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",false,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::getOutputFileNameTag(string type, string inputName=""){
52 string outputFileName = "";
53 map<string, vector<string> >::iterator it;
55 //is this a type this command creates
56 it = outputTypes.find(type);
57 if (it == outputTypes.end()) { m->mothurOut("[ERROR]: this command doesn't create a " + type + " output file.\n"); }
59 if (type == "otuheirarchy") { outputFileName = "otu.hierarchy"; }
60 else { m->mothurOut("[ERROR]: No definition for type " + type + " output file tag.\n"); m->control_pressed = true; }
62 return outputFileName;
65 m->errorOut(e, "OtuHierarchyCommand", "getOutputFileNameTag");
69 //**********************************************************************************************************************
70 OtuHierarchyCommand::OtuHierarchyCommand(){
72 abort = true; calledHelp = true;
74 vector<string> tempOutNames;
75 outputTypes["otuheirarchy"] = tempOutNames;
78 m->errorOut(e, "OtuHierarchyCommand", "OtuHierarchyCommand");
82 //**********************************************************************************************************************
83 OtuHierarchyCommand::OtuHierarchyCommand(string option) {
85 abort = false; calledHelp = false;
87 //allow user to run help
88 if(option == "help") { help(); abort = true; calledHelp = true; }
89 else if(option == "citation") { citation(); abort = true; calledHelp = true;}
92 vector<string> myArray = setParameters();
94 OptionParser parser(option);
95 map<string,string> parameters = parser.getParameters();
97 ValidParameters validParameter;
98 map<string,string>::iterator it;
100 //check to make sure all parameters are valid for command
101 for (it = parameters.begin(); it != parameters.end(); it++) {
102 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
105 //initialize outputTypes
106 vector<string> tempOutNames;
107 outputTypes["otuheirarchy"] = tempOutNames;
109 //if the user changes the input directory command factory will send this info to us in the output parameter
110 string inputDir = validParameter.validFile(parameters, "inputdir", false);
111 if (inputDir == "not found"){ inputDir = ""; }
114 it = parameters.find("list");
115 //user has given a template file
116 if(it != parameters.end()){
117 path = m->hasPath(it->second);
118 //if the user has not given a path then, add inputdir. else leave path alone.
119 if (path == "") { parameters["list"] = inputDir + it->second; }
123 listFile = validParameter.validFile(parameters, "list", true);
124 if (listFile == "not found") {
125 listFile = m->getListFile();
126 if (listFile != "") { m->mothurOut("Using " + listFile + " as input file for the list parameter."); m->mothurOutEndLine(); }
128 m->mothurOut("No valid current list file. You must provide a list file."); m->mothurOutEndLine();
131 }else if (listFile == "not open") { abort = true; }
132 else { m->setListFile(listFile); }
134 //if the user changes the output directory command factory will send this info to us in the output parameter
135 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
137 outputDir += m->hasPath(listFile); //if user entered a file with a path then preserve it
140 //check for optional parameter and set defaults
141 // ...at some point should added some additional type checking...
142 label = validParameter.validFile(parameters, "label", false);
143 if (label == "not found") { m->mothurOut("label is a required parameter for the otu.hierarchy command."); m->mothurOutEndLine(); abort = true; }
145 m->splitAtDash(label, labels);
146 if (labels.size() != 2) { m->mothurOut("You must provide 2 labels."); m->mothurOutEndLine(); abort = true; }
149 output = validParameter.validFile(parameters, "output", false); if (output == "not found") { output = "name"; }
151 if ((output != "name") && (output != "number")) { m->mothurOut("output options are name and number. I will use name."); m->mothurOutEndLine(); output = "name"; }
155 catch(exception& e) {
156 m->errorOut(e, "OtuHierarchyCommand", "OtuHierarchyCommand");
160 //**********************************************************************************************************************
162 int OtuHierarchyCommand::execute(){
165 if (abort == true) { if (calledHelp) { return 0; } return 2; }
167 //get listvectors that correspond to labels requested, (or use smart distancing to get closest listvector)
168 vector<ListVector> lists = getListVectors();
170 if (m->control_pressed) { outputTypes.clear(); return 0; }
172 //determine which is little and which is big, putting little first
173 if (lists.size() == 2) {
174 //if big is first swap them
175 if (lists[0].getNumBins() < lists[1].getNumBins()) {
176 reverse(lists.begin(), lists.end());
179 m->mothurOut("error getting listvectors, unable to read 2 different vectors, check your label inputs."); m->mothurOutEndLine(); return 0;
182 //map sequences to bin number in the "little" otu
183 map<string, int> littleBins;
184 for (int i = 0; i < lists[0].getNumBins(); i++) {
186 if (m->control_pressed) { return 0; }
188 string names = lists[0].get(i);
191 while (names.find_first_of(',') != -1) {
192 string name = names.substr(0,names.find_first_of(','));
193 names = names.substr(names.find_first_of(',')+1, names.length());
194 littleBins[name] = i;
198 littleBins[names] = i;
202 string outputFileName = outputDir + m->getRootName(m->getSimpleName(listFile)) + lists[0].getLabel() + "-" + lists[1].getLabel() + "." + getOutputFileNameTag("otuhierarchy");
203 m->openOutputFile(outputFileName, out);
205 //go through each bin in "big" otu and output the bins in "little" otu which created it
206 for (int i = 0; i < lists[1].getNumBins(); i++) {
208 if (m->control_pressed) { outputTypes.clear(); out.close(); m->mothurRemove(outputFileName); return 0; }
210 string names = lists[1].get(i);
213 if (output == "name") { out << names << '\t'; }
214 else { out << i << '\t'; }
216 map<int, int> bins; //bin numbers in little that are in this bin in big
217 map<int, int>::iterator it;
220 while (names.find_first_of(',') != -1) {
221 string name = names.substr(0,names.find_first_of(','));
222 names = names.substr(names.find_first_of(',')+1, names.length());
223 bins[littleBins[name]] = littleBins[name];
227 bins[littleBins[names]] = littleBins[names];
230 for (it = bins.begin(); it != bins.end(); it++) {
231 if (output == "name") { col2 += lists[0].get(it->first) + "\t"; }
232 else { col2 += toString(it->first) + "\t"; }
241 if (m->control_pressed) { outputTypes.clear(); m->mothurRemove(outputFileName); return 0; }
243 m->mothurOutEndLine();
244 m->mothurOut("Output File Name: "); m->mothurOutEndLine();
245 m->mothurOut(outputFileName); m->mothurOutEndLine(); outputNames.push_back(outputFileName); outputTypes["otuheirarchy"].push_back(outputFileName);
246 m->mothurOutEndLine();
250 catch(exception& e) {
251 m->errorOut(e, "OtuHierarchyCommand", "execute");
256 //**********************************************************************************************************************
257 //returns a vector of listVectors where "little" vector is first
258 vector<ListVector> OtuHierarchyCommand::getListVectors() {
261 int pos; //to use in smart distancing, position of last read in file
263 vector<ListVector> lists;
265 //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
266 set<string> processedLabels;
267 set<string> userLabels = labels;
271 m->openInputFile(listFile, in);
273 //get first list vector in file
274 ListVector* list = NULL;
275 string lastLabel = "";
279 list = new ListVector(in);
281 lastLabel = list->getLabel();
284 while ((list != NULL) && (userLabels.size() != 0)) {
286 if (m->control_pressed) { in.close(); delete list; return lists; }
288 //is this a listvector that we want?
289 if(labels.count(list->getLabel()) == 1){
291 //make copy of listvector
292 ListVector temp(*list);
293 lists.push_back(temp);
295 processedLabels.insert(list->getLabel());
296 userLabels.erase(list->getLabel());
299 //you have a label the user want that is smaller than this label and the last label has not already been processed
300 if ((m->anyLabelsToProcess(list->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
301 string saveLabel = list->getLabel();
302 int savePos = in.tellg();
304 //get smart distance line
308 list = new ListVector(in);
309 }else { list = NULL; }
311 //make copy of listvector
312 ListVector temp(*list);
313 lists.push_back(temp);
315 processedLabels.insert(list->getLabel());
316 userLabels.erase(list->getLabel());
318 //restore real lastlabel to save below
319 list->setLabel(saveLabel);
323 lastLabel = list->getLabel();
330 list = new ListVector(in);
332 }else { list = NULL; }
335 if (m->control_pressed) { in.close(); return lists; }
337 //output error messages about any remaining user labels
338 set<string>::iterator it;
339 bool needToRun = false;
340 for (it = userLabels.begin(); it != userLabels.end(); it++) {
341 m->mothurOut("Your file does not include the label " + *it);
342 if (processedLabels.count(lastLabel) != 1) {
343 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
346 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
350 if (m->control_pressed) { in.close(); return lists; }
352 //run last label if you need to
353 if (needToRun == true) {
354 if (list != NULL) { delete list; }
358 list = new ListVector(in);
360 //make copy of listvector
361 ListVector temp(*list);
362 lists.push_back(temp);
371 catch(exception& e) {
372 m->errorOut(e, "OtuHierarchyCommand", "getListVectors");
377 //**********************************************************************************************************************