2 * otuhierarchycommand.cpp
5 * Created by westcott on 1/19/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "otuhierarchycommand.h"
12 //**********************************************************************************************************************
13 vector<string> OtuHierarchyCommand::setParameters(){
15 CommandParameter poutput("output", "Multiple", "name-number", "name", "", "", "","",false,false); parameters.push_back(poutput);
16 CommandParameter plist("list", "InputTypes", "", "", "none", "none", "none","otuheirarchy",false,true,true); parameters.push_back(plist);
17 CommandParameter plabel("label", "String", "", "", "", "", "","",false,false); parameters.push_back(plabel);
18 CommandParameter pinputdir("inputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(pinputdir);
19 CommandParameter poutputdir("outputdir", "String", "", "", "", "", "","",false,false); parameters.push_back(poutputdir);
21 vector<string> myArray;
22 for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
26 m->errorOut(e, "OtuHierarchyCommand", "setParameters");
30 //**********************************************************************************************************************
31 string OtuHierarchyCommand::getHelpString(){
33 string helpString = "";
34 helpString += "The otu.hierarchy command is used to see how otus relate at two distances. \n";
35 helpString += "The otu.hierarchy command parameters are list, label and output. list and label parameters are required. \n";
36 helpString += "The output parameter allows you to output the names of the sequence in the OTUs or the OTU numbers. Options are name and number, default is name. \n";
37 helpString += "The otu.hierarchy command should be in the following format: \n";
38 helpString += "otu.hierarchy(list=yourListFile, label=yourLabels).\n";
39 helpString += "Example otu.hierarchy(list=amazon.fn.list, label=0.01-0.03).\n";
40 helpString += "The otu.hierarchy command outputs a .otu.hierarchy file which is described on the wiki.\n";
41 helpString += "Note: No spaces between parameter labels (i.e. list), '=' and parameters (i.e.yourListFile).\n";
45 m->errorOut(e, "OtuHierarchyCommand", "getHelpString");
49 //**********************************************************************************************************************
50 string OtuHierarchyCommand::getOutputPattern(string type) {
54 if (type == "otuheirarchy") { pattern = "[filename],[distance1],[tag],[distance2],otu.hierarchy"; }
55 else { m->mothurOut("[ERROR]: No definition for type " + type + " output pattern.\n"); m->control_pressed = true; }
60 m->errorOut(e, "OtuHierarchyCommand", "getOutputPattern");
64 //**********************************************************************************************************************
65 OtuHierarchyCommand::OtuHierarchyCommand(){
67 abort = true; calledHelp = true;
69 vector<string> tempOutNames;
70 outputTypes["otuheirarchy"] = tempOutNames;
73 m->errorOut(e, "OtuHierarchyCommand", "OtuHierarchyCommand");
77 //**********************************************************************************************************************
78 OtuHierarchyCommand::OtuHierarchyCommand(string option) {
80 abort = false; calledHelp = false;
82 //allow user to run help
83 if(option == "help") { help(); abort = true; calledHelp = true; }
84 else if(option == "citation") { citation(); abort = true; calledHelp = true;}
87 vector<string> myArray = setParameters();
89 OptionParser parser(option);
90 map<string,string> parameters = parser.getParameters();
92 ValidParameters validParameter;
93 map<string,string>::iterator it;
95 //check to make sure all parameters are valid for command
96 for (it = parameters.begin(); it != parameters.end(); it++) {
97 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
100 //initialize outputTypes
101 vector<string> tempOutNames;
102 outputTypes["otuheirarchy"] = tempOutNames;
104 //if the user changes the input directory command factory will send this info to us in the output parameter
105 string inputDir = validParameter.validFile(parameters, "inputdir", false);
106 if (inputDir == "not found"){ inputDir = ""; }
109 it = parameters.find("list");
110 //user has given a template file
111 if(it != parameters.end()){
112 path = m->hasPath(it->second);
113 //if the user has not given a path then, add inputdir. else leave path alone.
114 if (path == "") { parameters["list"] = inputDir + it->second; }
118 listFile = validParameter.validFile(parameters, "list", true);
119 if (listFile == "not found") {
120 listFile = m->getListFile();
121 if (listFile != "") { m->mothurOut("Using " + listFile + " as input file for the list parameter."); m->mothurOutEndLine(); }
123 m->mothurOut("No valid current list file. You must provide a list file."); m->mothurOutEndLine();
126 }else if (listFile == "not open") { abort = true; }
127 else { m->setListFile(listFile); }
129 //if the user changes the output directory command factory will send this info to us in the output parameter
130 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
132 outputDir += m->hasPath(listFile); //if user entered a file with a path then preserve it
135 //check for optional parameter and set defaults
136 // ...at some point should added some additional type checking...
137 label = validParameter.validFile(parameters, "label", false);
138 if (label == "not found") { m->mothurOut("label is a required parameter for the otu.hierarchy command."); m->mothurOutEndLine(); abort = true; }
140 m->splitAtDash(label, labels);
141 if (labels.size() != 2) { m->mothurOut("You must provide 2 labels."); m->mothurOutEndLine(); abort = true; }
144 output = validParameter.validFile(parameters, "output", false); if (output == "not found") { output = "name"; }
146 if ((output != "name") && (output != "number")) { m->mothurOut("output options are name and number. I will use name."); m->mothurOutEndLine(); output = "name"; }
150 catch(exception& e) {
151 m->errorOut(e, "OtuHierarchyCommand", "OtuHierarchyCommand");
155 //**********************************************************************************************************************
157 int OtuHierarchyCommand::execute(){
160 if (abort == true) { if (calledHelp) { return 0; } return 2; }
162 //get listvectors that correspond to labels requested, (or use smart distancing to get closest listvector)
163 vector<ListVector> lists = getListVectors();
165 if (m->control_pressed) { outputTypes.clear(); return 0; }
167 //determine which is little and which is big, putting little first
168 if (lists.size() == 2) {
169 //if big is first swap them
170 if (lists[0].getNumBins() < lists[1].getNumBins()) {
171 reverse(lists.begin(), lists.end());
174 m->mothurOut("error getting listvectors, unable to read 2 different vectors, check your label inputs."); m->mothurOutEndLine(); return 0;
177 //map sequences to bin number in the "little" otu
178 map<string, int> littleBins;
179 vector<string> binLabels0 = lists[0].getLabels();
180 for (int i = 0; i < lists[0].getNumBins(); i++) {
182 if (m->control_pressed) { return 0; }
183 string bin = lists[0].get(i);
184 vector<string> names; m->splitAtComma(bin, names);
185 for (int j = 0; j < names.size(); j++) { littleBins[names[j]] = i; }
189 map<string, string> variables;
190 variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(listFile));
191 variables["[distance1]"] = lists[0].getLabel();
192 variables["[tag]"] = "-";
193 variables["[distance2]"] = lists[1].getLabel();
194 string outputFileName = getOutputFileName("otuheirarchy",variables);
195 m->openOutputFile(outputFileName, out);
197 //go through each bin in "big" otu and output the bins in "little" otu which created it
198 vector<string> binLabels1 = lists[1].getLabels();
199 for (int i = 0; i < lists[1].getNumBins(); i++) {
201 if (m->control_pressed) { outputTypes.clear(); out.close(); m->mothurRemove(outputFileName); return 0; }
203 string binnames = lists[1].get(i);
204 vector<string> names; m->splitAtComma(binnames, names);
207 if (output == "name") { out << binnames << '\t'; }
208 else { out << binLabels1[i] << '\t'; }
210 map<int, int> bins; //bin numbers in little that are in this bin in big
211 map<int, int>::iterator it;
214 for (int j = 0; j < names.size(); j++) { bins[littleBins[names[j]]] = littleBins[names[j]]; }
217 for (it = bins.begin(); it != bins.end(); it++) {
218 if (output == "name") { col2 += lists[0].get(it->first) + "\t"; }
219 else { col2 += binLabels0[it->first] + "\t"; }
228 if (m->control_pressed) { outputTypes.clear(); m->mothurRemove(outputFileName); return 0; }
230 m->mothurOutEndLine();
231 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
232 m->mothurOut(outputFileName); m->mothurOutEndLine(); outputNames.push_back(outputFileName); outputTypes["otuheirarchy"].push_back(outputFileName);
233 m->mothurOutEndLine();
237 catch(exception& e) {
238 m->errorOut(e, "OtuHierarchyCommand", "execute");
243 //**********************************************************************************************************************
244 //returns a vector of listVectors where "little" vector is first
245 vector<ListVector> OtuHierarchyCommand::getListVectors() {
248 int pos; //to use in smart distancing, position of last read in file
250 vector<ListVector> lists;
252 //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
253 set<string> processedLabels;
254 set<string> userLabels = labels;
258 m->openInputFile(listFile, in);
260 //get first list vector in file
261 ListVector* list = NULL;
262 string lastLabel = "";
266 list = new ListVector(in);
268 lastLabel = list->getLabel();
271 while ((list != NULL) && (userLabels.size() != 0)) {
273 if (m->control_pressed) { in.close(); delete list; return lists; }
275 //is this a listvector that we want?
276 if(labels.count(list->getLabel()) == 1){
278 //make copy of listvector
279 ListVector temp(*list);
280 lists.push_back(temp);
282 processedLabels.insert(list->getLabel());
283 userLabels.erase(list->getLabel());
286 //you have a label the user want that is smaller than this label and the last label has not already been processed
287 if ((m->anyLabelsToProcess(list->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
288 string saveLabel = list->getLabel();
289 int savePos = in.tellg();
291 //get smart distance line
295 list = new ListVector(in);
296 }else { list = NULL; }
298 //make copy of listvector
299 ListVector temp(*list);
300 lists.push_back(temp);
302 processedLabels.insert(list->getLabel());
303 userLabels.erase(list->getLabel());
305 //restore real lastlabel to save below
306 list->setLabel(saveLabel);
310 lastLabel = list->getLabel();
317 list = new ListVector(in);
319 }else { list = NULL; }
322 if (m->control_pressed) { in.close(); return lists; }
324 //output error messages about any remaining user labels
325 set<string>::iterator it;
326 bool needToRun = false;
327 for (it = userLabels.begin(); it != userLabels.end(); it++) {
328 m->mothurOut("Your file does not include the label " + *it);
329 if (processedLabels.count(lastLabel) != 1) {
330 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
333 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
337 if (m->control_pressed) { in.close(); return lists; }
339 //run last label if you need to
340 if (needToRun == true) {
341 if (list != NULL) { delete list; }
345 list = new ListVector(in);
347 //make copy of listvector
348 ListVector temp(*list);
349 lists.push_back(temp);
358 catch(exception& e) {
359 m->errorOut(e, "OtuHierarchyCommand", "getListVectors");
364 //**********************************************************************************************************************