2 * matrixoutputcommand.cpp
5 * Created by Sarah Westcott on 5/20/09.
6 * Copyright 2009 Schloss Lab UMASS Amherst. All rights reserved.
10 #include "matrixoutputcommand.h"
11 #include "sharedjabund.h"
12 #include "sharedsorabund.h"
13 #include "sharedjclass.h"
14 #include "sharedsorclass.h"
15 #include "sharedjest.h"
16 #include "sharedsorest.h"
17 #include "sharedthetayc.h"
18 #include "sharedthetan.h"
19 #include "sharedmorisitahorn.h"
20 #include "sharedbraycurtis.h"
22 //**********************************************************************************************************************
23 vector<string> MatrixOutputCommand::getValidParameters(){
25 string Array[] = {"label","calc","groups","outputdir","inputdir", "output"};
26 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
30 m->errorOut(e, "MatrixOutputCommand", "getValidParameters");
34 //**********************************************************************************************************************
35 MatrixOutputCommand::MatrixOutputCommand(){
37 //initialize outputTypes
38 vector<string> tempOutNames;
39 outputTypes["phylip"] = tempOutNames;
42 m->errorOut(e, "MatrixOutputCommand", "MatrixOutputCommand");
46 //**********************************************************************************************************************
47 vector<string> MatrixOutputCommand::getRequiredParameters(){
49 vector<string> myArray;
53 m->errorOut(e, "MatrixOutputCommand", "getRequiredParameters");
57 //**********************************************************************************************************************
58 vector<string> MatrixOutputCommand::getRequiredFiles(){
60 string Array[] = {"shared"};
61 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
65 m->errorOut(e, "MatrixOutputCommand", "getRequiredFiles");
69 //**********************************************************************************************************************
71 MatrixOutputCommand::MatrixOutputCommand(string option) {
73 globaldata = GlobalData::getInstance();
80 //allow user to run help
81 if(option == "help") { validCalculator = new ValidCalculators(); help(); abort = true; }
84 //valid paramters for this command
85 string Array[] = {"label","calc","groups","outputdir","inputdir", "output"};
86 vector<string> myArray (Array, Array+(sizeof(Array)/sizeof(string)));
88 OptionParser parser(option);
89 map<string,string> parameters = parser.getParameters();
91 ValidParameters validParameter;
93 //check to make sure all parameters are valid for command
94 for (map<string,string>::iterator it = parameters.begin(); it != parameters.end(); it++) {
95 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
98 //initialize outputTypes
99 vector<string> tempOutNames;
100 outputTypes["phylip"] = tempOutNames;
102 //if the user changes the output directory command factory will send this info to us in the output parameter
103 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
105 outputDir += m->hasPath(globaldata->inputFileName); //if user entered a file with a path then preserve it
108 //make sure the user has already run the read.otu command
109 if (globaldata->getSharedFile() == "") {
110 if (globaldata->getListFile() == "") { m->mothurOut("You must read a list and a group, or a shared before you can use the dist.shared command."); m->mothurOutEndLine(); abort = true; }
111 else if (globaldata->getGroupFile() == "") { m->mothurOut("You must read a list and a group, or a shared before you can use the dist.shared command."); m->mothurOutEndLine(); abort = true; }
114 //check for optional parameter and set defaults
115 // ...at some point should added some additional type checking...
116 label = validParameter.validFile(parameters, "label", false);
117 if (label == "not found") { label = ""; }
119 if(label != "all") { m->splitAtDash(label, labels); allLines = 0; }
120 else { allLines = 1; }
123 output = validParameter.validFile(parameters, "output", false); if(output == "not found"){ output = "lt"; }
124 if ((output != "lt") && (output != "square")) { m->mothurOut(output + " is not a valid output form. Options are lt and square. I will use lt."); m->mothurOutEndLine(); output = "lt"; }
126 //if the user has not specified any labels use the ones from read.otu
128 allLines = globaldata->allLines;
129 labels = globaldata->labels;
132 groups = validParameter.validFile(parameters, "groups", false);
133 if (groups == "not found") { groups = ""; }
135 m->splitAtDash(groups, Groups);
136 globaldata->Groups = Groups;
139 calc = validParameter.validFile(parameters, "calc", false);
140 if (calc == "not found") { calc = "jclass-thetayc"; }
142 if (calc == "default") { calc = "jclass-thetayc"; }
144 m->splitAtDash(calc, Estimators);
146 if (abort == false) {
148 validCalculator = new ValidCalculators();
151 for (i=0; i<Estimators.size(); i++) {
152 if (validCalculator->isValidCalculator("matrix", Estimators[i]) == true) {
153 if (Estimators[i] == "jabund") {
154 matrixCalculators.push_back(new JAbund());
155 }else if (Estimators[i] == "sorabund") {
156 matrixCalculators.push_back(new SorAbund());
157 }else if (Estimators[i] == "jclass") {
158 matrixCalculators.push_back(new Jclass());
159 }else if (Estimators[i] == "sorclass") {
160 matrixCalculators.push_back(new SorClass());
161 }else if (Estimators[i] == "jest") {
162 matrixCalculators.push_back(new Jest());
163 }else if (Estimators[i] == "sorest") {
164 matrixCalculators.push_back(new SorEst());
165 }else if (Estimators[i] == "thetayc") {
166 matrixCalculators.push_back(new ThetaYC());
167 }else if (Estimators[i] == "thetan") {
168 matrixCalculators.push_back(new ThetaN());
169 }else if (Estimators[i] == "morisitahorn") {
170 matrixCalculators.push_back(new MorHorn());
171 }else if (Estimators[i] == "braycurtis") {
172 matrixCalculators.push_back(new BrayCurtis());
181 catch(exception& e) {
182 m->errorOut(e, "MatrixOutputCommand", "MatrixOutputCommand");
187 //**********************************************************************************************************************
189 void MatrixOutputCommand::help(){
191 m->mothurOut("The dist.shared command can only be executed after a successful read.otu command.\n");
192 m->mothurOut("The dist.shared command parameters are groups, calc, output and label. No parameters are required.\n");
193 m->mothurOut("The groups parameter allows you to specify which of the groups in your groupfile you would like included used.\n");
194 m->mothurOut("The group names are separated by dashes. The label parameter allows you to select what distance levels you would like distance matrices created for, and is also separated by dashes.\n");
195 m->mothurOut("The dist.shared command should be in the following format: dist.shared(groups=yourGroups, calc=yourCalcs, label=yourLabels).\n");
196 m->mothurOut("The output parameter allows you to specify format of your distance matrix. Options are lt, and square. The default is lt.\n");
197 m->mothurOut("Example dist.shared(groups=A-B-C, calc=jabund-sorabund).\n");
198 m->mothurOut("The default value for groups is all the groups in your groupfile.\n");
199 m->mothurOut("The default value for calc is jclass and thetayc.\n");
200 validCalculator->printCalc("matrix", cout);
201 m->mothurOut("The dist.shared command outputs a .dist file for each calculator you specify at each distance you choose.\n");
202 m->mothurOut("Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n\n");
204 catch(exception& e) {
205 m->errorOut(e, "MatrixOutputCommand", "help");
211 //**********************************************************************************************************************
213 MatrixOutputCommand::~MatrixOutputCommand(){
214 if (abort == false) {
215 delete input; globaldata->ginput = NULL;
217 delete validCalculator;
221 //**********************************************************************************************************************
223 int MatrixOutputCommand::execute(){
226 if (abort == true) { return 0; }
228 //if the users entered no valid calculators don't execute command
229 if (matrixCalculators.size() == 0) { m->mothurOut("No valid calculators."); m->mothurOutEndLine(); return 0; }
232 read = new ReadOTUFile(globaldata->inputFileName);
233 read->read(&*globaldata);
235 input = globaldata->ginput;
236 lookup = input->getSharedRAbundVectors();
237 string lastLabel = lookup[0]->getLabel();
239 //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
240 set<string> processedLabels;
241 set<string> userLabels = labels;
243 if (lookup.size() < 2) { m->mothurOut("You have not provided enough valid groups. I cannot run the command."); m->mothurOutEndLine(); return 0;}
245 numGroups = lookup.size();
247 if (m->control_pressed) { delete read; delete input; globaldata->ginput = NULL; for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } globaldata->Groups.clear(); return 0; }
249 //as long as you are not at the end of the file or done wih the lines you want
250 while((lookup[0] != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
252 if (m->control_pressed) { outputTypes.clear(); delete read; delete input; globaldata->ginput = NULL; for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; }
254 if(allLines == 1 || labels.count(lookup[0]->getLabel()) == 1){
255 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
258 processedLabels.insert(lookup[0]->getLabel());
259 userLabels.erase(lookup[0]->getLabel());
262 if ((m->anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
263 string saveLabel = lookup[0]->getLabel();
265 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
266 lookup = input->getSharedRAbundVectors(lastLabel);
268 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
271 processedLabels.insert(lookup[0]->getLabel());
272 userLabels.erase(lookup[0]->getLabel());
274 //restore real lastlabel to save below
275 lookup[0]->setLabel(saveLabel);
278 lastLabel = lookup[0]->getLabel();
280 //get next line to process
281 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
282 lookup = input->getSharedRAbundVectors();
285 if (m->control_pressed) { outputTypes.clear(); delete read; delete input; globaldata->ginput = NULL; for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; }
287 //output error messages about any remaining user labels
288 set<string>::iterator it;
289 bool needToRun = false;
290 for (it = userLabels.begin(); it != userLabels.end(); it++) {
291 m->mothurOut("Your file does not include the label " + *it);
292 if (processedLabels.count(lastLabel) != 1) {
293 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
296 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
300 if (m->control_pressed) { outputTypes.clear(); delete read; delete input; globaldata->ginput = NULL; for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; }
302 //run last label if you need to
303 if (needToRun == true) {
304 for (int i = 0; i < lookup.size(); i++) { if (lookup[i] != NULL) { delete lookup[i]; } }
305 lookup = input->getSharedRAbundVectors(lastLabel);
307 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
309 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
312 if (m->control_pressed) { outputTypes.clear(); delete read; delete input; globaldata->ginput = NULL; for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); } globaldata->Groups.clear(); return 0; }
314 //reset groups parameter
315 globaldata->Groups.clear();
317 m->mothurOutEndLine();
318 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
319 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
320 m->mothurOutEndLine();
325 catch(exception& e) {
326 m->errorOut(e, "MatrixOutputCommand", "execute");
330 /***********************************************************/
331 void MatrixOutputCommand::printSims(ostream& out) {
334 out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint);
337 out << simMatrix.size() << endl;
339 if (output == "lt") {
340 for (int m = 0; m < simMatrix.size(); m++) {
341 out << lookup[m]->getGroup() << '\t';
342 for (int n = 0; n < m; n++) {
343 out << simMatrix[m][n] << '\t';
348 for (int m = 0; m < simMatrix.size(); m++) {
349 out << lookup[m]->getGroup() << '\t';
350 for (int n = 0; n < simMatrix[m].size(); n++) {
351 out << simMatrix[m][n] << '\t';
357 catch(exception& e) {
358 m->errorOut(e, "MatrixOutputCommand", "printSims");
362 /***********************************************************/
363 int MatrixOutputCommand::process(vector<SharedRAbundVector*> thisLookup){
367 vector<SharedRAbundVector*> subset;
369 //for each calculator
370 for(int i = 0 ; i < matrixCalculators.size(); i++) {
372 //initialize simMatrix
374 simMatrix.resize(numGroups);
375 for (int p = 0; p < simMatrix.size(); p++) {
376 for (int j = 0; j < simMatrix.size(); j++) {
377 simMatrix[p].push_back(0.0);
381 for (int k = 0; k < thisLookup.size(); k++) {
382 for (int l = k; l < thisLookup.size(); l++) {
383 if (k != l) { //we dont need to similiarity of a groups to itself
384 //get estimated similarity between 2 groups
386 if (m->control_pressed) { return 0; }
388 subset.clear(); //clear out old pair of sharedrabunds
389 //add new pair of sharedrabunds
390 subset.push_back(thisLookup[k]); subset.push_back(thisLookup[l]);
392 data = matrixCalculators[i]->getValues(subset); //saves the calculator outputs
393 //save values in similarity matrix
394 simMatrix[k][l] = 1.0 - data[0]; //convert similiarity to distance
395 simMatrix[l][k] = 1.0 - data[0]; //convert similiarity to distance
400 exportFileName = outputDir + m->getRootName(m->getSimpleName(globaldata->inputFileName)) + matrixCalculators[i]->getName() + "." + thisLookup[0]->getLabel() + "." + output + ".dist";
401 m->openOutputFile(exportFileName, out);
402 outputNames.push_back(exportFileName); outputTypes["phylip"].push_back(exportFileName);
412 catch(exception& e) {
413 m->errorOut(e, "MatrixOutputCommand", "process");
417 /***********************************************************/