2 * getrelabundcommand.cpp
5 * Created by westcott on 6/21/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "getrelabundcommand.h"
12 //**********************************************************************************************************************
13 vector<string> GetRelAbundCommand::setParameters(){
15 CommandParameter pshared("shared", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pshared);
16 CommandParameter pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups);
17 CommandParameter pscale("scale", "Multiple", "totalgroup-totalotu-averagegroup-averageotu", "totalgroup", "", "", "",false,false); parameters.push_back(pscale);
18 CommandParameter plabel("label", "String", "", "", "", "", "",false,false); parameters.push_back(plabel);
19 CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
20 CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
22 vector<string> myArray;
23 for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
27 m->errorOut(e, "GetRelAbundCommand", "setParameters");
31 //**********************************************************************************************************************
32 string GetRelAbundCommand::getHelpString(){
34 string helpString = "";
35 helpString += "The get.relabund command parameters are shared, groups, scale and label. shared is required, unless you have a valid current file.\n";
36 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";
37 helpString += "The label parameter allows you to select what distance levels you would like, and are also separated by dashes.\n";
38 helpString += "The scale parameter allows you to select what scale you would like to use. Choices are totalgroup, totalotu, averagegroup, averageotu, default is totalgroup.\n";
39 helpString += "The get.relabund command should be in the following format: get.relabund(groups=yourGroups, label=yourLabels).\n";
40 helpString += "Example get.relabund(groups=A-B-C, scale=averagegroup).\n";
41 helpString += "The default value for groups is all the groups in your groupfile, and all labels in your inputfile will be used.\n";
42 helpString += "The get.relabund command outputs a .relabund file.\n";
43 helpString += "Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n";
47 m->errorOut(e, "GetRelAbundCommand", "getHelpString");
51 //**********************************************************************************************************************
52 GetRelAbundCommand::GetRelAbundCommand(){
54 abort = true; calledHelp = true;
56 vector<string> tempOutNames;
57 outputTypes["relabund"] = tempOutNames;
60 m->errorOut(e, "GetRelAbundCommand", "GetRelAbundCommand");
64 //**********************************************************************************************************************
65 GetRelAbundCommand::GetRelAbundCommand(string option) {
67 abort = false; calledHelp = false;
70 //allow user to run help
71 if(option == "help") { help(); abort = true; calledHelp = true; }
72 else if(option == "citation") { citation(); abort = true; calledHelp = true;}
75 vector<string> myArray = setParameters();
77 OptionParser parser(option);
78 map<string,string> parameters = parser.getParameters();
79 map<string,string>::iterator it;
81 ValidParameters validParameter;
83 //check to make sure all parameters are valid for command
84 for (it = parameters.begin(); it != parameters.end(); it++) {
85 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
88 //initialize outputTypes
89 vector<string> tempOutNames;
90 outputTypes["relabund"] = tempOutNames;
92 //if the user changes the input directory command factory will send this info to us in the output parameter
93 string inputDir = validParameter.validFile(parameters, "inputdir", false);
94 if (inputDir == "not found"){ inputDir = ""; }
97 it = parameters.find("shared");
98 //user has given a template file
99 if(it != parameters.end()){
100 path = m->hasPath(it->second);
101 //if the user has not given a path then, add inputdir. else leave path alone.
102 if (path == "") { parameters["shared"] = inputDir + it->second; }
107 sharedfile = validParameter.validFile(parameters, "shared", true);
108 if (sharedfile == "not open") { sharedfile = ""; abort = true; }
109 else if (sharedfile == "not found") {
110 //if there is a current shared file, use it
111 sharedfile = m->getSharedFile();
112 if (sharedfile != "") { m->mothurOut("Using " + sharedfile + " as input file for the shared parameter."); m->mothurOutEndLine(); }
113 else { m->mothurOut("You have no current sharedfile and the shared parameter is required."); m->mothurOutEndLine(); abort = true; }
114 }else { m->setSharedFile(sharedfile); }
117 //if the user changes the output directory command factory will send this info to us in the output parameter
118 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = m->hasPath(sharedfile); }
120 //check for optional parameter and set defaults
121 // ...at some point should added some additional type checking...
122 label = validParameter.validFile(parameters, "label", false);
123 if (label == "not found") { label = ""; }
125 if(label != "all") { m->splitAtDash(label, labels); allLines = 0; }
126 else { allLines = 1; }
129 groups = validParameter.validFile(parameters, "groups", false);
130 if (groups == "not found") { groups = ""; pickedGroups = false; }
133 m->splitAtDash(groups, Groups);
134 m->setGroups(Groups);
137 scale = validParameter.validFile(parameters, "scale", false); if (scale == "not found") { scale = "totalgroup"; }
139 if ((scale != "totalgroup") && (scale != "totalotu") && (scale != "averagegroup") && (scale != "averageotu")) {
140 m->mothurOut(scale + " is not a valid scaling option for the get.relabund command. Choices are totalgroup, totalotu, averagegroup, averageotu."); m->mothurOutEndLine(); abort = true;
145 catch(exception& e) {
146 m->errorOut(e, "GetRelAbundCommand", "GetRelAbundCommand");
150 //**********************************************************************************************************************
152 int GetRelAbundCommand::execute(){
155 if (abort == true) { if (calledHelp) { return 0; } return 2; }
157 string outputFileName = outputDir + m->getRootName(m->getSimpleName(sharedfile)) + "relabund";
159 m->openOutputFile(outputFileName, out);
160 out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint);
162 input = new InputData(sharedfile, "sharedfile");
163 lookup = input->getSharedRAbundVectors();
164 string lastLabel = lookup[0]->getLabel();
166 //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
167 set<string> processedLabels;
168 set<string> userLabels = labels;
170 //as long as you are not at the end of the file or done wih the lines you want
171 while((lookup[0] != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
173 if (m->control_pressed) { outputTypes.clear(); for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } m->clearGroups(); delete input; out.close(); m->mothurRemove(outputFileName); return 0; }
175 if(allLines == 1 || labels.count(lookup[0]->getLabel()) == 1){
177 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
178 if (!m->printedHeaders) { lookup[0]->printHeaders(out); }
179 getRelAbundance(lookup, out);
181 processedLabels.insert(lookup[0]->getLabel());
182 userLabels.erase(lookup[0]->getLabel());
185 if ((m->anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
186 string saveLabel = lookup[0]->getLabel();
188 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
189 lookup = input->getSharedRAbundVectors(lastLabel);
190 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
191 if (!m->printedHeaders) { lookup[0]->printHeaders(out); }
192 getRelAbundance(lookup, out);
194 processedLabels.insert(lookup[0]->getLabel());
195 userLabels.erase(lookup[0]->getLabel());
197 //restore real lastlabel to save below
198 lookup[0]->setLabel(saveLabel);
201 lastLabel = lookup[0]->getLabel();
202 //prevent memory leak
203 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; lookup[i] = NULL; }
205 if (m->control_pressed) { outputTypes.clear(); m->clearGroups(); delete input; out.close(); m->mothurRemove(outputFileName); return 0; }
207 //get next line to process
208 lookup = input->getSharedRAbundVectors();
211 if (m->control_pressed) { outputTypes.clear(); m->clearGroups(); delete input; out.close(); m->mothurRemove(outputFileName); return 0; }
213 //output error messages about any remaining user labels
214 set<string>::iterator it;
215 bool needToRun = false;
216 for (it = userLabels.begin(); it != userLabels.end(); it++) {
217 m->mothurOut("Your file does not include the label " + *it);
218 if (processedLabels.count(lastLabel) != 1) {
219 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
222 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
226 //run last label if you need to
227 if (needToRun == true) {
228 for (int i = 0; i < lookup.size(); i++) { if (lookup[i] != NULL) { delete lookup[i]; } }
229 lookup = input->getSharedRAbundVectors(lastLabel);
231 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
232 if (!m->printedHeaders) { lookup[0]->printHeaders(out); }
233 getRelAbundance(lookup, out);
235 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
238 //reset groups parameter
243 if (m->control_pressed) { outputTypes.clear(); m->mothurRemove(outputFileName); return 0;}
245 m->mothurOutEndLine();
246 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
247 m->mothurOut(outputFileName); m->mothurOutEndLine(); outputNames.push_back(outputFileName); outputTypes["relabund"].push_back(outputFileName);
248 m->mothurOutEndLine();
250 //set relabund file as new current relabundfile
252 itTypes = outputTypes.find("relabund");
253 if (itTypes != outputTypes.end()) {
254 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setRelAbundFile(current); }
259 catch(exception& e) {
260 m->errorOut(e, "GetRelAbundCommand", "execute");
264 //**********************************************************************************************************************
266 int GetRelAbundCommand::getRelAbundance(vector<SharedRAbundVector*>& thisLookUp, ofstream& out){
269 for (int i = 0; i < thisLookUp.size(); i++) {
270 out << thisLookUp[i]->getLabel() << '\t' << thisLookUp[i]->getGroup() << '\t' << thisLookUp[i]->getNumBins() << '\t';
272 for (int j = 0; j < thisLookUp[i]->getNumBins(); j++) {
274 if (m->control_pressed) { return 0; }
276 int abund = thisLookUp[i]->getAbundance(j);
278 float relabund = 0.0;
280 if (scale == "totalgroup") {
281 relabund = abund / (float) thisLookUp[i]->getNumSeqs();
282 }else if (scale == "totalotu") {
283 //calc the total in this otu
285 for (int l = 0; l < thisLookUp.size(); l++) { totalOtu += thisLookUp[l]->getAbundance(j); }
286 relabund = abund / (float) totalOtu;
287 }else if (scale == "averagegroup") {
288 relabund = abund / (float) (thisLookUp[i]->getNumSeqs() / (float) thisLookUp[i]->getNumBins());
289 }else if (scale == "averageotu") {
290 //calc the total in this otu
292 for (int l = 0; l < thisLookUp.size(); l++) { totalOtu += thisLookUp[l]->getAbundance(j); }
293 float averageOtu = totalOtu / (float) thisLookUp.size();
295 relabund = abund / (float) averageOtu;
296 }else{ m->mothurOut(scale + " is not a valid scaling option."); m->mothurOutEndLine(); m->control_pressed = true; return 0; }
298 out << relabund << '\t';
305 catch(exception& e) {
306 m->errorOut(e, "GetRelAbundCommand", "getRelAbundance");
310 //**********************************************************************************************************************