#include "readcolumn.h"
#include "formatphylip.h"
#include "formatcolumn.h"
-
+#include "sharedutilities.h"
//********************************************************************************************************************
help(); abort = true;
} else {
//valid paramters for this command
- string Array[] = {"fasta","list","label","name", "group", "sorted", "phylip","column","large","cutoff","precision","outputdir","inputdir"};
+ string Array[] = {"fasta","list","label","name", "group", "sorted", "phylip","column","large","cutoff","precision","groups","outputdir","inputdir"};
vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
OptionParser parser(option);
sorted = "";
}
+ groups = validParameter.validFile(parameters, "groups", false);
+ if (groups == "not found") { groups = ""; }
+ else {
+ if (groupfile == "") {
+ m->mothurOut("You must provide a groupfile to use groups."); m->mothurOutEndLine();
+ abort = true;
+ }else {
+ splitAtDash(groups, Groups);
+ }
+ }
+ globaldata->Groups = Groups;
+
string temp = validParameter.validFile(parameters, "large", false); if (temp == "not found") { temp = "F"; }
large = isTrue(temp);
void GetOTURepCommand::help(){
try {
- m->mothurOut("The get.oturep command parameters are phylip, column, list, fasta, name, group, large, cutoff, precision, sorted and label. The fasta and list parameters are required, as well as phylip or column and name.\n");
+ m->mothurOut("The get.oturep command parameters are phylip, column, list, fasta, name, group, large, cutoff, precision, groups, sorted and label. The fasta and list parameters are required, as well as phylip or column and name.\n");
m->mothurOut("The label parameter allows you to select what distance levels you would like a output files created for, and is separated by dashes.\n");
m->mothurOut("The phylip or column parameter is required, but only one may be used. If you use a column file the name filename is required. \n");
m->mothurOut("If you do not provide a cutoff value 10.00 is assumed. If you do not provide a precision value then 100 is assumed.\n");
m->mothurOut("The default value for label is all labels in your inputfile.\n");
m->mothurOut("The sorted parameter allows you to indicate you want the output sorted. You can sort by sequence name, bin number, bin size or group. The default is no sorting, but your options are name, number, size, or group.\n");
m->mothurOut("The large parameter allows you to indicate that your distance matrix is too large to fit in RAM. The default value is false.\n");
+ m->mothurOut("The groups parameter allows you to indicate that you want representative sequences for each group specified for each OTU, group name should be separated by dashes. ex. groups=A-B-C.\n");
m->mothurOut("The get.oturep command outputs a .fastarep and .rep.names file for each distance you specify, selecting one OTU representative for each bin.\n");
m->mothurOut("If you provide a groupfile, then it also appends the names of the groups present in that bin.\n");
m->mothurOut("Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFastaFile).\n\n");
if (large) { inRow.close(); remove(distFile.c_str()); }
return 0;
}
-
+
+ if (groupfile != "") {
+ //read in group map info.
+ groupMap = new GroupMap(groupfile);
+ int error = groupMap->readMap();
+ if (error == 1) { delete groupMap; m->mothurOut("Error reading your groupfile. Proceeding without groupfile."); m->mothurOutEndLine(); groupfile = ""; }
+
+ if (Groups.size() != 0) {
+ SharedUtil* util = new SharedUtil();
+ util->setGroups(Groups, groupMap->namesOfGroups, "getoturep");
+ delete util;
+ }
+ }
+
//set format to list so input can get listvector
globaldata->setFormat("list");
delete input; globaldata->ginput = NULL;
delete read;
- if (groupfile != "") {
- //read in group map info.
- groupMap = new GroupMap(groupfile);
- int error = groupMap->readMap();
- if (error == 1) { delete groupMap; m->mothurOut("Error reading your groupfile. Proceeding without groupfile."); m->mothurOutEndLine(); groupfile = ""; }
- }
-
//read fastafile
fasta = new FastaMap();
fasta->readFastaFile(fastafile);
}
}
//**********************************************************************************************************************
-string GetOTURepCommand::findRep(int bin, ListVector* thisList) {
+string GetOTURepCommand::findRep(vector<string> names) {
try{
- vector<string> names;
- map<string, string> groups;
- map<string, string>::iterator groupIt;
-
- //parse names into vector
- string binnames = thisList->get(bin);
- splitAtComma(binnames, names);
-
// if only 1 sequence in bin or processing the "unique" label, then
// the first sequence of the OTU is the representative one
if ((names.size() == 1) || (list->getLabel() == "unique")) {
}
}
}
-
+
return(names[minIndex]);
}
}
//**********************************************************************************************************************
int GetOTURepCommand::process(ListVector* processList) {
try{
- string nameRep, name, sequence;
+ string name, sequence;
+ string nameRep;
//create output file
if (outputDir == "") { outputDir += hasPath(listfile); }
ofstream newNamesOutput;
- string outputNamesFile = outputDir + getRootName(getSimpleName(listfile)) + processList->getLabel() + ".rep.names";
- openOutputFile(outputNamesFile, newNamesOutput);
- outputNames.push_back(outputNamesFile);
- outputNameFiles[outputNamesFile] = processList->getLabel();
+ string outputNamesFile;
+ map<string, ofstream*> filehandles;
+
+ if (Groups.size() == 0) { //you don't want to use groups
+ outputNamesFile = outputDir + getRootName(getSimpleName(listfile)) + processList->getLabel() + ".rep.names";
+ openOutputFile(outputNamesFile, newNamesOutput);
+ outputNames.push_back(outputNamesFile);
+ outputNameFiles[outputNamesFile] = processList->getLabel();
+ }else{ //you want to use groups
+ ofstream* temp;
+ for (int i=0; i<Groups.size(); i++) {
+ temp = new ofstream;
+ filehandles[Groups[i]] = temp;
+ outputNamesFile = outputDir + getRootName(getSimpleName(listfile)) + processList->getLabel() + "." + Groups[i] + ".rep.names";
+
+ openOutputFile(outputNamesFile, *(temp));
+ outputNames.push_back(outputNamesFile);
+ outputNameFiles[outputNamesFile] = processList->getLabel() + "." + Groups[i];
+ }
+ }
//for each bin in the list vector
for (int i = 0; i < processList->size(); i++) {
- nameRep = findRep(i, processList);
-
- if (m->control_pressed) { out.close(); newNamesOutput.close(); return 0; }
+ if (m->control_pressed) {
+ out.close();
+ if (Groups.size() == 0) { //you don't want to use groups
+ newNamesOutput.close();
+ }else{
+ for (int j=0; j<Groups.size(); j++) {
+ (*(filehandles[Groups[j]])).close();
+ delete filehandles[Groups[j]];
+ }
+ }
+ return 0;
+ }
- //output to new names file
- newNamesOutput << nameRep << '\t' << processList->get(i) << endl;
+ string temp = processList->get(i);
+ vector<string> namesInBin;
+ splitAtComma(temp, namesInBin);
+
+ if (Groups.size() == 0) {
+ nameRep = findRep(namesInBin);
+ newNamesOutput << i << '\t' << nameRep << '\t' << processList->get(i) << endl;
+ }else{
+ map<string, vector<string> > NamesInGroup;
+ for (int j=0; j<Groups.size(); j++) { //initialize groups
+ NamesInGroup[Groups[j]].resize(0);
+ }
+
+ for (int j=0; j<namesInBin.size(); j++) {
+ string thisgroup = groupMap->getGroup(namesInBin[j]);
+
+ if (thisgroup == "not found") { m->mothurOut(namesInBin[j] + " is not in your groupfile, please correct."); m->mothurOutEndLine(); m->control_pressed = true; }
+
+ if (inUsersGroups(thisgroup, Groups)) { //add this name to correct group
+ NamesInGroup[thisgroup].push_back(namesInBin[j]);
+ }
+ }
+
+ //get rep for each group in otu
+ for (int j=0; j<Groups.size(); j++) {
+ if (NamesInGroup[Groups[j]].size() != 0) { //are there members from this group in this otu?
+ //get rep for each group
+ nameRep = findRep(NamesInGroup[Groups[j]]);
+
+ //output group rep and other members of this group
+ (*(filehandles[Groups[j]])) << i << '\t' << nameRep << '\t';
+
+ for (int k=0; k<NamesInGroup[Groups[j]].size()-1; k++) {//output list of names in this otu from this group
+ (*(filehandles[Groups[j]])) << NamesInGroup[Groups[j]][k] << ",";
+ }
+ //output last name
+ (*(filehandles[Groups[j]])) << NamesInGroup[Groups[j]][NamesInGroup[Groups[j]].size()-1] << endl;
+ }
+ }
+ }
}
-
- newNamesOutput.close();
+
+ if (Groups.size() == 0) { //you don't want to use groups
+ newNamesOutput.close();
+ }else{
+ for (int i=0; i<Groups.size(); i++) {
+ (*(filehandles[Groups[i]])).close();
+ delete filehandles[Groups[i]];
+ }
+ }
+
return 0;
}
vector<repStruct> reps;
outputNames.push_back(outputFileName);
+ ofstream out2;
+ string tempNameFile = filename + ".temp";
+ openOutputFile(tempNameFile, out2);
+
ifstream in;
openInputFile(filename, in);
int i = 0;
while (!in.eof()) {
string rep, binnames;
- in >> rep >> binnames; gobble(in);
+ in >> i >> rep >> binnames; gobble(in);
+ out2 << rep << '\t' << binnames << endl;
vector<string> names;
splitAtComma(binnames, names);
}else {
m->mothurOut(rep + " is missing from your fasta or name file, ignoring. Please correct."); m->mothurOutEndLine();
}
- i++;
}
}
out.close();
+ out2.close();
+
+ rename(tempNameFile.c_str(), filename.c_str());
return 0;
projects / mothur.git / blobdiff