helpString += "Example sub.sample(list=abrecovery.fn.list, group=abrecovery.groups, groups=B-C, size=20).\n";
helpString += "The default value for groups is all the groups in your groupfile, and all labels in your inputfile will be used.\n";
helpString += "The sub.sample command outputs a .subsample file.\n";
- helpString += "Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n\n";
+ helpString += "Note: No spaces between parameter labels (i.e. groups), '=' and parameters (i.e.yourGroups).\n";
return helpString;
}
catch(exception& e) {
//allow user to run help
if(option == "help") { help(); abort = true; calledHelp = true; }
+ else if(option == "citation") { citation(); abort = true; calledHelp = true;}
else {
vector<string> myArray = setParameters();
//check for required parameters
listfile = validParameter.validFile(parameters, "list", true);
if (listfile == "not open") { listfile = ""; abort = true; }
- else if (listfile == "not found") { listfile = ""; }
+ else if (listfile == "not found") { listfile = ""; }
+ else { m->setListFile(listfile); }
sabundfile = validParameter.validFile(parameters, "sabund", true);
if (sabundfile == "not open") { sabundfile = ""; abort = true; }
else if (sabundfile == "not found") { sabundfile = ""; }
+ else { m->setSabundFile(sabundfile); }
rabundfile = validParameter.validFile(parameters, "rabund", true);
if (rabundfile == "not open") { rabundfile = ""; abort = true; }
else if (rabundfile == "not found") { rabundfile = ""; }
+ else { m->setRabundFile(rabundfile); }
fastafile = validParameter.validFile(parameters, "fasta", true);
if (fastafile == "not open") { fastafile = ""; abort = true; }
else if (fastafile == "not found") { fastafile = ""; }
+ else { m->setFastaFile(fastafile); }
sharedfile = validParameter.validFile(parameters, "shared", true);
if (sharedfile == "not open") { sharedfile = ""; abort = true; }
else if (sharedfile == "not found") { sharedfile = ""; }
+ else { m->setSharedFile(sharedfile); }
namefile = validParameter.validFile(parameters, "name", true);
if (namefile == "not open") { namefile = ""; abort = true; }
else if (namefile == "not found") { namefile = ""; }
+ else { m->setNameFile(namefile); }
groupfile = validParameter.validFile(parameters, "group", true);
if (groupfile == "not open") { groupfile = ""; abort = true; }
else if (groupfile == "not found") { groupfile = ""; }
-
+ else { m->setGroupFile(groupfile); }
//check for optional parameter and set defaults
// ...at some point should added some additional type checking...
if (abort == true) { if (calledHelp) { return 0; } return 2; }
if (sharedfile != "") { getSubSampleShared(); }
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); return 0; } }
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); return 0; } }
if (listfile != "") { getSubSampleList(); }
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); return 0; } }
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); return 0; } }
if (rabundfile != "") { getSubSampleRabund(); }
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); return 0; } }
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); return 0; } }
if (sabundfile != "") { getSubSampleSabund(); }
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); return 0; } }
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); return 0; } }
if (fastafile != "") { getSubSampleFasta(); }
- if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); return 0; } }
+ if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); return 0; } }
//set fasta file as new current fastafile
string current = "";
if (m->control_pressed) { return 0; }
- string thisOutputDir = outputDir;
- if (outputDir == "") { thisOutputDir += m->hasPath(fastafile); }
- string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(fastafile)) + "subsample" + m->getExtension(fastafile);
-
- ofstream out;
- m->openOutputFile(outputFileName, out);
- outputTypes["fasta"].push_back(outputFileName); outputNames.push_back(outputFileName);
//make sure that if your picked groups size is not too big
int thisSize = names.size();
if (thisSize < size) { size = thisSize; }
}
}else { //make sure size is not too large
- int smallestSize = groupMap->getNumSeqs(Groups[0]);
- for (int i = 1; i < Groups.size(); i++) {
+ vector<string> newGroups;
+ for (int i = 0; i < Groups.size(); i++) {
int thisSize = groupMap->getNumSeqs(Groups[i]);
- if (thisSize < smallestSize) { smallestSize = thisSize; }
+ if (thisSize >= size) { newGroups.push_back(Groups[i]); }
+ else { m->mothurOut("You have selected a size that is larger than " + Groups[i] + " number of sequences, removing " + Groups[i] + "."); m->mothurOutEndLine(); }
}
- if (smallestSize < size) { size = smallestSize; m->mothurOut("You have selected a size that is larger than your smallest sample, using your samllest sample size, " + toString(smallestSize) + "."); m->mothurOutEndLine(); }
+ Groups = newGroups;
}
m->mothurOut("Sampling " + toString(size) + " from each group."); m->mothurOutEndLine();
}
}
}
+
+ if (subset.size() == 0) { m->mothurOut("The size you selected is too large, skipping fasta file."); m->mothurOutEndLine(); return 0; }
+
+ string thisOutputDir = outputDir;
+ if (outputDir == "") { thisOutputDir += m->hasPath(fastafile); }
+ string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(fastafile)) + "subsample" + m->getExtension(fastafile);
+
+ ofstream out;
+ m->openOutputFile(outputFileName, out);
+ outputTypes["fasta"].push_back(outputFileName); outputNames.push_back(outputFileName);
+
//read through fasta file outputting only the names on the subsample list
ifstream in;
m->openInputFile(fastafile, in);
int SubSampleCommand::getSubSampleShared() {
try {
- string thisOutputDir = outputDir;
- if (outputDir == "") { thisOutputDir += m->hasPath(sharedfile); }
- string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(sharedfile)) + "subsample" + m->getExtension(sharedfile);
-
- ofstream out;
- m->openOutputFile(outputFileName, out);
- outputTypes["shared"].push_back(outputFileName); outputNames.push_back(outputFileName);
-
InputData* input = new InputData(sharedfile, "sharedfile");
vector<SharedRAbundVector*> lookup = input->getSharedRAbundVectors();
string lastLabel = lookup[0]->getLabel();
if (thisSize < size) { size = thisSize; }
}
+ }else {
+ m->Groups.clear();
+ vector<SharedRAbundVector*> temp;
+ for (int i = 0; i < lookup.size(); i++) {
+ if (lookup[i]->getNumSeqs() < size) {
+ m->mothurOut(lookup[i]->getGroup() + " contains " + toString(lookup[i]->getNumSeqs()) + ". Eliminating."); m->mothurOutEndLine();
+ delete lookup[i];
+ }else {
+ m->Groups.push_back(lookup[i]->getGroup());
+ temp.push_back(lookup[i]);
+ }
+ }
+ lookup = temp;
+ Groups = m->Groups;
}
- m->mothurOut("Sampling " + toString(size) + " from " + toString(lookup[0]->getNumSeqs()) + "."); m->mothurOutEndLine();
+ if (lookup.size() == 0) { m->mothurOut("The size you selected is too large, skipping shared file."); m->mothurOutEndLine(); delete input; return 0; }
+
+ m->mothurOut("Sampling " + toString(size) + " from each group."); m->mothurOutEndLine();
//as long as you are not at the end of the file or done wih the lines you want
while((lookup[0] != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
- if (m->control_pressed) { delete input; for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; lookup[i] = NULL; } out.close(); return 0; }
+ if (m->control_pressed) { delete input; for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; lookup[i] = NULL; } return 0; }
if(allLines == 1 || labels.count(lookup[0]->getLabel()) == 1){
m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
- processShared(lookup, out);
+ processShared(lookup);
processedLabels.insert(lookup[0]->getLabel());
userLabels.erase(lookup[0]->getLabel());
lookup = input->getSharedRAbundVectors(lastLabel);
m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
- processShared(lookup, out);
+ processShared(lookup);
processedLabels.insert(lookup[0]->getLabel());
userLabels.erase(lookup[0]->getLabel());
}
- if (m->control_pressed) { out.close(); return 0; }
+ if (m->control_pressed) { return 0; }
//output error messages about any remaining user labels
set<string>::iterator it;
m->mothurOut(lookup[0]->getLabel()); m->mothurOutEndLine();
- processShared(lookup, out);
+ processShared(lookup);
for (int i = 0; i < lookup.size(); i++) { delete lookup[i]; }
}
delete input;
- out.close();
return 0;
}
}
//**********************************************************************************************************************
-int SubSampleCommand::processShared(vector<SharedRAbundVector*>& thislookup, ofstream& out) {
+int SubSampleCommand::processShared(vector<SharedRAbundVector*>& thislookup) {
try {
+ //save mothurOut's binLabels to restore for next label
+ vector<string> saveBinLabels = m->currentBinLabels;
+
+ string thisOutputDir = outputDir;
+ if (outputDir == "") { thisOutputDir += m->hasPath(sharedfile); }
+ string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(sharedfile)) + thislookup[0]->getLabel() + ".subsample" + m->getExtension(sharedfile);
+
+
+ ofstream out;
+ m->openOutputFile(outputFileName, out);
+ outputTypes["shared"].push_back(outputFileName); outputNames.push_back(outputFileName);
+
int numBins = thislookup[0]->getNumBins();
for (int i = 0; i < thislookup.size(); i++) {
int thisSize = thislookup[i]->getNumSeqs();
for (int j = 0; j < size; j++) {
- if (m->control_pressed) { delete order; return 0; }
+ if (m->control_pressed) { delete order; out.close(); return 0; }
//get random number to sample from order between 0 and thisSize-1.
int myrand = int((float)(thisSize) * (float)(rand()) / ((float)RAND_MAX+1.0));
//subsampling may have created some otus with no sequences in them
eliminateZeroOTUS(thislookup);
- if (m->control_pressed) { return 0; }
+ if (m->control_pressed) { out.close(); return 0; }
+
+ thislookup[0]->printHeaders(out);
for (int i = 0; i < thislookup.size(); i++) {
out << thislookup[i]->getLabel() << '\t' << thislookup[i]->getGroup() << '\t';
thislookup[i]->print(out);
}
+ out.close();
+
+ //save mothurOut's binLabels to restore for next label
+ m->currentBinLabels = saveBinLabels;
+
return 0;
}
if (thisSize < size) { size = thisSize; }
}
}else { //make sure size is not too large
- int smallestSize = groupMap->getNumSeqs(Groups[0]);
- for (int i = 1; i < Groups.size(); i++) {
+ vector<string> newGroups;
+ for (int i = 0; i < Groups.size(); i++) {
int thisSize = groupMap->getNumSeqs(Groups[i]);
- if (thisSize < smallestSize) { smallestSize = thisSize; }
+ if (thisSize >= size) { newGroups.push_back(Groups[i]); }
+ else { m->mothurOut("You have selected a size that is larger than " + Groups[i] + " number of sequences, removing " + Groups[i] + "."); m->mothurOutEndLine(); }
}
- if (smallestSize < size) { size = smallestSize; m->mothurOut("You have selected a size that is larger than your smallest sample, using your samllest sample size, " + toString(smallestSize) + "."); m->mothurOutEndLine(); }
+ Groups = newGroups;
}
m->mothurOut("Sampling " + toString(size) + " from each group."); m->mothurOutEndLine();
individual += binnames[j];
}
}
- if (subset.count(individual) != 0) { newNames += individual; }
+ if (subset.count(individual) != 0) { newNames += individual + ","; }
//if there are names in this bin add to new list
//**********************************************************************************************************************
int SubSampleCommand::getSubSampleRabund() {
try {
-
- string thisOutputDir = outputDir;
- if (outputDir == "") { thisOutputDir += m->hasPath(rabundfile); }
- string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(rabundfile)) + "subsample" + m->getExtension(rabundfile);
-
- ofstream out;
- m->openOutputFile(outputFileName, out);
- outputTypes["rabund"].push_back(outputFileName); outputNames.push_back(outputFileName);
-
InputData* input = new InputData(rabundfile, "rabund");
RAbundVector* rabund = input->getRAbundVector();
string lastLabel = rabund->getLabel();
if (size == 0) { //user has not set size, set size = 10%
size = int((rabund->getNumSeqs()) * 0.10);
- }
+ }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; }
m->mothurOut("Sampling " + toString(size) + " from " + toString(rabund->getNumSeqs()) + "."); m->mothurOutEndLine();
+ string thisOutputDir = outputDir;
+ if (outputDir == "") { thisOutputDir += m->hasPath(rabundfile); }
+ string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(rabundfile)) + "subsample" + m->getExtension(rabundfile);
+
+ ofstream out;
+ m->openOutputFile(outputFileName, out);
+ outputTypes["rabund"].push_back(outputFileName); outputNames.push_back(outputFileName);
+
//as long as you are not at the end of the file or done wih the lines you want
while((rabund != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
if (m->control_pressed) { delete input; delete rabund; out.close(); return 0; }
//**********************************************************************************************************************
int SubSampleCommand::getSubSampleSabund() {
try {
-
- string thisOutputDir = outputDir;
- if (outputDir == "") { thisOutputDir += m->hasPath(sabundfile); }
- string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(sabundfile)) + "subsample" + m->getExtension(sabundfile);
-
- ofstream out;
- m->openOutputFile(outputFileName, out);
- outputTypes["sabund"].push_back(outputFileName); outputNames.push_back(outputFileName);
-
+
InputData* input = new InputData(sabundfile, "sabund");
SAbundVector* sabund = input->getSAbundVector();
string lastLabel = sabund->getLabel();
if (size == 0) { //user has not set size, set size = 10%
size = int((sabund->getNumSeqs()) * 0.10);
- }
+ }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; }
+
m->mothurOut("Sampling " + toString(size) + " from " + toString(sabund->getNumSeqs()) + "."); m->mothurOutEndLine();
+ string thisOutputDir = outputDir;
+ if (outputDir == "") { thisOutputDir += m->hasPath(sabundfile); }
+ string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(sabundfile)) + "subsample" + m->getExtension(sabundfile);
+
+ ofstream out;
+ m->openOutputFile(outputFileName, out);
+ outputTypes["sabund"].push_back(outputFileName); outputNames.push_back(outputFileName);
+
+
//as long as you are not at the end of the file or done wih the lines you want
while((sabund != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
if (m->control_pressed) { delete input; delete sabund; out.close(); return 0; }
}
//for each bin
+ vector<string> newBinLabels;
for (int i = 0; i < thislookup[0]->getNumBins(); i++) {
if (m->control_pressed) { for (int j = 0; j < newLookup.size(); j++) { delete newLookup[j]; } return 0; }
for (int j = 0; j < thislookup.size(); j++) {
newLookup[j]->push_back(thislookup[j]->getAbundance(i), thislookup[j]->getGroup());
}
+ //if there is a bin label use it otherwise make one
+ string binLabel = "Otu" + (i+1);
+ if (i < m->currentBinLabels.size()) { binLabel = m->currentBinLabels[i]; }
+
+ newBinLabels.push_back(binLabel);
}
}
thislookup.clear();
thislookup = newLookup;
+ m->currentBinLabels = newBinLabels;
return 0;