2 // GetCoreMicroBiomeCommand.cpp
5 // Created by John Westcott on 5/8/12.
6 // Copyright (c) 2012 Schloss Lab. All rights reserved.
9 #include "GetCoreMicroBiomeCommand.h"
12 //**********************************************************************************************************************
13 vector<string> GetCoreMicroBiomeCommand::setParameters(){
15 CommandParameter pshared("shared", "InputTypes", "", "", "SharedRel", "SharedRel", "none",false,false); parameters.push_back(pshared);
16 CommandParameter prelabund("relabund", "InputTypes", "", "", "SharedRel", "SharedRel", "none",false,false); parameters.push_back(prelabund);
17 CommandParameter pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups);
18 CommandParameter plabel("label", "String", "", "", "", "", "",false,false); parameters.push_back(plabel);
19 CommandParameter poutput("output", "Multiple", "fraction-count", "fraction", "", "", "",false,false); parameters.push_back(poutput);
20 CommandParameter pabund("abundance", "Number", "", "-1", "", "", "",false,false); parameters.push_back(pabund);
21 CommandParameter psamples("samples", "Number", "", "-1", "", "", "",false,false); parameters.push_back(psamples);
22 CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
23 CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
25 vector<string> myArray;
26 for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
30 m->errorOut(e, "GetCoreMicroBiomeCommand", "setParameters");
34 //**********************************************************************************************************************
35 string GetCoreMicroBiomeCommand::getHelpString(){
37 string helpString = "";
38 helpString += "The get.coremicrobiome determines the fraction of OTUs that are found in varying numbers of samples for different minimum relative abundances.\n";
39 helpString += "The get.coremicrobiome parameters are: shared, relabund, groups, label, output, abundance and samples. Shared or relabund is required.\n";
40 helpString += "The label parameter is used to analyze specific labels in your input.\n";
41 helpString += "The groups parameter allows you to specify which of the groups you would like analyzed.\n";
42 helpString += "The output parameter is used to specify whether you would like the fraction of OTU's or OTU count outputted. Options are fraction or count. Default=fraction.\n";
43 helpString += "The abundance parameter allows you to specify an abundance you would like the OTU names outputted for. Must be an integer between 0 and 100, indicating the relative abundance. \n";
44 helpString += "The samples parameter allows you to specify the minimum number of samples you would like the OTU names outputted for. Must be an interger between 1 and number of samples in your file.\n";
45 helpString += "The new command should be in the following format: get.coremicrobiome(shared=yourSharedFile)\n";
46 helpString += "get.coremicrobiom(shared=final.an.shared, abund=30)\n";
50 m->errorOut(e, "GetCoreMicroBiomeCommand", "getHelpString");
54 //**********************************************************************************************************************
55 GetCoreMicroBiomeCommand::GetCoreMicroBiomeCommand(){
57 abort = true; calledHelp = true;
59 vector<string> tempOutNames;
60 outputTypes["coremicrobiom"] = tempOutNames;
63 m->errorOut(e, "GetCoreMicroBiomeCommand", "GetCoreMicroBiomeCommand");
67 //**********************************************************************************************************************
68 GetCoreMicroBiomeCommand::GetCoreMicroBiomeCommand(string option) {
70 abort = false; calledHelp = false; allLines = 1;
72 //allow user to run help
73 if(option == "help") { help(); abort = true; calledHelp = true; }
74 else if(option == "citation") { citation(); abort = true; calledHelp = true;}
77 //valid paramters for this command
78 vector<string> myArray = setParameters();
80 OptionParser parser(option);
81 map<string,string> parameters = parser.getParameters();
83 ValidParameters validParameter;
84 map<string,string>::iterator it;
85 //check to make sure all parameters are valid for command
86 for (it = parameters.begin(); it != parameters.end(); it++) {
87 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
91 //if the user changes the input directory command factory will send this info to us in the output parameter
92 string inputDir = validParameter.validFile(parameters, "inputdir", false);
93 if (inputDir == "not found"){ inputDir = ""; }
97 it = parameters.find("relabund");
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["relabund"] = inputDir + it->second; }
105 it = parameters.find("shared");
106 //user has given a template file
107 if(it != parameters.end()){
108 path = m->hasPath(it->second);
109 //if the user has not given a path then, add inputdir. else leave path alone.
110 if (path == "") { parameters["shared"] = inputDir + it->second; }
115 //check for parameters
116 sharedfile = validParameter.validFile(parameters, "shared", true);
117 if (sharedfile == "not open") { abort = true; }
118 else if (sharedfile == "not found") { sharedfile = ""; }
119 else { inputFileName = sharedfile; format = "sharedfile"; m->setSharedFile(sharedfile); }
121 relabundfile = validParameter.validFile(parameters, "relabund", true);
122 if (relabundfile == "not open") { abort = true; }
123 else if (relabundfile == "not found") { relabundfile = ""; }
124 else { inputFileName = relabundfile; format = "relabund"; m->setRelAbundFile(relabundfile); }
126 if ((relabundfile == "") && (sharedfile == "")) {
127 //is there are current file available for either of these?
128 //give priority to shared, then relabund
129 sharedfile = m->getSharedFile();
130 if (sharedfile != "") { inputFileName = sharedfile; format="sharedfile"; m->mothurOut("Using " + sharedfile + " as input file for the shared parameter."); m->mothurOutEndLine(); }
132 relabundfile = m->getRelAbundFile();
133 if (relabundfile != "") { inputFileName = relabundfile; format="relabund"; m->mothurOut("Using " + relabundfile + " as input file for the relabund parameter."); m->mothurOutEndLine(); }
135 m->mothurOut("No valid current files. You must provide a shared or relabund."); m->mothurOutEndLine();
141 //if the user changes the output directory command factory will send this info to us in the output parameter
142 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
143 outputDir = m->hasPath(inputFileName); //if user entered a file with a path then preserve it
146 string groups = validParameter.validFile(parameters, "groups", false);
147 if (groups == "not found") { groups = ""; }
148 else { m->splitAtDash(groups, Groups); }
149 m->setGroups(Groups);
151 string label = validParameter.validFile(parameters, "label", false);
152 if (label == "not found") { label = ""; }
154 if(label != "all") { m->splitAtDash(label, labels); allLines = 0; }
155 else { allLines = 1; }
158 output = validParameter.validFile(parameters, "output", false); if(output == "not found"){ output = "fraction"; }
160 if ((output != "fraction") && (output != "count")) { m->mothurOut(output + " is not a valid output form. Options are fraction and count. I will use fraction."); m->mothurOutEndLine(); output = "fraction"; }
162 string temp = validParameter.validFile(parameters, "abundance", false); if (temp == "not found"){ temp = "-1"; }
163 m->mothurConvert(temp, abund);
166 if ((abund < 0) || (abund > 100)) { m->mothurOut(toString(abund) + " is not a valid number for abund. Must be an integer between 0 and 100.\n"); }
169 temp = validParameter.validFile(parameters, "samples", false); if (temp == "not found"){ temp = "-1"; }
170 m->mothurConvert(temp, samples);
175 catch(exception& e) {
176 m->errorOut(e, "GetCoreMicroBiomeCommand", "GetCoreMicroBiomeCommand");
180 //**********************************************************************************************************************
182 int GetCoreMicroBiomeCommand::execute(){
185 if (abort == true) { if (calledHelp) { return 0; } return 2; }
187 InputData input(inputFileName, format);
188 vector<SharedRAbundFloatVector*> lookup = input.getSharedRAbundFloatVectors();
189 string lastLabel = lookup[0]->getLabel();
192 if ((samples < 1) || (samples > lookup.size())) { m->mothurOut(toString(samples) + " is not a valid number for samples. Must be an integer between 1 and the number of samples in your file. Your file contains " + toString(lookup.size()) + " samples, so I will use that.\n"); samples = lookup.size(); }
196 //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
197 set<string> processedLabels;
198 set<string> userLabels = labels;
200 //as long as you are not at the end of the file or done wih the lines you want
201 while((lookup[0] != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
203 if (m->control_pressed) { for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; } for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; }
205 if(allLines == 1 || labels.count(lookup[0]->getLabel()) == 1){
207 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
211 processedLabels.insert(lookup[0]->getLabel());
212 userLabels.erase(lookup[0]->getLabel());
215 if ((m->anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
216 string saveLabel = lookup[0]->getLabel();
218 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
219 lookup = input.getSharedRAbundFloatVectors(lastLabel);
220 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
224 processedLabels.insert(lookup[0]->getLabel());
225 userLabels.erase(lookup[0]->getLabel());
227 //restore real lastlabel to save below
228 lookup[0]->setLabel(saveLabel);
231 lastLabel = lookup[0]->getLabel();
232 //prevent memory leak
233 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; lookup[i] = NULL; }
235 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; }
237 //get next line to process
238 lookup = input.getSharedRAbundFloatVectors();
241 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; }
243 //output error messages about any remaining user labels
244 set<string>::iterator it;
245 bool needToRun = false;
246 for (it = userLabels.begin(); it != userLabels.end(); it++) {
247 m->mothurOut("Your file does not include the label " + *it);
248 if (processedLabels.count(lastLabel) != 1) {
249 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
252 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
256 //run last label if you need to
257 if (needToRun == true) {
258 for (int i = 0; i < lookup.size(); i++) { if (lookup[i] != NULL) { delete lookup[i]; } }
259 lookup = input.getSharedRAbundFloatVectors(lastLabel);
261 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
265 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
268 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; }
270 //output files created by command
271 m->mothurOutEndLine();
272 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
273 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
274 m->mothurOutEndLine();
278 catch(exception& e) {
279 m->errorOut(e, "GetCoreMicroBiomeCommand", "execute");
283 //**********************************************************************************************************************
285 int GetCoreMicroBiomeCommand::createTable(vector<SharedRAbundFloatVector*>& lookup){
288 string outputFileName = outputDir + m->getRootName(m->getSimpleName(inputFileName)) + lookup[0]->getLabel() + ".core.microbiom";
289 outputNames.push_back(outputFileName); outputTypes["coremicrobiom"].push_back(outputFileName);
291 m->openOutputFile(outputFileName, out);
293 int numSamples = lookup.size();
294 int numOtus = lookup[0]->getNumBins();
296 //table is 100 by numsamples
297 //question we are answering is: what fraction of OTUs in a study have a relative abundance at or above %X
298 //in at least %Y samples. x goes from 0 to 100, y from 1 to numSamples
299 vector< vector<double> > table; table.resize(101);
300 for (int i = 0; i < table.size(); i++) { table[i].resize(numSamples, 0.0); }
302 map<int, vector<string> > otuNames;
303 if ((abund != -1) && (samples == -1)) { //fill with all samples
304 for (int i = 0; i < numSamples; i++) {
306 otuNames[i+1] = temp;
308 }else if ((abund == -1) && (samples != -1)) { //fill with all relabund
309 for (int i = 0; i < 101; i++) {
313 }else if ((abund != -1) && (samples != -1)) { //only one line is wanted
315 otuNames[abund] = temp;
318 for (int i = 0; i < numOtus; i++) {
320 if (m->control_pressed) { break; }
322 //count number of samples in this otu with a relabund >= spot in count
323 vector<int> counts; counts.resize(101, 0);
325 for (int j = 0; j < lookup.size(); j++) {
326 double relabund = lookup[j]->getAbundance(i);
327 int wholeRelabund = (int) (floor(relabund*100));
328 for (int k = 0; k < wholeRelabund+1; k++) { counts[k]++; }
331 //add this otus info to table
332 for (int j = 0; j < table.size(); j++) {
333 for (int k = 0; k < counts[j]; k++) { table[j][k]++; }
335 if ((abund == -1) && (samples != -1)) { //we want all OTUs with this number of samples
336 if (counts[j] >= samples) { otuNames[j].push_back(m->currentBinLabels[i]); }
337 }else if ((abund != -1) && (samples == -1)) { //we want all OTUs with this relabund
339 for (int k = 0; k < counts[j]; k++) { otuNames[k+1].push_back(m->currentBinLabels[i]); }
341 }else if ((abund != -1) && (samples != -1)) { //we want only OTUs with this relabund for this number of samples
342 if ((j == abund) && (counts[j] >= samples)) {
343 otuNames[j].push_back(m->currentBinLabels[i]);
350 if (output == "fraction") { out.setf(ios::fixed, ios::floatfield); out.setf(ios::showpoint); }
351 out << "NumSamples\t";
353 //convert table counts to percents
354 for (int i = 0; i < table.size(); i++) {
355 out << "Relabund-" << i << "%\t";
356 if (m->control_pressed) { break; }
357 for (int j = 0; j < table[i].size(); j++) { if (output == "fraction") { table[i][j] /= (double) numOtus; } }
361 for (int i = 0; i < numSamples; i++) {
362 if (m->control_pressed) { break; }
364 for (int j = 0; j < table.size(); j++) { out << table[j][i] << '\t'; }
370 if (m->control_pressed) { return 0; }
372 if ((samples != -1) || (abund != -1)) {
373 string outputFileName2 = outputDir + m->getRootName(m->getSimpleName(inputFileName)) + lookup[0]->getLabel() + ".core.microbiomlist";
374 outputNames.push_back(outputFileName2); outputTypes["coremicrobiom"].push_back(outputFileName2);
376 m->openOutputFile(outputFileName2, out2);
378 if ((abund == -1) && (samples != -1)) { //we want all OTUs with this number of samples
379 out2 << "Relabund%\tOTUList_for_samples=" << samples << "\n";
380 }else if ((abund != -1) && (samples == -1)) { //we want all OTUs with this relabund
381 out2 << "Samples\tOTUList_for_abund=" << abund << "\n";
382 }else if ((abund != -1) && (samples != -1)) { //we want only OTUs with this relabund for this number of samples
383 out2 << "Relabund%\tOTUList_for_samples=" << samples << "\n";
386 for (map<int, vector<string> >::iterator it = otuNames.begin(); it != otuNames.end(); it++) {
387 if (m->control_pressed) { break; }
389 vector<string> temp = it->second;
390 string list = m->makeList(temp);
392 out2 << it->first << '\t' << list << endl;
400 catch(exception& e) {
401 m->errorOut(e, "GetCoreMicroBiomeCommand", "createTable");
406 //**********************************************************************************************************************