5 * Created by westcott on 10/27/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "subsamplecommand.h"
11 #include "sharedutilities.h"
13 //**********************************************************************************************************************
14 vector<string> SubSampleCommand::setParameters(){
16 CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "FLSSR", "none",false,false); parameters.push_back(pfasta);
17 CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname);
18 CommandParameter pgroup("group", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pgroup);
19 CommandParameter plist("list", "InputTypes", "", "", "none", "FLSSR", "none",false,false); parameters.push_back(plist);
20 CommandParameter pshared("shared", "InputTypes", "", "", "none", "FLSSR", "none",false,false); parameters.push_back(pshared);
21 CommandParameter prabund("rabund", "InputTypes", "", "", "none", "FLSSR", "none",false,false); parameters.push_back(prabund);
22 CommandParameter psabund("sabund", "InputTypes", "", "", "none", "FLSSR", "none",false,false); parameters.push_back(psabund);
23 CommandParameter plabel("label", "String", "", "", "", "", "",false,false); parameters.push_back(plabel);
24 CommandParameter pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups);
25 CommandParameter psize("size", "Number", "", "0", "", "", "",false,false); parameters.push_back(psize);
26 CommandParameter ppersample("persample", "Boolean", "", "F", "", "", "",false,false); parameters.push_back(ppersample);
27 CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
28 CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
30 vector<string> myArray;
31 for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
35 m->errorOut(e, "SubSampleCommand", "setParameters");
39 //**********************************************************************************************************************
40 string SubSampleCommand::getHelpString(){
42 string helpString = "";
43 helpString += "The sub.sample command is designed to be used as a way to normalize your data, or create a smaller set from your original set.\n";
44 helpString += "The sub.sample command parameters are fasta, name, list, group, rabund, sabund, shared, groups, size, persample and label. You must provide a fasta, list, sabund, rabund or shared file as an input file.\n";
45 helpString += "The namefile is only used with the fasta file, not with the listfile, because the list file should contain all sequences.\n";
46 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";
47 helpString += "The label parameter allows you to select what distance levels you would like, and are also separated by dashes.\n";
48 helpString += "The size parameter allows you indicate the size of your subsample.\n";
49 helpString += "The persample parameter allows you indicate you want to select subsample of the same size from each of your groups, default=false. It is only used with the list and fasta files if a groupfile is given.\n";
50 helpString += "persample=false will select a random set of sequences of the size you select, but the number of seqs from each group may differ.\n";
51 helpString += "The size parameter is not set: with shared file size=number of seqs in smallest sample, with all other files if a groupfile is given and persample=true, then size=number of seqs in smallest sample, otherwise size=10% of number of seqs.\n";
52 helpString += "The sub.sample command should be in the following format: sub.sample(list=yourListFile, group=yourGroupFile, groups=yourGroups, label=yourLabels).\n";
53 helpString += "Example sub.sample(list=abrecovery.fn.list, group=abrecovery.groups, groups=B-C, size=20).\n";
54 helpString += "The default value for groups is all the groups in your groupfile, and all labels in your inputfile will be used.\n";
55 helpString += "The sub.sample command outputs a .subsample file.\n";
56 helpString += "Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n";
60 m->errorOut(e, "SubSampleCommand", "getHelpString");
64 //**********************************************************************************************************************
65 SubSampleCommand::SubSampleCommand(){
67 abort = true; calledHelp = true;
69 vector<string> tempOutNames;
70 outputTypes["shared"] = tempOutNames;
71 outputTypes["list"] = tempOutNames;
72 outputTypes["rabund"] = tempOutNames;
73 outputTypes["sabund"] = tempOutNames;
74 outputTypes["fasta"] = tempOutNames;
75 outputTypes["name"] = tempOutNames;
76 outputTypes["group"] = tempOutNames;
79 m->errorOut(e, "SubSampleCommand", "GetRelAbundCommand");
83 //**********************************************************************************************************************
84 SubSampleCommand::SubSampleCommand(string option) {
86 abort = false; calledHelp = false;
89 //allow user to run help
90 if(option == "help") { help(); abort = true; calledHelp = true; }
91 else if(option == "citation") { citation(); abort = true; calledHelp = true;}
94 vector<string> myArray = setParameters();
96 OptionParser parser(option);
97 map<string,string> parameters = parser.getParameters();
99 ValidParameters validParameter;
101 //check to make sure all parameters are valid for command
102 map<string,string>::iterator it;
103 for (it = parameters.begin(); it != parameters.end(); it++) {
104 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
107 //initialize outputTypes
108 vector<string> tempOutNames;
109 outputTypes["shared"] = tempOutNames;
110 outputTypes["list"] = tempOutNames;
111 outputTypes["rabund"] = tempOutNames;
112 outputTypes["sabund"] = tempOutNames;
113 outputTypes["fasta"] = tempOutNames;
114 outputTypes["name"] = tempOutNames;
115 outputTypes["group"] = tempOutNames;
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 = ""; }
120 //if the user changes the input directory command factory will send this info to us in the output parameter
121 string inputDir = validParameter.validFile(parameters, "inputdir", false);
122 if (inputDir == "not found"){ inputDir = ""; }
125 it = parameters.find("list");
126 //user has given a template file
127 if(it != parameters.end()){
128 path = m->hasPath(it->second);
129 //if the user has not given a path then, add inputdir. else leave path alone.
130 if (path == "") { parameters["list"] = inputDir + it->second; }
133 it = parameters.find("fasta");
134 //user has given a template file
135 if(it != parameters.end()){
136 path = m->hasPath(it->second);
137 //if the user has not given a path then, add inputdir. else leave path alone.
138 if (path == "") { parameters["fasta"] = inputDir + it->second; }
141 it = parameters.find("shared");
142 //user has given a template file
143 if(it != parameters.end()){
144 path = m->hasPath(it->second);
145 //if the user has not given a path then, add inputdir. else leave path alone.
146 if (path == "") { parameters["shared"] = inputDir + it->second; }
149 it = parameters.find("group");
150 //user has given a template file
151 if(it != parameters.end()){
152 path = m->hasPath(it->second);
153 //if the user has not given a path then, add inputdir. else leave path alone.
154 if (path == "") { parameters["group"] = inputDir + it->second; }
157 it = parameters.find("sabund");
158 //user has given a template file
159 if(it != parameters.end()){
160 path = m->hasPath(it->second);
161 //if the user has not given a path then, add inputdir. else leave path alone.
162 if (path == "") { parameters["sabund"] = inputDir + it->second; }
165 it = parameters.find("rabund");
166 //user has given a template file
167 if(it != parameters.end()){
168 path = m->hasPath(it->second);
169 //if the user has not given a path then, add inputdir. else leave path alone.
170 if (path == "") { parameters["rabund"] = inputDir + it->second; }
173 it = parameters.find("name");
174 //user has given a template file
175 if(it != parameters.end()){
176 path = m->hasPath(it->second);
177 //if the user has not given a path then, add inputdir. else leave path alone.
178 if (path == "") { parameters["name"] = inputDir + it->second; }
182 //check for required parameters
183 listfile = validParameter.validFile(parameters, "list", true);
184 if (listfile == "not open") { listfile = ""; abort = true; }
185 else if (listfile == "not found") { listfile = ""; }
186 else { m->setListFile(listfile); }
188 sabundfile = validParameter.validFile(parameters, "sabund", true);
189 if (sabundfile == "not open") { sabundfile = ""; abort = true; }
190 else if (sabundfile == "not found") { sabundfile = ""; }
191 else { m->setSabundFile(sabundfile); }
193 rabundfile = validParameter.validFile(parameters, "rabund", true);
194 if (rabundfile == "not open") { rabundfile = ""; abort = true; }
195 else if (rabundfile == "not found") { rabundfile = ""; }
196 else { m->setRabundFile(rabundfile); }
198 fastafile = validParameter.validFile(parameters, "fasta", true);
199 if (fastafile == "not open") { fastafile = ""; abort = true; }
200 else if (fastafile == "not found") { fastafile = ""; }
201 else { m->setFastaFile(fastafile); }
203 sharedfile = validParameter.validFile(parameters, "shared", true);
204 if (sharedfile == "not open") { sharedfile = ""; abort = true; }
205 else if (sharedfile == "not found") { sharedfile = ""; }
206 else { m->setSharedFile(sharedfile); }
208 namefile = validParameter.validFile(parameters, "name", true);
209 if (namefile == "not open") { namefile = ""; abort = true; }
210 else if (namefile == "not found") { namefile = ""; }
211 else { m->setNameFile(namefile); }
213 groupfile = validParameter.validFile(parameters, "group", true);
214 if (groupfile == "not open") { groupfile = ""; abort = true; }
215 else if (groupfile == "not found") { groupfile = ""; }
216 else { m->setGroupFile(groupfile); }
218 //check for optional parameter and set defaults
219 // ...at some point should added some additional type checking...
220 label = validParameter.validFile(parameters, "label", false);
221 if (label == "not found") { label = ""; }
223 if(label != "all") { m->splitAtDash(label, labels); allLines = 0; }
224 else { allLines = 1; }
227 groups = validParameter.validFile(parameters, "groups", false);
228 if (groups == "not found") { groups = ""; pickedGroups = false; }
231 m->splitAtDash(groups, Groups);
232 m->setGroups(Groups);
235 string temp = validParameter.validFile(parameters, "size", false); if (temp == "not found"){ temp = "0"; }
238 temp = validParameter.validFile(parameters, "persample", false); if (temp == "not found"){ temp = "f"; }
239 persample = m->isTrue(temp);
241 if (groupfile == "") { persample = false; }
243 if ((namefile != "") && (fastafile == "")) { m->mothurOut("You may only use a namefile with a fastafile."); m->mothurOutEndLine(); abort = true; }
245 if ((fastafile == "") && (listfile == "") && (sabundfile == "") && (rabundfile == "") && (sharedfile == "")) {
246 m->mothurOut("You must provide a fasta, list, sabund, rabund or shared file as an input file."); m->mothurOutEndLine(); abort = true; }
248 if (pickedGroups && ((groupfile == "") && (sharedfile == ""))) {
249 m->mothurOut("You cannot pick groups without a valid group file or shared file."); m->mothurOutEndLine(); abort = true; }
251 if ((groupfile != "") && ((fastafile == "") && (listfile == ""))) {
252 m->mothurOut("Group file only valid with listfile or fastafile."); m->mothurOutEndLine(); abort = true; }
254 if ((groupfile != "") && ((fastafile != "") && (listfile != ""))) {
255 m->mothurOut("A new group file can only be made from the subsample of a listfile or fastafile, not both. Please correct."); m->mothurOutEndLine(); abort = true; }
260 catch(exception& e) {
261 m->errorOut(e, "SubSampleCommand", "SubSampleCommand");
265 //**********************************************************************************************************************
267 int SubSampleCommand::execute(){
270 if (abort == true) { if (calledHelp) { return 0; } return 2; }
272 if (sharedfile != "") { getSubSampleShared(); }
273 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); return 0; } }
275 if (listfile != "") { getSubSampleList(); }
276 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); return 0; } }
278 if (rabundfile != "") { getSubSampleRabund(); }
279 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); return 0; } }
281 if (sabundfile != "") { getSubSampleSabund(); }
282 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); return 0; } }
284 if (fastafile != "") { getSubSampleFasta(); }
285 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); return 0; } }
287 //set fasta file as new current fastafile
289 itTypes = outputTypes.find("fasta");
290 if (itTypes != outputTypes.end()) {
291 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setFastaFile(current); }
294 itTypes = outputTypes.find("name");
295 if (itTypes != outputTypes.end()) {
296 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setNameFile(current); }
299 itTypes = outputTypes.find("group");
300 if (itTypes != outputTypes.end()) {
301 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setGroupFile(current); }
304 itTypes = outputTypes.find("list");
305 if (itTypes != outputTypes.end()) {
306 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setListFile(current); }
309 itTypes = outputTypes.find("shared");
310 if (itTypes != outputTypes.end()) {
311 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSharedFile(current); }
314 itTypes = outputTypes.find("rabund");
315 if (itTypes != outputTypes.end()) {
316 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setRabundFile(current); }
319 itTypes = outputTypes.find("sabund");
320 if (itTypes != outputTypes.end()) {
321 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSabundFile(current); }
325 m->mothurOutEndLine();
326 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
327 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
328 m->mothurOutEndLine();
332 catch(exception& e) {
333 m->errorOut(e, "SubSampleCommand", "execute");
337 //**********************************************************************************************************************
338 int SubSampleCommand::getSubSampleFasta() {
341 if (namefile != "") { readNames(); } //fills names with all names in namefile.
342 else { getNames(); }//no name file, so get list of names to pick from
345 if (groupfile != "") {
347 groupMap = new GroupMap(groupfile);
350 //takes care of user setting groupNames that are invalid or setting groups=all
351 SharedUtil* util = new SharedUtil();
352 vector<string> namesGroups = groupMap->getNamesOfGroups();
353 util->setGroups(Groups, namesGroups);
357 if (names.size() != groupMap->getNumSeqs()) {
358 m->mothurOut("[ERROR]: your fasta file contains " + toString(names.size()) + " sequences, and your groupfile contains " + toString(groupMap->getNumSeqs()) + ", please correct.");
359 m->mothurOutEndLine();
365 if (m->control_pressed) { return 0; }
368 //make sure that if your picked groups size is not too big
369 int thisSize = names.size();
371 if (size == 0) { //user has not set size, set size = smallest samples size
372 size = groupMap->getNumSeqs(Groups[0]);
373 for (int i = 1; i < Groups.size(); i++) {
374 int thisSize = groupMap->getNumSeqs(Groups[i]);
376 if (thisSize < size) { size = thisSize; }
378 }else { //make sure size is not too large
379 vector<string> newGroups;
380 for (int i = 0; i < Groups.size(); i++) {
381 int thisSize = groupMap->getNumSeqs(Groups[i]);
383 if (thisSize >= size) { newGroups.push_back(Groups[i]); }
384 else { m->mothurOut("You have selected a size that is larger than " + Groups[i] + " number of sequences, removing " + Groups[i] + "."); m->mothurOutEndLine(); }
389 m->mothurOut("Sampling " + toString(size) + " from each group."); m->mothurOutEndLine();
393 for(int i = 0; i < Groups.size(); i++) {
394 total += groupMap->getNumSeqs(Groups[i]);
397 if (size == 0) { //user has not set size, set size = 10% samples size
398 size = int (total * 0.10);
403 m->mothurOut("Your size is too large for the number of groups you selected. Adjusting to " + toString(int (total * 0.10)) + "."); m->mothurOutEndLine();
405 size = int (total * 0.10);
408 m->mothurOut("Sampling " + toString(size) + " from " + toString(total) + "."); m->mothurOutEndLine();
411 if (size == 0) { //user has not set size, set size = 10% samples size
412 size = int (names.size() * 0.10);
415 if (size > thisSize) { m->mothurOut("Your fasta file only contains " + toString(thisSize) + " sequences. Setting size to " + toString(thisSize) + "."); m->mothurOutEndLine();
419 if (!pickedGroups) { m->mothurOut("Sampling " + toString(size) + " from " + toString(thisSize) + "."); m->mothurOutEndLine(); }
422 random_shuffle(names.begin(), names.end());
424 set<string> subset; //dont want repeat sequence names added
426 for (int i = 0; i < Groups.size(); i++) {
428 //randomly select a subset of those names from this group to include in the subsample
429 for (int j = 0; j < size; j++) {
431 if (m->control_pressed) { return 0; }
433 //get random sequence to add, making sure we have not already added it
437 myrand = int((float)(thisSize) * (float)(rand()) / ((float)RAND_MAX+1.0));
439 if (subset.count(names[myrand]) == 0) {
441 string group = groupMap->getGroup(names[myrand]);
442 if (group == "not found") { m->mothurOut("[ERROR]: " + names[myrand] + " is not in your groupfile. please correct."); m->mothurOutEndLine(); group = "NOTFOUND"; }
444 if (group == Groups[i]) { subset.insert(names[myrand]); break; }
451 //randomly select a subset of those names to include in the subsample
452 for (int j = 0; j < size; j++) {
454 if (m->control_pressed) { return 0; }
456 //get random sequence to add, making sure we have not already added it
460 myrand = int((float)(thisSize) * (float)(rand()) / ((float)RAND_MAX+1.0));
462 if (subset.count(names[myrand]) == 0) {
464 if (groupfile != "") { //if there is a groupfile given fill in group info
465 string group = groupMap->getGroup(names[myrand]);
466 if (group == "not found") { m->mothurOut("[ERROR]: " + names[myrand] + " is not in your groupfile. please correct."); m->mothurOutEndLine(); group = "NOTFOUND"; }
468 if (pickedGroups) { //if hte user picked groups, we only want to keep the names of sequences from those groups
469 if (m->inUsersGroups(group, Groups)) {
470 subset.insert(names[myrand]); break;
473 subset.insert(names[myrand]); break;
475 }else{ //save everyone, group
476 subset.insert(names[myrand]); break;
483 if (subset.size() == 0) { m->mothurOut("The size you selected is too large, skipping fasta file."); m->mothurOutEndLine(); return 0; }
485 string thisOutputDir = outputDir;
486 if (outputDir == "") { thisOutputDir += m->hasPath(fastafile); }
487 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(fastafile)) + "subsample" + m->getExtension(fastafile);
490 m->openOutputFile(outputFileName, out);
491 outputTypes["fasta"].push_back(outputFileName); outputNames.push_back(outputFileName);
493 //read through fasta file outputting only the names on the subsample list
495 m->openInputFile(fastafile, in);
499 map<string, vector<string> >::iterator itNameMap;
503 if (m->control_pressed) { in.close(); out.close(); return 0; }
505 Sequence currSeq(in);
506 thisname = currSeq.getName();
508 if (thisname != "") {
510 //does the subset contain a sequence that this sequence represents
511 itNameMap = nameMap.find(thisname);
512 if (itNameMap != nameMap.end()) {
513 vector<string> nameRepresents = itNameMap->second;
515 for (int i = 0; i < nameRepresents.size(); i++){
516 if (subset.count(nameRepresents[i]) != 0) {
517 out << ">" << nameRepresents[i] << endl << currSeq.getAligned() << endl;
522 m->mothurOut("[ERROR]: " + thisname + " is not in your namefile, please correct."); m->mothurOutEndLine();
530 if (count != subset.size()) {
531 m->mothurOut("[ERROR]: The subset selected contained " + toString(subset.size()) + " sequences, but I only found " + toString(count) + " of those in the fastafile."); m->mothurOutEndLine();
534 //if a groupfile is provided read through the group file only outputting the names on the subsample list
535 if (groupfile != "") {
537 string groupOutputDir = outputDir;
538 if (outputDir == "") { groupOutputDir += m->hasPath(groupfile); }
539 string groupOutputFileName = groupOutputDir + m->getRootName(m->getSimpleName(groupfile)) + "subsample" + m->getExtension(groupfile);
542 m->openOutputFile(groupOutputFileName, outGroup);
543 outputTypes["group"].push_back(groupOutputFileName); outputNames.push_back(groupOutputFileName);
546 m->openInputFile(groupfile, inGroup);
549 while(!inGroup.eof()){
551 if (m->control_pressed) { inGroup.close(); outGroup.close(); return 0; }
553 inGroup >> name; m->gobble(inGroup); //read from first column
554 inGroup >> group; //read from second column
556 //if this name is in the accnos file
557 if (subset.count(name) != 0) {
558 outGroup << name << '\t' << group << endl;
568 if (subset.size() != 0) {
569 m->mothurOut("Your groupfile does not match your fasta file."); m->mothurOutEndLine();
570 for (set<string>::iterator it = subset.begin(); it != subset.end(); it++) {
571 m->mothurOut("[ERROR]: " + *it + " is missing from your groupfile."); m->mothurOutEndLine();
579 catch(exception& e) {
580 m->errorOut(e, "SubSampleCommand", "getSubSampleFasta");
584 //**********************************************************************************************************************
585 int SubSampleCommand::getNames() {
589 m->openInputFile(fastafile, in);
594 if (m->control_pressed) { in.close(); return 0; }
596 Sequence currSeq(in);
597 thisname = currSeq.getName();
599 if (thisname != "") {
600 vector<string> temp; temp.push_back(thisname);
601 nameMap[thisname] = temp;
602 names.push_back(thisname);
611 catch(exception& e) {
612 m->errorOut(e, "SubSampleCommand", "getNames");
616 //**********************************************************************************************************************
617 int SubSampleCommand::readNames() {
621 m->openInputFile(namefile, in);
623 string thisname, repnames;
624 map<string, vector<string> >::iterator it;
628 if (m->control_pressed) { in.close(); return 0; }
630 in >> thisname; m->gobble(in); //read from first column
631 in >> repnames; //read from second column
633 it = nameMap.find(thisname);
634 if (it == nameMap.end()) {
636 vector<string> splitRepNames;
637 m->splitAtComma(repnames, splitRepNames);
639 nameMap[thisname] = splitRepNames;
640 for (int i = 0; i < splitRepNames.size(); i++) { names.push_back(splitRepNames[i]); }
642 }else{ m->mothurOut(thisname + " is already in namesfile. I will use first definition."); m->mothurOutEndLine(); }
651 catch(exception& e) {
652 m->errorOut(e, "SubSampleCommand", "readNames");
656 //**********************************************************************************************************************
657 int SubSampleCommand::getSubSampleShared() {
660 InputData* input = new InputData(sharedfile, "sharedfile");
661 vector<SharedRAbundVector*> lookup = input->getSharedRAbundVectors();
662 string lastLabel = lookup[0]->getLabel();
664 //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
665 set<string> processedLabels;
666 set<string> userLabels = labels;
668 if (size == 0) { //user has not set size, set size = smallest samples size
669 size = lookup[0]->getNumSeqs();
670 for (int i = 1; i < lookup.size(); i++) {
671 int thisSize = lookup[i]->getNumSeqs();
673 if (thisSize < size) { size = thisSize; }
678 vector<SharedRAbundVector*> temp;
679 for (int i = 0; i < lookup.size(); i++) {
680 if (lookup[i]->getNumSeqs() < size) {
681 m->mothurOut(lookup[i]->getGroup() + " contains " + toString(lookup[i]->getNumSeqs()) + ". Eliminating."); m->mothurOutEndLine();
684 Groups.push_back(lookup[i]->getGroup());
685 temp.push_back(lookup[i]);
689 m->setGroups(Groups);
692 if (lookup.size() == 0) { m->mothurOut("The size you selected is too large, skipping shared file."); m->mothurOutEndLine(); delete input; return 0; }
694 m->mothurOut("Sampling " + toString(size) + " from each group."); m->mothurOutEndLine();
696 //as long as you are not at the end of the file or done wih the lines you want
697 while((lookup[0] != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
698 if (m->control_pressed) { delete input; for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; lookup[i] = NULL; } return 0; }
700 if(allLines == 1 || labels.count(lookup[0]->getLabel()) == 1){
702 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
704 processShared(lookup);
706 processedLabels.insert(lookup[0]->getLabel());
707 userLabels.erase(lookup[0]->getLabel());
710 if ((m->anyLabelsToProcess(lookup[0]->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
711 string saveLabel = lookup[0]->getLabel();
713 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
715 lookup = input->getSharedRAbundVectors(lastLabel);
716 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
718 processShared(lookup);
720 processedLabels.insert(lookup[0]->getLabel());
721 userLabels.erase(lookup[0]->getLabel());
723 //restore real lastlabel to save below
724 lookup[0]->setLabel(saveLabel);
727 lastLabel = lookup[0]->getLabel();
728 //prevent memory leak
729 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; lookup[i] = NULL; }
731 //get next line to process
732 lookup = input->getSharedRAbundVectors();
736 if (m->control_pressed) { return 0; }
738 //output error messages about any remaining user labels
739 set<string>::iterator it;
740 bool needToRun = false;
741 for (it = userLabels.begin(); it != userLabels.end(); it++) {
742 m->mothurOut("Your file does not include the label " + *it);
743 if (processedLabels.count(lastLabel) != 1) {
744 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
747 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
751 //run last label if you need to
752 if (needToRun == true) {
753 for (int i = 0; i < lookup.size(); i++) { if (lookup[i] != NULL) { delete lookup[i]; } }
754 lookup = input->getSharedRAbundVectors(lastLabel);
756 m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
758 processShared(lookup);
760 for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
768 catch(exception& e) {
769 m->errorOut(e, "SubSampleCommand", "getSubSampleShared");
773 //**********************************************************************************************************************
774 int SubSampleCommand::processShared(vector<SharedRAbundVector*>& thislookup) {
777 //save mothurOut's binLabels to restore for next label
778 vector<string> saveBinLabels = m->currentBinLabels;
780 string thisOutputDir = outputDir;
781 if (outputDir == "") { thisOutputDir += m->hasPath(sharedfile); }
782 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(sharedfile)) + thislookup[0]->getLabel() + ".subsample" + m->getExtension(sharedfile);
786 m->openOutputFile(outputFileName, out);
787 outputTypes["shared"].push_back(outputFileName); outputNames.push_back(outputFileName);
789 int numBins = thislookup[0]->getNumBins();
790 for (int i = 0; i < thislookup.size(); i++) {
791 int thisSize = thislookup[i]->getNumSeqs();
793 if (thisSize != size) {
795 string thisgroup = thislookup[i]->getGroup();
797 OrderVector* order = new OrderVector();
798 for(int p=0;p<numBins;p++){
799 for(int j=0;j<thislookup[i]->getAbundance(p);j++){
803 random_shuffle(order->begin(), order->end());
805 SharedRAbundVector* temp = new SharedRAbundVector(numBins);
806 temp->setLabel(thislookup[i]->getLabel());
807 temp->setGroup(thislookup[i]->getGroup());
809 delete thislookup[i];
810 thislookup[i] = temp;
813 for (int j = 0; j < size; j++) {
815 if (m->control_pressed) { delete order; out.close(); return 0; }
817 //get random number to sample from order between 0 and thisSize-1.
818 int myrand = int((float)(thisSize) * (float)(rand()) / ((float)RAND_MAX+1.0));
820 int bin = order->get(myrand);
822 int abund = thislookup[i]->getAbundance(bin);
823 thislookup[i]->set(bin, (abund+1), thisgroup);
829 //subsampling may have created some otus with no sequences in them
830 eliminateZeroOTUS(thislookup);
832 if (m->control_pressed) { out.close(); return 0; }
834 thislookup[0]->printHeaders(out);
836 for (int i = 0; i < thislookup.size(); i++) {
837 out << thislookup[i]->getLabel() << '\t' << thislookup[i]->getGroup() << '\t';
838 thislookup[i]->print(out);
843 //save mothurOut's binLabels to restore for next label
844 m->currentBinLabels = saveBinLabels;
849 catch(exception& e) {
850 m->errorOut(e, "SubSampleCommand", "processShared");
854 //**********************************************************************************************************************
855 int SubSampleCommand::getSubSampleList() {
858 string thisOutputDir = outputDir;
859 if (outputDir == "") { thisOutputDir += m->hasPath(listfile); }
860 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(listfile)) + "subsample" + m->getExtension(listfile);
863 m->openOutputFile(outputFileName, out);
864 outputTypes["list"].push_back(outputFileName); outputNames.push_back(outputFileName);
866 InputData* input = new InputData(listfile, "list");
867 ListVector* list = input->getListVector();
868 string lastLabel = list->getLabel();
870 //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
871 set<string> processedLabels;
872 set<string> userLabels = labels;
876 if (groupfile != "") {
878 groupMap = new GroupMap(groupfile);
881 //takes care of user setting groupNames that are invalid or setting groups=all
882 SharedUtil* util = new SharedUtil();
883 vector<string> namesGroups = groupMap->getNamesOfGroups();
884 util->setGroups(Groups, namesGroups);
888 string groupOutputDir = outputDir;
889 if (outputDir == "") { groupOutputDir += m->hasPath(groupfile); }
890 string groupOutputFileName = groupOutputDir + m->getRootName(m->getSimpleName(groupfile)) + "subsample" + m->getExtension(groupfile);
892 m->openOutputFile(groupOutputFileName, outGroup);
893 outputTypes["group"].push_back(groupOutputFileName); outputNames.push_back(groupOutputFileName);
896 if (list->getNumSeqs() != groupMap->getNumSeqs()) {
897 m->mothurOut("[ERROR]: your list file contains " + toString(list->getNumSeqs()) + " sequences, and your groupfile contains " + toString(groupMap->getNumSeqs()) + ", please correct.");
898 m->mothurOutEndLine();
908 //make sure that if your picked groups size is not too big
910 if (size == 0) { //user has not set size, set size = smallest samples size
911 size = groupMap->getNumSeqs(Groups[0]);
912 for (int i = 1; i < Groups.size(); i++) {
913 int thisSize = groupMap->getNumSeqs(Groups[i]);
915 if (thisSize < size) { size = thisSize; }
917 }else { //make sure size is not too large
918 vector<string> newGroups;
919 for (int i = 0; i < Groups.size(); i++) {
920 int thisSize = groupMap->getNumSeqs(Groups[i]);
922 if (thisSize >= size) { newGroups.push_back(Groups[i]); }
923 else { m->mothurOut("You have selected a size that is larger than " + Groups[i] + " number of sequences, removing " + Groups[i] + "."); m->mothurOutEndLine(); }
928 m->mothurOut("Sampling " + toString(size) + " from each group."); m->mothurOutEndLine();
932 for(int i = 0; i < Groups.size(); i++) {
933 total += groupMap->getNumSeqs(Groups[i]);
936 if (size == 0) { //user has not set size, set size = 10% samples size
937 size = int (total * 0.10);
941 m->mothurOut("Your size is too large for the number of groups you selected. Adjusting to " + toString(int (total * 0.10)) + "."); m->mothurOutEndLine();
942 size = int (total * 0.10);
945 m->mothurOut("Sampling " + toString(size) + " from " + toString(total) + "."); m->mothurOutEndLine();
948 if (size == 0) { //user has not set size, set size = 10% samples size
949 size = int (list->getNumSeqs() * 0.10);
952 int thisSize = list->getNumSeqs();
953 if (size > thisSize) { m->mothurOut("Your list file only contains " + toString(thisSize) + " sequences. Setting size to " + toString(thisSize) + "."); m->mothurOutEndLine();
957 m->mothurOut("Sampling " + toString(size) + " from " + toString(list->getNumSeqs()) + "."); m->mothurOutEndLine();
963 for (int i = 0; i < list->getNumBins(); i++) {
964 string binnames = list->get(i);
967 string individual = "";
968 int length = binnames.length();
969 for(int j=0;j<length;j++){
970 if(binnames[j] == ','){
972 if (groupfile != "") { //if there is a groupfile given fill in group info
973 string group = groupMap->getGroup(individual);
974 if (group == "not found") { m->mothurOut("[ERROR]: " + individual + " is not in your groupfile. please correct."); m->mothurOutEndLine(); group = "NOTFOUND"; }
976 if (pickedGroups) { //if hte user picked groups, we only want to keep the names of sequences from those groups
977 if (m->inUsersGroups(group, Groups)) {
978 names.push_back(individual);
981 names.push_back(individual);
983 }else{ //save everyone, group
984 names.push_back(individual);
989 individual += binnames[j];
993 if (groupfile != "") { //if there is a groupfile given fill in group info
994 string group = groupMap->getGroup(individual);
995 if (group == "not found") { m->mothurOut("[ERROR]: " + individual + " is not in your groupfile. please correct."); m->mothurOutEndLine(); group = "NOTFOUND"; }
997 if (pickedGroups) { //if hte user picked groups, we only want to keep the names of sequences from those groups
998 if (m->inUsersGroups(group, Groups)) {
999 names.push_back(individual);
1002 names.push_back(individual);
1004 }else{ //save everyone, group
1005 names.push_back(individual);
1009 random_shuffle(names.begin(), names.end());
1011 //randomly select a subset of those names to include in the subsample
1012 set<string> subset; //dont want repeat sequence names added
1014 for (int i = 0; i < Groups.size(); i++) {
1016 for (int j = 0; j < size; j++) {
1018 if (m->control_pressed) { break; }
1020 //get random sequence to add, making sure we have not already added it
1024 myrand = int((float)(names.size()) * (float)(rand()) / ((float)RAND_MAX+1.0));
1026 if (subset.count(names[myrand]) == 0) { //you are not already added
1027 if (groupMap->getGroup(names[myrand]) == Groups[i]) { subset.insert(names[myrand]); break; }
1033 for (int j = 0; j < size; j++) {
1035 if (m->control_pressed) { break; }
1037 //get random sequence to add, making sure we have not already added it
1041 myrand = int((float)(names.size()) * (float)(rand()) / ((float)RAND_MAX+1.0));
1043 if (subset.count(names[myrand]) == 0) { subset.insert(names[myrand]); break; }
1048 if (groupfile != "") {
1049 //write out new groupfile
1050 for (set<string>::iterator it = subset.begin(); it != subset.end(); it++) {
1051 string group = groupMap->getGroup(*it);
1052 if (group == "not found") { group = "NOTFOUND"; }
1054 outGroup << *it << '\t' << group << endl;
1056 outGroup.close(); delete groupMap;
1060 //as long as you are not at the end of the file or done wih the lines you want
1061 while((list != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
1063 if (m->control_pressed) { delete list; delete input; out.close(); return 0; }
1065 if(allLines == 1 || labels.count(list->getLabel()) == 1){
1067 m->mothurOut(list->getLabel()); m->mothurOutEndLine();
1069 processList(list, out, subset);
1071 processedLabels.insert(list->getLabel());
1072 userLabels.erase(list->getLabel());
1075 if ((m->anyLabelsToProcess(list->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
1076 string saveLabel = list->getLabel();
1080 list = input->getListVector(lastLabel);
1081 m->mothurOut(list->getLabel()); m->mothurOutEndLine();
1083 processList(list, out, subset);
1085 processedLabels.insert(list->getLabel());
1086 userLabels.erase(list->getLabel());
1088 //restore real lastlabel to save below
1089 list->setLabel(saveLabel);
1092 lastLabel = list->getLabel();
1094 delete list; list = NULL;
1096 //get next line to process
1097 list = input->getListVector();
1101 if (m->control_pressed) { if (list != NULL) { delete list; } delete input; out.close(); return 0; }
1103 //output error messages about any remaining user labels
1104 set<string>::iterator it;
1105 bool needToRun = false;
1106 for (it = userLabels.begin(); it != userLabels.end(); it++) {
1107 m->mothurOut("Your file does not include the label " + *it);
1108 if (processedLabels.count(lastLabel) != 1) {
1109 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
1112 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
1116 //run last label if you need to
1117 if (needToRun == true) {
1118 if (list != NULL) { delete list; }
1120 list = input->getListVector(lastLabel);
1122 m->mothurOut(list->getLabel()); m->mothurOutEndLine();
1124 processList(list, out, subset);
1126 delete list; list = NULL;
1130 if (list != NULL) { delete list; }
1136 catch(exception& e) {
1137 m->errorOut(e, "SubSampleCommand", "getSubSampleList");
1141 //**********************************************************************************************************************
1142 int SubSampleCommand::processList(ListVector*& list, ofstream& out, set<string>& subset) {
1145 int numBins = list->getNumBins();
1147 ListVector* temp = new ListVector();
1148 temp->setLabel(list->getLabel());
1150 for (int i = 0; i < numBins; i++) {
1152 if (m->control_pressed) { break; }
1154 string binnames = list->get(i);
1157 string individual = "";
1158 string newNames = "";
1159 int length = binnames.length();
1160 for(int j=0;j<length;j++){
1161 if(binnames[j] == ','){
1162 if (subset.count(individual) != 0) { newNames += individual + ","; }
1165 individual += binnames[j];
1168 if (subset.count(individual) != 0) { newNames += individual + ","; }
1171 //if there are names in this bin add to new list
1172 if (newNames != "") {
1173 newNames = newNames.substr(0, newNames.length()-1); //rip off extra comma
1174 temp->push_back(newNames);
1181 if (m->control_pressed) { return 0; }
1188 catch(exception& e) {
1189 m->errorOut(e, "SubSampleCommand", "processList");
1193 //**********************************************************************************************************************
1194 int SubSampleCommand::getSubSampleRabund() {
1196 InputData* input = new InputData(rabundfile, "rabund");
1197 RAbundVector* rabund = input->getRAbundVector();
1198 string lastLabel = rabund->getLabel();
1200 //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
1201 set<string> processedLabels;
1202 set<string> userLabels = labels;
1204 if (size == 0) { //user has not set size, set size = 10%
1205 size = int((rabund->getNumSeqs()) * 0.10);
1206 }else if (size > rabund->getNumSeqs()) { m->mothurOut("The size you selected is too large, skipping rabund file."); m->mothurOutEndLine(); delete input; delete rabund; return 0; }
1208 m->mothurOut("Sampling " + toString(size) + " from " + toString(rabund->getNumSeqs()) + "."); m->mothurOutEndLine();
1210 string thisOutputDir = outputDir;
1211 if (outputDir == "") { thisOutputDir += m->hasPath(rabundfile); }
1212 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(rabundfile)) + "subsample" + m->getExtension(rabundfile);
1215 m->openOutputFile(outputFileName, out);
1216 outputTypes["rabund"].push_back(outputFileName); outputNames.push_back(outputFileName);
1218 //as long as you are not at the end of the file or done wih the lines you want
1219 while((rabund != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
1220 if (m->control_pressed) { delete input; delete rabund; out.close(); return 0; }
1222 if(allLines == 1 || labels.count(rabund->getLabel()) == 1){
1224 m->mothurOut(rabund->getLabel()); m->mothurOutEndLine();
1226 processRabund(rabund, out);
1228 processedLabels.insert(rabund->getLabel());
1229 userLabels.erase(rabund->getLabel());
1232 if ((m->anyLabelsToProcess(rabund->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
1233 string saveLabel = rabund->getLabel();
1237 rabund = input->getRAbundVector(lastLabel);
1238 m->mothurOut(rabund->getLabel()); m->mothurOutEndLine();
1240 processRabund(rabund, out);
1242 processedLabels.insert(rabund->getLabel());
1243 userLabels.erase(rabund->getLabel());
1245 //restore real lastlabel to save below
1246 rabund->setLabel(saveLabel);
1249 lastLabel = rabund->getLabel();
1251 //prevent memory leak
1252 delete rabund; rabund = NULL;
1254 //get next line to process
1255 rabund = input->getRAbundVector();
1259 if (m->control_pressed) { out.close(); return 0; }
1261 //output error messages about any remaining user labels
1262 set<string>::iterator it;
1263 bool needToRun = false;
1264 for (it = userLabels.begin(); it != userLabels.end(); it++) {
1265 m->mothurOut("Your file does not include the label " + *it);
1266 if (processedLabels.count(lastLabel) != 1) {
1267 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
1270 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
1274 //run last label if you need to
1275 if (needToRun == true) {
1276 if (rabund != NULL) { delete rabund; }
1278 rabund = input->getRAbundVector(lastLabel);
1280 m->mothurOut(rabund->getLabel()); m->mothurOutEndLine();
1282 processRabund(rabund, out);
1293 catch(exception& e) {
1294 m->errorOut(e, "SubSampleCommand", "getSubSampleRabund");
1298 //**********************************************************************************************************************
1299 int SubSampleCommand::processRabund(RAbundVector*& rabund, ofstream& out) {
1302 int numBins = rabund->getNumBins();
1303 int thisSize = rabund->getNumSeqs();
1305 if (thisSize != size) {
1307 OrderVector* order = new OrderVector();
1308 for(int p=0;p<numBins;p++){
1309 for(int j=0;j<rabund->get(p);j++){
1310 order->push_back(p);
1313 random_shuffle(order->begin(), order->end());
1315 RAbundVector* temp = new RAbundVector(numBins);
1316 temp->setLabel(rabund->getLabel());
1321 for (int j = 0; j < size; j++) {
1323 if (m->control_pressed) { delete order; return 0; }
1325 //get random number to sample from order between 0 and thisSize-1.
1326 int myrand = int((float)(thisSize) * (float)(rand()) / ((float)RAND_MAX+1.0));
1328 int bin = order->get(myrand);
1330 int abund = rabund->get(bin);
1331 rabund->set(bin, (abund+1));
1337 if (m->control_pressed) { return 0; }
1344 catch(exception& e) {
1345 m->errorOut(e, "SubSampleCommand", "processRabund");
1349 //**********************************************************************************************************************
1350 int SubSampleCommand::getSubSampleSabund() {
1353 InputData* input = new InputData(sabundfile, "sabund");
1354 SAbundVector* sabund = input->getSAbundVector();
1355 string lastLabel = sabund->getLabel();
1357 //if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
1358 set<string> processedLabels;
1359 set<string> userLabels = labels;
1361 if (size == 0) { //user has not set size, set size = 10%
1362 size = int((sabund->getNumSeqs()) * 0.10);
1363 }else if (size > sabund->getNumSeqs()) { m->mothurOut("The size you selected is too large, skipping sabund file."); m->mothurOutEndLine(); delete input; delete sabund; return 0; }
1366 m->mothurOut("Sampling " + toString(size) + " from " + toString(sabund->getNumSeqs()) + "."); m->mothurOutEndLine();
1368 string thisOutputDir = outputDir;
1369 if (outputDir == "") { thisOutputDir += m->hasPath(sabundfile); }
1370 string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(sabundfile)) + "subsample" + m->getExtension(sabundfile);
1373 m->openOutputFile(outputFileName, out);
1374 outputTypes["sabund"].push_back(outputFileName); outputNames.push_back(outputFileName);
1377 //as long as you are not at the end of the file or done wih the lines you want
1378 while((sabund != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
1379 if (m->control_pressed) { delete input; delete sabund; out.close(); return 0; }
1381 if(allLines == 1 || labels.count(sabund->getLabel()) == 1){
1383 m->mothurOut(sabund->getLabel()); m->mothurOutEndLine();
1385 processSabund(sabund, out);
1387 processedLabels.insert(sabund->getLabel());
1388 userLabels.erase(sabund->getLabel());
1391 if ((m->anyLabelsToProcess(sabund->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
1392 string saveLabel = sabund->getLabel();
1396 sabund = input->getSAbundVector(lastLabel);
1397 m->mothurOut(sabund->getLabel()); m->mothurOutEndLine();
1399 processSabund(sabund, out);
1401 processedLabels.insert(sabund->getLabel());
1402 userLabels.erase(sabund->getLabel());
1404 //restore real lastlabel to save below
1405 sabund->setLabel(saveLabel);
1408 lastLabel = sabund->getLabel();
1410 //prevent memory leak
1411 delete sabund; sabund = NULL;
1413 //get next line to process
1414 sabund = input->getSAbundVector();
1418 if (m->control_pressed) { out.close(); return 0; }
1420 //output error messages about any remaining user labels
1421 set<string>::iterator it;
1422 bool needToRun = false;
1423 for (it = userLabels.begin(); it != userLabels.end(); it++) {
1424 m->mothurOut("Your file does not include the label " + *it);
1425 if (processedLabels.count(lastLabel) != 1) {
1426 m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
1429 m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
1433 //run last label if you need to
1434 if (needToRun == true) {
1435 if (sabund != NULL) { delete sabund; }
1437 sabund = input->getSAbundVector(lastLabel);
1439 m->mothurOut(sabund->getLabel()); m->mothurOutEndLine();
1441 processSabund(sabund, out);
1452 catch(exception& e) {
1453 m->errorOut(e, "SubSampleCommand", "getSubSampleSabund");
1457 //**********************************************************************************************************************
1458 int SubSampleCommand::processSabund(SAbundVector*& sabund, ofstream& out) {
1461 RAbundVector* rabund = new RAbundVector();
1462 *rabund = sabund->getRAbundVector();
1464 int numBins = rabund->getNumBins();
1465 int thisSize = rabund->getNumSeqs();
1467 if (thisSize != size) {
1469 OrderVector* order = new OrderVector();
1470 for(int p=0;p<numBins;p++){
1471 for(int j=0;j<rabund->get(p);j++){
1472 order->push_back(p);
1475 random_shuffle(order->begin(), order->end());
1477 RAbundVector* temp = new RAbundVector(numBins);
1478 temp->setLabel(rabund->getLabel());
1483 for (int j = 0; j < size; j++) {
1485 if (m->control_pressed) { delete order; return 0; }
1487 //get random number to sample from order between 0 and thisSize-1.
1488 int myrand = int((float)(thisSize) * (float)(rand()) / ((float)RAND_MAX+1.0));
1490 int bin = order->get(myrand);
1492 int abund = rabund->get(bin);
1493 rabund->set(bin, (abund+1));
1499 if (m->control_pressed) { return 0; }
1502 sabund = new SAbundVector();
1503 *sabund = rabund->getSAbundVector();
1511 catch(exception& e) {
1512 m->errorOut(e, "SubSampleCommand", "processSabund");
1516 //**********************************************************************************************************************
1517 int SubSampleCommand::eliminateZeroOTUS(vector<SharedRAbundVector*>& thislookup) {
1520 vector<SharedRAbundVector*> newLookup;
1521 for (int i = 0; i < thislookup.size(); i++) {
1522 SharedRAbundVector* temp = new SharedRAbundVector();
1523 temp->setLabel(thislookup[i]->getLabel());
1524 temp->setGroup(thislookup[i]->getGroup());
1525 newLookup.push_back(temp);
1529 vector<string> newBinLabels;
1530 for (int i = 0; i < thislookup[0]->getNumBins(); i++) {
1531 if (m->control_pressed) { for (int j = 0; j < newLookup.size(); j++) { delete newLookup[j]; } return 0; }
1533 //look at each sharedRabund and make sure they are not all zero
1534 bool allZero = true;
1535 for (int j = 0; j < thislookup.size(); j++) {
1536 if (thislookup[j]->getAbundance(i) != 0) { allZero = false; break; }
1539 //if they are not all zero add this bin
1541 for (int j = 0; j < thislookup.size(); j++) {
1542 newLookup[j]->push_back(thislookup[j]->getAbundance(i), thislookup[j]->getGroup());
1544 //if there is a bin label use it otherwise make one
1545 string binLabel = "Otu" + toString(i+1);
1546 if (i < m->currentBinLabels.size()) { binLabel = m->currentBinLabels[i]; }
1548 newBinLabels.push_back(binLabel);
1552 for (int j = 0; j < thislookup.size(); j++) { delete thislookup[j]; }
1555 thislookup = newLookup;
1556 m->currentBinLabels = newBinLabels;
1561 catch(exception& e) {
1562 m->errorOut(e, "SubSampleCommand", "eliminateZeroOTUS");
1567 //**********************************************************************************************************************