5 * Created by Sarah Westcott on 1/2/09.
6 * Copyright 2009 Schloss Lab UMASS Amherst. All rights reserved.
10 #include "collectcommand.h"
15 #include "bootstrap.h"
17 #include "npshannon.h"
19 #include "jackknife.h"
23 #include "bergerparker.h"
25 #include "goodscoverage.h"
33 //**********************************************************************************************************************
34 CollectCommand::CollectCommand(string option){
36 globaldata = GlobalData::getInstance();
42 //allow user to run help
43 if(option == "help") { validCalculator = new ValidCalculators(); help(); delete validCalculator; abort = true; }
46 //valid paramters for this command
47 string Array[] = {"freq","label","calc","abund","size","outputdir","inputdir"};
48 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
50 OptionParser parser(option);
51 map<string,string> parameters = parser.getParameters();
53 ValidParameters validParameter;
55 //check to make sure all parameters are valid for command
56 for (map<string,string>::iterator it = parameters.begin(); it != parameters.end(); it++) {
57 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
60 //if the user changes the output directory command factory will send this info to us in the output parameter
61 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
63 //make sure the user has already run the read.otu command
64 if ((globaldata->getSharedFile() == "") && (globaldata->getListFile() == "") && (globaldata->getRabundFile() == "") && (globaldata->getSabundFile() == "")) { mothurOut("You must read a list, sabund, rabund or shared file before you can use the collect.single command."); mothurOutEndLine(); abort = true; }
66 //check for optional parameter and set defaults
67 // ...at some point should added some additional type checking...
68 label = validParameter.validFile(parameters, "label", false);
69 if (label == "not found") { label = ""; }
71 if(label != "all") { splitAtDash(label, labels); allLines = 0; }
72 else { allLines = 1; }
75 //if the user has not specified any labels use the ones from read.otu
77 allLines = globaldata->allLines;
78 labels = globaldata->labels;
81 calc = validParameter.validFile(parameters, "calc", false);
82 if (calc == "not found") { calc = "sobs-chao-ace-jack-shannon-npshannon-simpson"; }
84 if (calc == "default") { calc = "sobs-chao-ace-jack-shannon-npshannon-simpson"; }
86 splitAtDash(calc, Estimators);
89 temp = validParameter.validFile(parameters, "freq", false); if (temp == "not found") { temp = "100"; }
92 temp = validParameter.validFile(parameters, "abund", false); if (temp == "not found") { temp = "10"; }
95 temp = validParameter.validFile(parameters, "size", false); if (temp == "not found") { temp = "0"; }
100 catch(exception& e) {
101 errorOut(e, "CollectCommand", "CollectCommand");
105 //**********************************************************************************************************************
107 void CollectCommand::help(){
109 mothurOut("The collect.single command can only be executed after a successful read.otu command. WITH ONE EXECEPTION. \n");
110 mothurOut("The collect.single command can be executed after a successful cluster command. It will use the .list file from the output of the cluster.\n");
111 mothurOut("The collect.single command parameters are label, freq, calc and abund. No parameters are required. \n");
112 mothurOut("The collect.single command should be in the following format: \n");
113 mothurOut("collect.single(label=yourLabel, iters=yourIters, freq=yourFreq, calc=yourEstimators).\n");
114 mothurOut("Example collect(label=unique-.01-.03, iters=10000, freq=10, calc=sobs-chao-ace-jack).\n");
115 mothurOut("The default values for freq is 100, and calc are sobs-chao-ace-jack-shannon-npshannon-simpson.\n");
116 validCalculator->printCalc("single", cout);
117 mothurOut("The label parameter is used to analyze specific labels in your input.\n");
118 mothurOut("Note: No spaces between parameter labels (i.e. freq), '=' and parameters (i.e.yourFreq).\n\n");
120 catch(exception& e) {
121 errorOut(e, "CollectCommand", "help");
126 //**********************************************************************************************************************
128 CollectCommand::~CollectCommand(){}
130 //**********************************************************************************************************************
132 int CollectCommand::execute(){
135 if (abort == true) { return 0; }
137 if ((globaldata->getFormat() != "sharedfile")) { inputFileNames.push_back(globaldata->inputFileName); }
138 else { inputFileNames = parseSharedFile(globaldata->getSharedFile()); globaldata->setFormat("rabund"); }
140 for (int p = 0; p < inputFileNames.size(); p++) {
142 if (outputDir == "") { outputDir += hasPath(inputFileNames[p]); }
143 string fileNameRoot = outputDir + getRootName(getSimpleName(inputFileNames[p]));
144 globaldata->inputFileName = inputFileNames[p];
146 if (inputFileNames.size() > 1) {
147 mothurOutEndLine(); mothurOut("Processing group " + groups[p]); mothurOutEndLine(); mothurOutEndLine();
150 validCalculator = new ValidCalculators();
152 for (int i=0; i<Estimators.size(); i++) {
153 if (validCalculator->isValidCalculator("single", Estimators[i]) == true) {
154 if (Estimators[i] == "sobs") {
155 cDisplays.push_back(new CollectDisplay(new Sobs(), new OneColumnFile(fileNameRoot+"sobs")));
156 }else if (Estimators[i] == "chao") {
157 cDisplays.push_back(new CollectDisplay(new Chao1(), new ThreeColumnFile(fileNameRoot+"chao")));
158 }else if (Estimators[i] == "nseqs") {
159 cDisplays.push_back(new CollectDisplay(new NSeqs(), new OneColumnFile(fileNameRoot+"nseqs")));
160 }else if (Estimators[i] == "coverage") {
161 cDisplays.push_back(new CollectDisplay(new Coverage(), new OneColumnFile(fileNameRoot+"coverage")));
162 }else if (Estimators[i] == "ace") {
163 cDisplays.push_back(new CollectDisplay(new Ace(abund), new ThreeColumnFile(fileNameRoot+"ace")));
164 }else if (Estimators[i] == "jack") {
165 cDisplays.push_back(new CollectDisplay(new Jackknife(), new ThreeColumnFile(fileNameRoot+"jack")));
166 }else if (Estimators[i] == "shannon") {
167 cDisplays.push_back(new CollectDisplay(new Shannon(), new ThreeColumnFile(fileNameRoot+"shannon")));
168 }else if (Estimators[i] == "npshannon") {
169 cDisplays.push_back(new CollectDisplay(new NPShannon(), new OneColumnFile(fileNameRoot+"np_shannon")));
170 }else if (Estimators[i] == "simpson") {
171 cDisplays.push_back(new CollectDisplay(new Simpson(), new ThreeColumnFile(fileNameRoot+"simpson")));
172 }else if (Estimators[i] == "bootstrap") {
173 cDisplays.push_back(new CollectDisplay(new Bootstrap(), new OneColumnFile(fileNameRoot+"bootstrap")));
174 }else if (Estimators[i] == "geometric") {
175 cDisplays.push_back(new CollectDisplay(new Geom(), new OneColumnFile(fileNameRoot+"geometric")));
176 }else if (Estimators[i] == "qstat") {
177 cDisplays.push_back(new CollectDisplay(new QStat(), new OneColumnFile(fileNameRoot+"qstat")));
178 }else if (Estimators[i] == "logseries") {
179 cDisplays.push_back(new CollectDisplay(new LogSD(), new OneColumnFile(fileNameRoot+"logseries")));
180 }else if (Estimators[i] == "bergerparker") {
181 cDisplays.push_back(new CollectDisplay(new BergerParker(), new OneColumnFile(fileNameRoot+"bergerparker")));
182 }else if (Estimators[i] == "bstick") {
183 cDisplays.push_back(new CollectDisplay(new BStick(), new ThreeColumnFile(fileNameRoot+"bstick")));
184 }else if (Estimators[i] == "goodscoverage") {
185 cDisplays.push_back(new CollectDisplay(new GoodsCoverage(), new OneColumnFile(fileNameRoot+"goodscoverage")));
186 }else if (Estimators[i] == "efron") {
187 cDisplays.push_back(new CollectDisplay(new Efron(size), new OneColumnFile(fileNameRoot+"efron")));
188 }else if (Estimators[i] == "boneh") {
189 cDisplays.push_back(new CollectDisplay(new Boneh(size), new OneColumnFile(fileNameRoot+"boneh")));
190 }else if (Estimators[i] == "solow") {
191 cDisplays.push_back(new CollectDisplay(new Solow(size), new OneColumnFile(fileNameRoot+"solow")));
192 }else if (Estimators[i] == "shen") {
193 cDisplays.push_back(new CollectDisplay(new Shen(size, abund), new OneColumnFile(fileNameRoot+"shen")));
199 //if the users entered no valid calculators don't execute command
200 if (cDisplays.size() == 0) { return 0; }
202 read = new ReadOTUFile(inputFileNames[p]);
203 read->read(&*globaldata);
205 order = globaldata->gorder;
206 string lastLabel = order->getLabel();
207 input = globaldata->ginput;
209 //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
210 set<string> processedLabels;
211 set<string> userLabels = labels;
213 while((order != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
215 if(allLines == 1 || labels.count(order->getLabel()) == 1){
217 cCurve = new Collect(order, cDisplays);
218 cCurve->getCurve(freq);
221 mothurOut(order->getLabel()); mothurOutEndLine();
222 processedLabels.insert(order->getLabel());
223 userLabels.erase(order->getLabel());
227 //you have a label the user want that is smaller than this label and the last label has not already been processed
228 if ((anyLabelsToProcess(order->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
229 string saveLabel = order->getLabel();
232 order = (input->getOrderVector(lastLabel));
234 cCurve = new Collect(order, cDisplays);
235 cCurve->getCurve(freq);
238 mothurOut(order->getLabel()); mothurOutEndLine();
239 processedLabels.insert(order->getLabel());
240 userLabels.erase(order->getLabel());
242 //restore real lastlabel to save below
243 order->setLabel(saveLabel);
246 lastLabel = order->getLabel();
249 order = (input->getOrderVector());
252 //output error messages about any remaining user labels
253 set<string>::iterator it;
254 bool needToRun = false;
255 for (it = userLabels.begin(); it != userLabels.end(); it++) {
256 mothurOut("Your file does not include the label " + *it);
257 if (processedLabels.count(lastLabel) != 1) {
258 mothurOut(". I will use " + lastLabel + "."); mothurOutEndLine();
261 mothurOut(". Please refer to " + lastLabel + "."); mothurOutEndLine();
265 //run last label if you need to
266 if (needToRun == true) {
267 if (order != NULL) { delete order; }
268 order = (input->getOrderVector(lastLabel));
270 mothurOut(order->getLabel()); mothurOutEndLine();
272 cCurve = new Collect(order, cDisplays);
273 cCurve->getCurve(freq);
278 for(int i=0;i<cDisplays.size();i++){ delete cDisplays[i]; }
280 delete input; globaldata->ginput = NULL;
282 globaldata->gorder = NULL;
283 delete validCalculator;
290 catch(exception& e) {
291 errorOut(e, "CollectCommand", "execute");
296 //**********************************************************************************************************************
297 vector<string> CollectCommand::parseSharedFile(string filename) {
299 vector<string> filenames;
301 map<string, ofstream*> filehandles;
302 map<string, ofstream*>::iterator it3;
306 read = new ReadOTUFile(filename);
307 read->read(&*globaldata);
309 input = globaldata->ginput;
310 vector<SharedRAbundVector*> lookup = input->getSharedRAbundVectors();
312 string sharedFileRoot = getRootName(filename);
314 //clears file before we start to write to it below
315 for (int i=0; i<lookup.size(); i++) {
316 remove((sharedFileRoot + lookup[i]->getGroup() + ".rabund").c_str());
317 filenames.push_back((sharedFileRoot + lookup[i]->getGroup() + ".rabund"));
321 for (int i=0; i<lookup.size(); i++) {
323 filehandles[lookup[i]->getGroup()] = temp;
324 groups.push_back(lookup[i]->getGroup());
327 while(lookup[0] != NULL) {
329 for (int i = 0; i < lookup.size(); i++) {
330 RAbundVector rav = lookup[i]->getRAbundVector();
331 openOutputFileAppend(sharedFileRoot + lookup[i]->getGroup() + ".rabund", *(filehandles[lookup[i]->getGroup()]));
332 rav.print(*(filehandles[lookup[i]->getGroup()]));
333 (*(filehandles[lookup[i]->getGroup()])).close();
336 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
337 lookup = input->getSharedRAbundVectors();
341 for (it3 = filehandles.begin(); it3 != filehandles.end(); it3++) {
346 globaldata->ginput = NULL;
350 catch(exception& e) {
351 errorOut(e, "CollectCommand", "parseSharedFile");
355 //**********************************************************************************************************************