+
+ if (m->control_pressed) { return 0; }
map<int, string> file2Group; //index in outputNames[i] -> group
for (int p = 0; p < inputFileNames.size(); p++) {
for (i=0; i<Estimators.size(); i++) {
if (validCalculator.isValidCalculator("rarefaction", Estimators[i]) == true) {
if (Estimators[i] == "sobs") {
- rDisplays.push_back(new RareDisplay(new Sobs(), new ThreeColumnFile(fileNameRoot+"rarefaction")));
- outputNames.push_back(fileNameRoot+"rarefaction"); outputTypes["rarefaction"].push_back(fileNameRoot+"rarefaction");
+ rDisplays.push_back(new RareDisplay(new Sobs(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("rarefaction"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("rarefaction")); outputTypes["rarefaction"].push_back(fileNameRoot+getOutputFileNameTag("rarefaction"));
}else if (Estimators[i] == "chao") {
- rDisplays.push_back(new RareDisplay(new Chao1(), new ThreeColumnFile(fileNameRoot+"r_chao")));
- outputNames.push_back(fileNameRoot+"r_chao"); outputTypes["r_chao"].push_back(fileNameRoot+"r_chao");
+ rDisplays.push_back(new RareDisplay(new Chao1(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_chao"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_chao")); outputTypes["r_chao"].push_back(fileNameRoot+getOutputFileNameTag("r_chao"));
}else if (Estimators[i] == "ace") {
if(abund < 5)
abund = 10;
- rDisplays.push_back(new RareDisplay(new Ace(abund), new ThreeColumnFile(fileNameRoot+"r_ace")));
- outputNames.push_back(fileNameRoot+"r_ace"); outputTypes["r_ace"].push_back(fileNameRoot+"r_ace");
+ rDisplays.push_back(new RareDisplay(new Ace(abund), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_ace"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_ace")); outputTypes["r_ace"].push_back(fileNameRoot+getOutputFileNameTag("r_ace"));
}else if (Estimators[i] == "jack") {
- rDisplays.push_back(new RareDisplay(new Jackknife(), new ThreeColumnFile(fileNameRoot+"r_jack")));
- outputNames.push_back(fileNameRoot+"r_jack"); outputTypes["r_jack"].push_back(fileNameRoot+"r_jack");
+ rDisplays.push_back(new RareDisplay(new Jackknife(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_jack"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_jack")); outputTypes["r_jack"].push_back(fileNameRoot+getOutputFileNameTag("r_jack"));
}else if (Estimators[i] == "shannon") {
- rDisplays.push_back(new RareDisplay(new Shannon(), new ThreeColumnFile(fileNameRoot+"r_shannon")));
- outputNames.push_back(fileNameRoot+"r_shannon"); outputTypes["r_shannon"].push_back(fileNameRoot+"r_shannon");
+ rDisplays.push_back(new RareDisplay(new Shannon(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_shannon"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_shannon")); outputTypes["r_shannon"].push_back(fileNameRoot+getOutputFileNameTag("r_shannon"));
}else if (Estimators[i] == "shannoneven") {
- rDisplays.push_back(new RareDisplay(new ShannonEven(), new ThreeColumnFile(fileNameRoot+"r_shannoneven")));
- outputNames.push_back(fileNameRoot+"r_shannoneven"); outputTypes["r_shannoneven"].push_back(fileNameRoot+"r_shannoneven");
+ rDisplays.push_back(new RareDisplay(new ShannonEven(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_shannoneven"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_shannoneven")); outputTypes["r_shannoneven"].push_back(fileNameRoot+getOutputFileNameTag("r_shannoneven"));
}else if (Estimators[i] == "heip") {
- rDisplays.push_back(new RareDisplay(new Heip(), new ThreeColumnFile(fileNameRoot+"r_heip")));
- outputNames.push_back(fileNameRoot+"r_heip"); outputTypes["r_heip"].push_back(fileNameRoot+"r_heip");
+ rDisplays.push_back(new RareDisplay(new Heip(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_heip"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_heip")); outputTypes["r_heip"].push_back(fileNameRoot+getOutputFileNameTag("r_heip"));
}else if (Estimators[i] == "smithwilson") {
- rDisplays.push_back(new RareDisplay(new SmithWilson(), new ThreeColumnFile(fileNameRoot+"r_smithwilson")));
- outputNames.push_back(fileNameRoot+"r_smithwilson"); outputTypes["r_smithwilson"].push_back(fileNameRoot+"r_smithwilson");
+ rDisplays.push_back(new RareDisplay(new SmithWilson(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_smithwilson"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_smithwilson")); outputTypes["r_smithwilson"].push_back(fileNameRoot+getOutputFileNameTag("r_smithwilson"));
}else if (Estimators[i] == "npshannon") {
- rDisplays.push_back(new RareDisplay(new NPShannon(), new ThreeColumnFile(fileNameRoot+"r_npshannon")));
- outputNames.push_back(fileNameRoot+"r_npshannon"); outputTypes["r_npshannon"].push_back(fileNameRoot+"r_npshannon");
+ rDisplays.push_back(new RareDisplay(new NPShannon(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_npshannon"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_npshannon")); outputTypes["r_npshannon"].push_back(fileNameRoot+getOutputFileNameTag("r_npshannon"));
}else if (Estimators[i] == "simpson") {
- rDisplays.push_back(new RareDisplay(new Simpson(), new ThreeColumnFile(fileNameRoot+"r_simpson")));
- outputNames.push_back(fileNameRoot+"r_simpson"); outputTypes["r_simpson"].push_back(fileNameRoot+"r_simpson");
+ rDisplays.push_back(new RareDisplay(new Simpson(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_simpson"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_simpson")); outputTypes["r_simpson"].push_back(fileNameRoot+getOutputFileNameTag("r_simpson"));
}else if (Estimators[i] == "simpsoneven") {
- rDisplays.push_back(new RareDisplay(new SimpsonEven(), new ThreeColumnFile(fileNameRoot+"r_simpsoneven")));
- outputNames.push_back(fileNameRoot+"r_simpsoneven"); outputTypes["r_simpsoneven"].push_back(fileNameRoot+"r_simpsoneven");
+ rDisplays.push_back(new RareDisplay(new SimpsonEven(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_simpsoneven"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_simpsoneven")); outputTypes["r_simpsoneven"].push_back(fileNameRoot+getOutputFileNameTag("r_simpsoneven"));
}else if (Estimators[i] == "invsimpson") {
- rDisplays.push_back(new RareDisplay(new InvSimpson(), new ThreeColumnFile(fileNameRoot+"r_invsimpson")));
- outputNames.push_back(fileNameRoot+"r_invsimpson"); outputTypes["r_invsimpson"].push_back(fileNameRoot+"r_invsimpson");
+ rDisplays.push_back(new RareDisplay(new InvSimpson(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_invsimpson"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_invsimpson")); outputTypes["r_invsimpson"].push_back(fileNameRoot+getOutputFileNameTag("r_invsimpson"));
}else if (Estimators[i] == "bootstrap") {
- rDisplays.push_back(new RareDisplay(new Bootstrap(), new ThreeColumnFile(fileNameRoot+"r_bootstrap")));
- outputNames.push_back(fileNameRoot+"r_bootstrap"); outputTypes["r_bootstrap"].push_back(fileNameRoot+"r_bootstrap");
+ rDisplays.push_back(new RareDisplay(new Bootstrap(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_bootstrap"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_bootstrap")); outputTypes["r_bootstrap"].push_back(fileNameRoot+getOutputFileNameTag("r_bootstrap"));
}else if (Estimators[i] == "coverage") {
- rDisplays.push_back(new RareDisplay(new Coverage(), new ThreeColumnFile(fileNameRoot+"r_coverage")));
- outputNames.push_back(fileNameRoot+"r_coverage"); outputTypes["r_coverage"].push_back(fileNameRoot+"r_coverage");
+ rDisplays.push_back(new RareDisplay(new Coverage(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_coverage"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_coverage")); outputTypes["r_coverage"].push_back(fileNameRoot+getOutputFileNameTag("r_coverage"));
}else if (Estimators[i] == "nseqs") {
- rDisplays.push_back(new RareDisplay(new NSeqs(), new ThreeColumnFile(fileNameRoot+"r_nseqs")));
- outputNames.push_back(fileNameRoot+"r_nseqs"); outputTypes["r_nseqs"].push_back(fileNameRoot+"r_nseqs");
+ rDisplays.push_back(new RareDisplay(new NSeqs(), new ThreeColumnFile(fileNameRoot+getOutputFileNameTag("r_nseqs"))));
+ outputNames.push_back(fileNameRoot+getOutputFileNameTag("r_nseqs")); outputTypes["r_nseqs"].push_back(fileNameRoot+getOutputFileNameTag("r_nseqs"));
}
if (inputFileNames.size() > 1) { file2Group[outputNames.size()-1] = groups[p]; }
}
if(allLines == 1 || labels.count(order->getLabel()) == 1){
m->mothurOut(order->getLabel()); m->mothurOutEndLine();
- rCurve = new Rarefact(order, rDisplays, processors);
+ map<string, set<int> >::iterator itEndings = labelToEnds.find(order->getLabel());
+ set<int> ends;
+ if (itEndings != labelToEnds.end()) { ends = itEndings->second; }
+ rCurve = new Rarefact(order, rDisplays, processors, ends);
rCurve->getCurve(freq, nIters);
delete rCurve;
order = (input->getOrderVector(lastLabel));
m->mothurOut(order->getLabel()); m->mothurOutEndLine();
- rCurve = new Rarefact(order, rDisplays, processors);
+ map<string, set<int> >::iterator itEndings = labelToEnds.find(order->getLabel());
+ set<int> ends;
+ if (itEndings != labelToEnds.end()) { ends = itEndings->second; }
+ rCurve = new Rarefact(order, rDisplays, processors, ends);
+
rCurve->getCurve(freq, nIters);
delete rCurve;
order = (input->getOrderVector(lastLabel));
m->mothurOut(order->getLabel()); m->mothurOutEndLine();
- rCurve = new Rarefact(order, rDisplays, processors);
+ map<string, set<int> >::iterator itEndings = labelToEnds.find(order->getLabel());
+ set<int> ends;
+ if (itEndings != labelToEnds.end()) { ends = itEndings->second; }
+ rCurve = new Rarefact(order, rDisplays, processors, ends);
+
rCurve->getCurve(freq, nIters);
delete rCurve;
//find different types of files
map<string, map<string, string> > typesFiles;
+ map<string, vector< vector<string> > > fileLabels; //combofile name to labels. each label is a vector because it may be unique lci hci.
+ vector<string> groupNames;
for (int i = 0; i < outputNames.size(); i++) {
+
string extension = m->getExtension(outputNames[i]);
-
+ string combineFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + "groups" + extension;
+ m->mothurRemove(combineFileName); //remove old file
+
ifstream in;
m->openInputFile(outputNames[i], in);
string labels = m->getline(in);
- string newLine = labels.substr(0, labels.find_first_of('\t'));
-
- newLine += "\tGroup" + labels.substr(labels.find_first_of('\t'));
+ istringstream iss (labels,istringstream::in);
+ string newLabel = ""; vector<string> theseLabels;
+ while(!iss.eof()) { iss >> newLabel; m->gobble(iss); theseLabels.push_back(newLabel); }
+ vector< vector<string> > allLabels;
+ vector<string> thisSet; thisSet.push_back(theseLabels[0]); allLabels.push_back(thisSet); thisSet.clear(); //makes "numSampled" its own grouping
+ for (int j = 1; j < theseLabels.size()-1; j++) {
+ if (theseLabels[j+1] == "lci") {
+ thisSet.push_back(theseLabels[j]);
+ thisSet.push_back(theseLabels[j+1]);
+ thisSet.push_back(theseLabels[j+2]);
+ j++; j++;
+ }else{ //no lci or hci for this calc.
+ thisSet.push_back(theseLabels[j]);
+ }
+ allLabels.push_back(thisSet);
+ thisSet.clear();
+ }
+ fileLabels[combineFileName] = allLabels;
+
map<string, map<string, string> >::iterator itfind = typesFiles.find(extension);
if (itfind != typesFiles.end()) {
(itfind->second)[outputNames[i]] = file2Group[i];
temp[outputNames[i]] = file2Group[i];
typesFiles[extension] = temp;
}
-
- string combineFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + "groups" + extension;
-
- //print headers
- ofstream out;
- m->openOutputFile(combineFileName, out);
- out << newLine << endl;
- out.close();
-
+ if (!(m->inUsersGroups(file2Group[i], groupNames))) { groupNames.push_back(file2Group[i]); }
}
//for each type create a combo file
- map<int, int> lineToNumber;
+
for (map<string, map<string, string> >::iterator it = typesFiles.begin(); it != typesFiles.end(); it++) {
ofstream out;
string combineFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + "groups" + it->first;
m->openOutputFileAppend(combineFileName, out);
newFileNames.push_back(combineFileName);
- map<string, string> thisTypesFiles = it->second;
-
+ map<string, string> thisTypesFiles = it->second; //it->second maps filename to group
+ set<int> numSampledSet;
+
//open each type summary file
- map<string, vector<string> > files; //maps file name to lines in file
+ map<string, map<int, vector< vector<string> > > > files; //maps file name to lines in file
int maxLines = 0;
- int numColumns = 0;
for (map<string, string>::iterator itFileNameGroup = thisTypesFiles.begin(); itFileNameGroup != thisTypesFiles.end(); itFileNameGroup++) {
string thisfilename = itFileNameGroup->first;
string group = itFileNameGroup->second;
-
+
ifstream temp;
m->openInputFile(thisfilename, temp);
//read through first line - labels
m->getline(temp); m->gobble(temp);
- vector<string> thisFilesLines;
-
- thisFilesLines.push_back(group);
- int count = 1;
+ map<int, vector< vector<string> > > thisFilesLines;
while (!temp.eof()){
-
- string thisLine = m->getline(temp);
-
- string numSampled = thisLine.substr(0, thisLine.find_first_of('\t'));
- int num = 0;
- convert(numSampled, num);
- numColumns = m->getNumChar(thisLine, '\t');
- lineToNumber[count] = num;
- count++;
-
- thisFilesLines.push_back(thisLine);
- m->gobble(temp);
+ int numSampled = 0;
+ temp >> numSampled; m->gobble(temp);
+
+ vector< vector<string> > theseReads;
+ vector<string> thisSet; thisSet.push_back(toString(numSampled)); theseReads.push_back(thisSet); thisSet.clear();
+ for (int k = 1; k < fileLabels[combineFileName].size(); k++) { //output thing like 0.03-A lci-A hci-A
+ vector<string> reads;
+ string next = "";
+ for (int l = 0; l < fileLabels[combineFileName][k].size(); l++) { //output modified labels
+ temp >> next; m->gobble(temp);
+ reads.push_back(next);
+ }
+ theseReads.push_back(reads);
+ }
+ thisFilesLines[numSampled] = theseReads;
+ m->gobble(temp);
+
+ numSampledSet.insert(numSampled);
}
- files[thisfilename] = thisFilesLines;
+ files[group] = thisFilesLines;
//save longest file for below
if (maxLines < thisFilesLines.size()) { maxLines = thisFilesLines.size(); }
m->mothurRemove(thisfilename);
}
-
+ //output new labels line
+ out << fileLabels[combineFileName][0][0] << '\t';
+ for (int k = 1; k < fileLabels[combineFileName].size(); k++) { //output thing like 0.03-A lci-A hci-A
+ for (int n = 0; n < groupNames.size(); n++) { // for each group
+ for (int l = 0; l < fileLabels[combineFileName][k].size(); l++) { //output modified labels
+ out << fileLabels[combineFileName][k][l] << '-' << groupNames[n] << '\t';
+ }
+ }
+ }
+ out << endl;
+
//for each label
- for (int k = 1; k < maxLines; k++) {
+ for (set<int>::iterator itNumSampled = numSampledSet.begin(); itNumSampled != numSampledSet.end(); itNumSampled++) {
- //grab data for each group
- for (map<string, string>::iterator itFileNameGroup = thisTypesFiles.begin(); itFileNameGroup != thisTypesFiles.end(); itFileNameGroup++) {
-
- string thisfilename = itFileNameGroup->first;
- map<int, int>::iterator itLine = lineToNumber.find(k);
- if (itLine != lineToNumber.end()) {
- string output = toString(itLine->second);
- if (k < files[thisfilename].size()) {
- string line = files[thisfilename][k];
- output = line.substr(0, line.find_first_of('\t'));
- output += '\t' + files[thisfilename][0] + '\t' + line.substr(line.find_first_of('\t'));
- }else{
- output += '\t' + files[thisfilename][0] + '\t';
- for (int h = 0; h < numColumns; h++) {
- output += "NA\t";
- }
- }
- out << output << endl;
- }else { m->mothurOut("[ERROR]: parsing results, cant find " + toString(k)); m->mothurOutEndLine(); }
- }
+ out << (*itNumSampled) << '\t';
+
+ if (m->control_pressed) { break; }
+
+ for (int k = 1; k < fileLabels[combineFileName].size(); k++) { //each chunk
+ //grab data for each group
+ for (map<string, map<int, vector< vector<string> > > >::iterator itFileNameGroup = files.begin(); itFileNameGroup != files.end(); itFileNameGroup++) {
+
+ string group = itFileNameGroup->first;
+
+ map<int, vector< vector<string> > >::iterator itLine = files[group].find(*itNumSampled);
+ if (itLine != files[group].end()) {
+ for (int l = 0; l < (itLine->second)[k].size(); l++) {
+ out << (itLine->second)[k][l] << '\t';
+
+ }
+ }else {
+ for (int l = 0; l < fileLabels[combineFileName][k].size(); l++) {
+ out << "NA" << '\t';
+ }
+ }
+ }
+ }
+ out << endl;
}
-
out.close();
-
}
//return combine file name
}
}
//**********************************************************************************************************************
-vector<string> RareFactCommand::parseSharedFile(string filename) {
+vector<string> RareFactCommand::parseSharedFile(string filename, map<string, set<int> >& label2Ends) {
try {
vector<string> filenames;
m->openOutputFileAppend(sharedFileRoot + lookup[i]->getGroup() + ".rabund", *(filehandles[lookup[i]->getGroup()]));
rav.print(*(filehandles[lookup[i]->getGroup()]));
(*(filehandles[lookup[i]->getGroup()])).close();
+ label2Ends[lookup[i]->getLabel()].insert(rav.getNumSeqs());
}
for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
projects / mothur.git / blobdiff