SequenceCountParser::SequenceCountParser(string countfile, string fastafile) {
try {
m = MothurOut::getInstance();
//read count file
CountTable countTable;
countTable.readTable(countfile, true, false);
//initialize maps
namesOfGroups = countTable.getNamesOfGroups();
for (int i = 0; i < namesOfGroups.size(); i++) {
vector<Sequence> temp;
map<string, int> tempMap;
seqs[namesOfGroups[i]] = temp;
countTablePerGroup[namesOfGroups[i]] = tempMap;
}
//read fasta file making sure each sequence is in the group file
ifstream in;
m->openInputFile(fastafile, in);
int fastaCount = 0;
while (!in.eof()) {
if (m->control_pressed) { break; }
Sequence seq(in); m->gobble(in);
fastaCount++;
if (m->debug) { if((fastaCount) % 1000 == 0){ m->mothurOut("[DEBUG]: reading seq " + toString(fastaCount) + "\n."); } }
if (seq.getName() != "") {
allSeqsMap[seq.getName()] = seq.getName();
vector<int> groupCounts = countTable.getGroupCounts(seq.getName());
for (int i = 0; i < namesOfGroups.size(); i++) {
if (groupCounts[i] != 0) {
seqs[namesOfGroups[i]].push_back(seq);
countTablePerGroup[namesOfGroups[i]][seq.getName()] = groupCounts[i];
}
}
}
}
in.close();
}
catch(exception& e) {
m->errorOut(e, "SequenceCountParser", "SequenceCountParser");
exit(1);
}
}
SequenceCountParser::SequenceCountParser(string fastafile, CountTable& countTable) {
try {
m = MothurOut::getInstance();
//initialize maps
if (countTable.hasGroupInfo()) {
namesOfGroups = countTable.getNamesOfGroups();
for (int i = 0; i < namesOfGroups.size(); i++) {
vector<Sequence> temp;
map<string, int> tempMap;
seqs[namesOfGroups[i]] = temp;
countTablePerGroup[namesOfGroups[i]] = tempMap;
}
//read fasta file making sure each sequence is in the group file
ifstream in;
m->openInputFile(fastafile, in);
int fastaCount = 0;
while (!in.eof()) {
if (m->control_pressed) { break; }
Sequence seq(in); m->gobble(in);
fastaCount++;
if (m->debug) { if((fastaCount) % 1000 == 0){ m->mothurOut("[DEBUG]: reading seq " + toString(fastaCount) + "\n."); } }
if (seq.getName() != "") {
allSeqsMap[seq.getName()] = seq.getName();
vector<int> groupCounts = countTable.getGroupCounts(seq.getName());
for (int i = 0; i < namesOfGroups.size(); i++) {
if (groupCounts[i] != 0) {
seqs[namesOfGroups[i]].push_back(seq);
countTablePerGroup[namesOfGroups[i]][seq.getName()] = groupCounts[i];
}
}
}
}
in.close();
}else { m->control_pressed = true; m->mothurOut("[ERROR]: cannot parse fasta file by group with a count table that does not include group data, please correct.\n"); }
}
catch(exception& e) {
m->errorOut(e, "SequenceCountParser", "SequenceCountParser");
exit(1);
}
}
//**********************************************************************************************************************
int SplitGroupCommand::runCount(){
try {
CountTable ct;
ct.readTable(countfile, true, false);
if (!ct.hasGroupInfo()) { m->mothurOut("[ERROR]: your count file does not contain group info, cannot split by group.\n"); m->control_pressed = true; }
if (m->control_pressed) { return 0; }
vector<string> namesGroups = ct.getNamesOfGroups();
SharedUtil util; util.setGroups(Groups, namesGroups);
//fill filehandles with neccessary ofstreams
map<string, string> ffiles; //group -> filename
map<string, string> cfiles; //group -> filename
for (int i=0; i<Groups.size(); i++) {
ofstream ftemp, ctemp;
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastafile));
variables["[group]"] = Groups[i];
string newFasta = getOutputFileName("fasta",variables);
outputNames.push_back(newFasta); outputTypes["fasta"].push_back(newFasta);
ffiles[Groups[i]] = newFasta;
m->openOutputFile(newFasta, ftemp); ftemp.close();
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(countfile));
string newCount = getOutputFileName("count",variables);
outputNames.push_back(newCount); outputTypes["count"].push_back(newCount);
cfiles[Groups[i]] = newCount;
m->openOutputFile(newCount, ctemp);
ctemp << "Representative_Sequence\ttotal\t" << Groups[i] << endl; ctemp.close();
}
ifstream in;
m->openInputFile(fastafile, in);
while (!in.eof()) {
Sequence seq(in); m->gobble(in);
if (m->control_pressed) { break; }
if (seq.getName() != "") {
vector<string> thisSeqsGroups = ct.getGroups(seq.getName());
for (int i = 0; i < thisSeqsGroups.size(); i++) {
if (m->inUsersGroups(thisSeqsGroups[i], Groups)) { //if this sequence belongs to a group we want them print
ofstream outf, outc;
m->openOutputFileAppend(ffiles[thisSeqsGroups[i]], outf);
seq.printSequence(outf); outf.close();
int numSeqs = ct.getGroupCount(seq.getName(), thisSeqsGroups[i]);
m->openOutputFileAppend(cfiles[thisSeqsGroups[i]], outc);
outc << seq.getName() << '\t' << numSeqs << '\t' << numSeqs << endl; outc.close();
}
}
}
}
in.close();
return 0;
}
catch(exception& e) {
m->errorOut(e, "SplitGroupCommand", "runCount");
exit(1);
}
}
int MergeGroupsCommand::processCountFile(DesignMap*& designMap){
try {
CountTable countTable;
if (!countTable.testGroups(countfile)) { m->mothurOut("[ERROR]: your countfile contains no group information, please correct.\n"); m->setControl_pressed(true); return 0; }
//read countTable
countTable.readTable(countfile, true, false);
//fill Groups - checks for "all" and for any typo groups
vector<string> nameGroups = countTable.getNamesOfGroups();
if (Groups.size() == 0) { Groups = nameGroups; }
vector<string> dnamesGroups = designMap->getNamesGroups();
//sanity check
bool error = false;
if (nameGroups.size() == dnamesGroups.size()) { //at least there are the same number
//is every group in counttable also in designmap
for (int i = 0; i < nameGroups.size(); i++) {
if (m->getControl_pressed()) { break; }
if (!util.inUsersGroups(nameGroups[i], dnamesGroups)) { error = true; break; }
}
}
if (error) { m->mothurOut("[ERROR]: Your countfile does not contain the same groups as your design file, please correct\n"); m->setControl_pressed(true); return 0; }
//user selected groups - remove some groups from table
if (Groups.size() != nameGroups.size()) {
for (int i = 0; i < nameGroups.size(); i++) {
if (!util.inUsersGroups(nameGroups[i], Groups)) { countTable.removeGroup(nameGroups[i]); }
}
}
//ask again in case order changed
nameGroups = countTable.getNamesOfGroups();
int numGroups = nameGroups.size();
//create new table
CountTable newTable;
vector<string> treatments = designMap->getCategory();
map<string, vector<int> > clearedMap;
for (int i = 0; i < treatments.size(); i++) {
newTable.addGroup(treatments[i]);
vector<int> temp;
clearedMap[treatments[i]] = temp;
}
treatments = newTable.getNamesOfGroups();
set<string> namesToRemove;
vector<string> namesOfSeqs = countTable.getNamesOfSeqs();
for (int i = 0; i < namesOfSeqs.size(); i++) {
if (m->getControl_pressed()) { break; }
vector<int> thisSeqsCounts = countTable.getGroupCounts(namesOfSeqs[i]);
map<string, vector<int> > thisSeqsMap = clearedMap;
for (int j = 0; j < numGroups; j++) {
thisSeqsMap[designMap->get(nameGroups[j])].push_back(thisSeqsCounts[j]);
}
//create new counts for seq for new table
vector<int> newCounts; int totalAbund = 0;
for (int j = 0; j < treatments.size(); j++){
int abund = mergeAbund(thisSeqsMap[treatments[j]]);
newCounts.push_back(abund); //order matters, add in count for each treatment in new table.
totalAbund += abund;
}
//add seq to new table
if(totalAbund == 0) {
namesToRemove.insert(namesOfSeqs[i]);
}else { newTable.push_back(namesOfSeqs[i], newCounts); }
}
if (error) { m->setControl_pressed(true); return 0; }
//remove sequences zeroed out by median method
if (namesToRemove.size() != 0) {
//print names
ofstream out;
string accnosFile = "accnosFile.temp";
util.openOutputFile(accnosFile, out);
//output to .accnos file
for (set<string>::iterator it = namesToRemove.begin(); it != namesToRemove.end(); it++) {
if (m->getControl_pressed()) { out.close(); util.mothurRemove(accnosFile); return 0; }
out << *it << endl;
}
out.close();
//run remove.seqs
string inputString = "accnos=" + accnosFile + ", fasta=" + fastafile;
m->mothurOut("/******************************************/"); m->mothurOutEndLine();
m->mothurOut("Running command: remove.seqs(" + inputString + ")"); m->mothurOutEndLine();
current->setMothurCalling(true);
Command* removeCommand = new RemoveSeqsCommand(inputString);
removeCommand->execute();
map<string, vector<string> > filenames = removeCommand->getOutputFiles();
delete removeCommand;
current->setMothurCalling(false);
m->mothurOut("/******************************************/"); m->mothurOutEndLine();
util.mothurRemove(accnosFile);
}
string thisOutputDir = outputDir;
if (outputDir == "") { thisOutputDir += util.hasPath(countfile); }
map<string, string> variables;
variables["[filename]"] = thisOutputDir + util.getRootName(util.getSimpleName(countfile));
variables["[extension]"] = util.getExtension(countfile);
string outputFileName = getOutputFileName("count", variables);
outputTypes["count"].push_back(outputFileName); outputNames.push_back(outputFileName);
newTable.printTable(outputFileName);
return 0;
}
catch(exception& e) {
m->errorOut(e, "MergeGroupsCommand", "processCountFile");
exit(1);
}
}
int RemoveGroupsCommand::execute(){
try {
if (abort == true) { if (calledHelp) { return 0; } return 2; }
//get groups you want to remove
if (accnosfile != "") { m->readAccnos(accnosfile, Groups); m->setGroups(Groups); }
if (groupfile != "") {
groupMap = new GroupMap(groupfile);
groupMap->readMap();
//make sure groups are valid
//takes care of user setting groupNames that are invalid or setting groups=all
vector<string> namesGroups = groupMap->getNamesOfGroups();
vector<string> checkedGroups;
for (int i = 0; i < Groups.size(); i++) {
if (m->inUsersGroups(Groups[i], namesGroups)) { checkedGroups.push_back(Groups[i]); }
else { m->mothurOut("[WARNING]: " + Groups[i] + " is not a valid group in your groupfile, ignoring.\n"); }
}
if (checkedGroups.size() == 0) { m->mothurOut("[ERROR]: no valid groups, aborting.\n"); delete groupMap; return 0; }
else {
Groups = checkedGroups;
m->setGroups(Groups);
}
//fill names with names of sequences that are from the groups we want to remove
fillNames();
delete groupMap;
}else if (countfile != ""){
if ((fastafile != "") || (listfile != "") || (taxfile != "")) {
m->mothurOut("\n[NOTE]: The count file should contain only unique names, so mothur assumes your fasta, list and taxonomy files also contain only uniques.\n\n");
}
CountTable ct;
ct.readTable(countfile, true, false);
if (!ct.hasGroupInfo()) { m->mothurOut("[ERROR]: your count file does not contain group info, aborting.\n"); return 0; }
vector<string> gNamesOfGroups = ct.getNamesOfGroups();
SharedUtil util;
util.setGroups(Groups, gNamesOfGroups);
vector<string> namesOfSeqs = ct.getNamesOfSeqs();
sort(Groups.begin(), Groups.end());
for (int i = 0; i < namesOfSeqs.size(); i++) {
vector<string> thisSeqsGroups = ct.getGroups(namesOfSeqs[i]);
if (m->isSubset(Groups, thisSeqsGroups)) { //you only have seqs from these groups so remove you
names.insert(namesOfSeqs[i]);
}
}
}
if (m->control_pressed) { return 0; }
//read through the correct file and output lines you want to keep
if (namefile != "") { readName(); }
if (fastafile != "") { readFasta(); }
if (groupfile != "") { readGroup(); }
if (countfile != "") { readCount(); }
if (listfile != "") { readList(); }
if (taxfile != "") { readTax(); }
if (sharedfile != "") { readShared(); }
if (designfile != "") { readDesign(); }
if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; }
if (outputNames.size() != 0) {
m->mothurOutEndLine();
m->mothurOut("Output File names: "); m->mothurOutEndLine();
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
m->mothurOutEndLine();
//set fasta file as new current fastafile
string current = "";
itTypes = outputTypes.find("fasta");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setFastaFile(current); }
}
itTypes = outputTypes.find("name");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setNameFile(current); }
}
itTypes = outputTypes.find("group");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setGroupFile(current); }
}
itTypes = outputTypes.find("list");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setListFile(current); }
}
itTypes = outputTypes.find("taxonomy");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setTaxonomyFile(current); }
}
itTypes = outputTypes.find("shared");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSharedFile(current); }
}
itTypes = outputTypes.find("design");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setDesignFile(current); }
}
itTypes = outputTypes.find("count");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setCountTableFile(current); }
}
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "RemoveGroupsCommand", "execute");
exit(1);
}
}
EstOutput Parsimony::getValues(Tree* t, int p, string o) {
try {
processors = p;
outputDir = o;
CountTable* ct = t->getCountTable();
//if the users enters no groups then give them the score of all groups
vector<string> mGroups = m->getGroups();
int numGroups = mGroups.size();
//calculate number of comparsions
int numComp = 0;
vector< vector<string> > namesOfGroupCombos;
for (int r=0; r<numGroups; r++) {
for (int l = 0; l < r; l++) {
numComp++;
vector<string> groups; groups.push_back(mGroups[r]); groups.push_back(mGroups[l]);
//cout << globaldata->Groups[r] << '\t' << globaldata->Groups[l] << endl;
namesOfGroupCombos.push_back(groups);
}
}
//numComp+1 for AB, AC, BC, ABC
if (numComp != 1) {
vector<string> groups;
if (numGroups == 0) {
//get score for all users groups
vector<string> tGroups = ct->getNamesOfGroups();
for (int i = 0; i < tGroups.size(); i++) {
if (tGroups[i] != "xxx") {
groups.push_back(tGroups[i]);
//cout << tmap->namesOfGroups[i] << endl;
}
}
namesOfGroupCombos.push_back(groups);
}else {
for (int i = 0; i < mGroups.size(); i++) {
groups.push_back(mGroups[i]);
//cout << globaldata->Groups[i] << endl;
}
namesOfGroupCombos.push_back(groups);
}
}
lines.clear();
int remainingPairs = namesOfGroupCombos.size();
int startIndex = 0;
for (int remainingProcessors = processors; remainingProcessors > 0; remainingProcessors--) {
int numPairs = remainingPairs; //case for last processor
if (remainingProcessors != 1) { numPairs = ceil(remainingPairs / remainingProcessors); }
lines.push_back(linePair(startIndex, numPairs)); //startIndex, numPairs
startIndex = startIndex + numPairs;
remainingPairs = remainingPairs - numPairs;
}
data = createProcesses(t, namesOfGroupCombos, ct);
return data;
}
catch(exception& e) {
m->errorOut(e, "Parsimony", "getValues");
exit(1);
}
}
//**********************************************************************************************************************
int SharedCommand::createSharedFromListGroup() {
try {
GroupMap* groupMap = NULL;
CountTable* countTable = NULL;
pickedGroups = false;
if (groupfile != "") {
groupMap = new GroupMap(groupfile);
int groupError = groupMap->readMap();
if (groupError == 1) { delete groupMap; return 0; }
vector<string> allGroups = groupMap->getNamesOfGroups();
if (Groups.size() == 0) { Groups = allGroups; }
else { pickedGroups = true; }
}else{
countTable = new CountTable();
countTable->readTable(countfile, true, false);
vector<string> allGroups = countTable->getNamesOfGroups();
if (Groups.size() == 0) { Groups = allGroups; }
else { pickedGroups = true; }
}
int numGroups = Groups.size();
if (m->getControl_pressed()) { return 0; }
ofstream out;
string filename = "";
if (!pickedGroups) {
string filename = listfile;
if (outputDir == "") { outputDir += util.hasPath(filename); }
map<string, string> variables;
variables["[filename]"] = outputDir + util.getRootName(util.getSimpleName(filename));
filename = getOutputFileName("shared",variables);
outputNames.push_back(filename); outputTypes["shared"].push_back(filename);
util.openOutputFile(filename, out);
}
//set fileroot
fileroot = outputDir + util.getRootName(util.getSimpleName(listfile));
map<string, string> variables;
variables["[filename]"] = fileroot;
string errorOff = "no error";
InputData input(listfile, "shared", Groups);
SharedListVector* SharedList = input.getSharedListVector();
string lastLabel = SharedList->getLabel();
SharedRAbundVectors* lookup;
if (m->getControl_pressed()) {
delete SharedList; if (groupMap != NULL) { delete groupMap; } if (countTable != NULL) { delete countTable; }
out.close(); if (!pickedGroups) { util.mothurRemove(filename); }
return 0;
}
//sanity check
vector<string> namesSeqs;
int numGroupNames = 0;
if (current->getGroupMode() == "group") { namesSeqs = groupMap->getNamesSeqs(); numGroupNames = groupMap->getNumSeqs(); }
else { namesSeqs = countTable->getNamesOfSeqs(); numGroupNames = countTable->getNumUniqueSeqs(); }
int error = ListGroupSameSeqs(namesSeqs, SharedList);
if ((!pickedGroups) && (SharedList->getNumSeqs() != numGroupNames)) { //if the user has not specified any groups and their files don't match exit with error
m->mothurOut("Your group file contains " + toString(numGroupNames) + " sequences and list file contains " + toString(SharedList->getNumSeqs()) + " sequences. Please correct.\n"); m->setControl_pressed(true);
out.close(); if (!pickedGroups) { util.mothurRemove(filename); } //remove blank shared file you made
//delete memory
delete SharedList; if (groupMap != NULL) { delete groupMap; } if (countTable != NULL) { delete countTable; }
return 0;
}
if (error == 1) { m->setControl_pressed(true); }
//if user has specified groups make new groupfile for them
if ((pickedGroups) && (current->getGroupMode() == "group")) { //make new group file
string groups = "";
if (numGroups < 4) {
for (int i = 0; i < numGroups-1; i++) {
groups += Groups[i] + ".";
}
groups+=Groups[numGroups-1];
}else { groups = "merge"; }
map<string, string> variables;
variables["[filename]"] = outputDir + util.getRootName(util.getSimpleName(listfile));
variables["[group]"] = groups;
string newGroupFile = getOutputFileName("group",variables);
outputTypes["group"].push_back(newGroupFile);
outputNames.push_back(newGroupFile);
ofstream outGroups;
util.openOutputFile(newGroupFile, outGroups);
vector<string> names = groupMap->getNamesSeqs();
string groupName;
for (int i = 0; i < names.size(); i++) {
groupName = groupMap->getGroup(names[i]);
if (isValidGroup(groupName, Groups)) {
outGroups << names[i] << '\t' << groupName << endl;
}
}
outGroups.close();
}
//if the users enters label "0.06" and there is no "0.06" in their file use the next lowest label.
set<string> processedLabels;
set<string> userLabels = labels;
bool printHeaders = true;
while((SharedList != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
if (m->getControl_pressed()) {
delete SharedList; if (groupMap != NULL) { delete groupMap; } if (countTable != NULL) { delete countTable; }
if (!pickedGroups) { out.close(); util.mothurRemove(filename); }
return 0;
}
if(allLines == 1 || labels.count(SharedList->getLabel()) == 1){
lookup = SharedList->getSharedRAbundVector();
m->mothurOut(lookup->getLabel()+"\n");
if (m->getControl_pressed()) {
delete SharedList; if (groupMap != NULL) { delete groupMap; } if (countTable != NULL) { delete countTable; }
delete lookup;
if (!pickedGroups) { out.close(); util.mothurRemove(filename); }
return 0;
}
//if picked groups must split the shared file by label
if (pickedGroups) {
string filename = listfile;
if (outputDir == "") { outputDir += util.hasPath(filename); }
map<string, string> variables;
variables["[filename]"] = outputDir + util.getRootName(util.getSimpleName(filename));
variables["[distance]"] = lookup->getLabel();
filename = getOutputFileName("shared",variables);
outputNames.push_back(filename); outputTypes["shared"].push_back(filename);
ofstream out2;
util.openOutputFile(filename, out2);
lookup->eliminateZeroOTUS();
printSharedData(lookup, out2, printHeaders);
out2.close();
}else {
printSharedData(lookup, out, printHeaders); //prints info to the .shared file
}
delete lookup;
processedLabels.insert(SharedList->getLabel());
userLabels.erase(SharedList->getLabel());
}
if ((util.anyLabelsToProcess(SharedList->getLabel(), userLabels, errorOff) ) && (processedLabels.count(lastLabel) != 1)) {
string saveLabel = SharedList->getLabel();
delete SharedList;
SharedList = input.getSharedListVector(lastLabel); //get new list vector to process
lookup = SharedList->getSharedRAbundVector();
m->mothurOut(lookup->getLabel()+"\n");
if (m->getControl_pressed()) {
delete SharedList; if (groupMap != NULL) { delete groupMap; } if (countTable != NULL) { delete countTable; }
delete lookup;
if (!pickedGroups) { out.close(); util.mothurRemove(filename); }
return 0;
}
//if picked groups must split the shared file by label
if (pickedGroups) {
string filename = listfile;
if (outputDir == "") { outputDir += util.hasPath(filename); }
map<string, string> variables;
variables["[filename]"] = outputDir + util.getRootName(util.getSimpleName(filename));
variables["[distance]"] = lookup->getLabel();
filename = getOutputFileName("shared",variables);
outputNames.push_back(filename); outputTypes["shared"].push_back(filename);
ofstream out2;
util.openOutputFile(filename, out2);
lookup->eliminateZeroOTUS();
printSharedData(lookup, out2, printHeaders);
out2.close();
}else {
printSharedData(lookup, out, printHeaders); //prints info to the .shared file
}
delete lookup;
processedLabels.insert(SharedList->getLabel());
userLabels.erase(SharedList->getLabel());
//restore real lastlabel to save below
SharedList->setLabel(saveLabel);
}
lastLabel = SharedList->getLabel();
delete SharedList;
SharedList = input.getSharedListVector(); //get new list vector to process
}
//output error messages about any remaining user labels
set<string>::iterator it;
bool needToRun = false;
for (it = userLabels.begin(); it != userLabels.end(); it++) {
if (processedLabels.count(lastLabel) != 1) {
needToRun = true;
}
}
//run last label if you need to
if (needToRun ) {
if (SharedList != NULL) { delete SharedList; }
SharedList = input.getSharedListVector(lastLabel); //get new list vector to process
lookup = SharedList->getSharedRAbundVector();
m->mothurOut(lookup->getLabel()+"\n");
if (m->getControl_pressed()) {
if (groupMap != NULL) { delete groupMap; } if (countTable != NULL) { delete countTable; }
if (!pickedGroups) { out.close(); util.mothurRemove(filename); }
return 0;
}
//if picked groups must split the shared file by label
if (pickedGroups) {
string filename = listfile;
if (outputDir == "") { outputDir += util.hasPath(filename); }
map<string, string> variables;
variables["[filename]"] = outputDir + util.getRootName(util.getSimpleName(filename));
variables["[distance]"] = lookup->getLabel();
filename = getOutputFileName("shared",variables);
outputNames.push_back(filename); outputTypes["shared"].push_back(filename);
ofstream out2;
util.openOutputFile(filename, out2);
lookup->eliminateZeroOTUS();
printSharedData(lookup, out2, printHeaders);
out2.close();
}else {
printSharedData(lookup, out, printHeaders); //prints info to the .shared file
}
delete lookup;
delete SharedList;
}
if (!pickedGroups) { out.close(); }
if (groupMap != NULL) { delete groupMap; } if (countTable != NULL) { delete countTable; }
if (m->getControl_pressed()) {
if (!pickedGroups) { util.mothurRemove(filename); }
return 0;
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "SharedCommand", "createSharedFromListGroup");
exit(1);
}
}
int TreeGroupCommand::execute(){
try {
if (abort) { if (calledHelp) { return 0; } return 2; }
if (format == "sharedfile") {
InputData input(sharedfile, "sharedfile", Groups);
SharedRAbundVectors* lookup = input.getSharedRAbundVectors();
lastLabel = lookup->getLabel();
Groups = lookup->getNamesGroups();
if (lookup->size() < 2) { m->mothurOut("You have not provided enough valid groups. I cannot run the command.\n"); return 0; }
//create treemap class from groupmap for tree class to use
CountTable ct;
set<string> nameMap;
map<string, string> groupMap;
set<string> gps;
for (int i = 0; i < Groups.size(); i++) {
nameMap.insert(Groups[i]);
gps.insert(Groups[i]);
groupMap[Groups[i]] = Groups[i];
}
ct.createTable(nameMap, groupMap, gps);
//fills tree names with shared files groups
Treenames = lookup->getNamesGroups();
if (m->getControl_pressed()) { return 0; }
//create tree file
makeSimsShared(input, lookup, ct);
if (m->getControl_pressed()) { for (int i = 0; i < outputNames.size(); i++) { util.mothurRemove(outputNames[i]); } return 0; }
}else{
//read in dist file
filename = inputfile;
ReadMatrix* readMatrix;
if (format == "column") { readMatrix = new ReadColumnMatrix(filename); }
else if (format == "phylip") { readMatrix = new ReadPhylipMatrix(filename); }
readMatrix->setCutoff(cutoff);
ListVector* list;
if(namefile != ""){
NameAssignment* nameMap = new NameAssignment(namefile);
nameMap->readMap();
readMatrix->read(nameMap);
list = readMatrix->getListVector();
delete nameMap;
}else if (countfile != "") {
CountTable* ct = new CountTable();
ct->readTable(countfile, true, false);
readMatrix->read(ct);
list = readMatrix->getListVector();
delete ct;
}else { NameAssignment* nameMap = NULL; readMatrix->read(nameMap); list = readMatrix->getListVector(); }
SparseDistanceMatrix* dMatrix = readMatrix->getDMatrix();
Treenames.clear();
//make treemap
CountTable ct;
set<string> nameMap;
map<string, string> groupMap;
set<string> gps;
for (int i = 0; i < list->getNumBins(); i++) {
string bin = list->get(i);
nameMap.insert(bin);
gps.insert(bin);
groupMap[bin] = bin;
Treenames.push_back(bin);
}
ct.createTable(nameMap, groupMap, gps);
vector<string> namesGroups = ct.getNamesOfGroups();
if (m->getControl_pressed()) { return 0; }
vector< vector<double> > matrix = makeSimsDist(dMatrix, list->getNumBins());
delete readMatrix;
delete dMatrix;
if (m->getControl_pressed()) { return 0; }
//create a new filename
map<string, string> variables;
variables["[filename]"] = outputDir + util.getRootName(util.getSimpleName(inputfile));
string outputFile = getOutputFileName("tree",variables);
outputNames.push_back(outputFile); outputTypes["tree"].push_back(outputFile);
Tree* newTree = new Tree(&ct, matrix, Treenames);
if (m->getControl_pressed()) { delete newTree; newTree = NULL; }
else { newTree->assembleTree(); }
if (newTree != NULL) { newTree->createNewickFile(outputFile); delete newTree; }
if (m->getControl_pressed()) { return 0; } m->mothurOut("Tree complete.\n");
}
//set tree file as new current treefile
string currentName = "";
itTypes = outputTypes.find("tree");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setTreeFile(currentName); }
}
m->mothurOut("\nOutput File Names: \n");
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i] +"\n"); } m->mothurOutEndLine();
return 0;
}
catch(exception& e) {
m->errorOut(e, "TreeGroupCommand", "execute");
exit(1);
}
}
//**********************************************************************************************************************
int RemoveRareCommand::processList(){
try {
//you must provide a label because the names in the listfile need to be consistent
string thisLabel = "";
if (allLines) { m->mothurOut("For the listfile you must select one label, using first label in your listfile."); m->mothurOutEndLine(); }
else if (labels.size() > 1) { m->mothurOut("For the listfile you must select one label, using " + (*labels.begin()) + "."); m->mothurOutEndLine(); thisLabel = *labels.begin(); }
else { thisLabel = *labels.begin(); }
InputData input(listfile, "list");
ListVector* list = input.getListVector();
//get first one or the one we want
if (thisLabel != "") {
//use smart distancing
set<string> userLabels; userLabels.insert(thisLabel);
set<string> processedLabels;
string lastLabel = list->getLabel();
while((list != NULL) && (userLabels.size() != 0)) {
if(userLabels.count(list->getLabel()) == 1){
processedLabels.insert(list->getLabel());
userLabels.erase(list->getLabel());
break;
}
if ((m->anyLabelsToProcess(list->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
processedLabels.insert(list->getLabel());
userLabels.erase(list->getLabel());
delete list;
list = input.getListVector(lastLabel);
break;
}
lastLabel = list->getLabel();
delete list;
list = input.getListVector();
}
if (userLabels.size() != 0) {
m->mothurOut("Your file does not include the label " + thisLabel + ". I will use " + lastLabel + "."); m->mothurOutEndLine();
list = input.getListVector(lastLabel);
}
}
string thisOutputDir = outputDir;
if (outputDir == "") { thisOutputDir += m->hasPath(listfile); }
map<string, string> variables;
variables["[filename]"] = thisOutputDir + m->getRootName(m->getSimpleName(listfile));
variables["[extension]"] = m->getExtension(listfile);
variables["[tag]"] = list->getLabel();
string outputFileName = getOutputFileName("list", variables);
variables["[filename]"] = thisOutputDir + m->getRootName(m->getSimpleName(groupfile));
variables["[extension]"] = m->getExtension(groupfile);
string outputGroupFileName = getOutputFileName("group", variables);
variables["[filename]"] = thisOutputDir + m->getRootName(m->getSimpleName(countfile));
variables["[extension]"] = m->getExtension(countfile);
string outputCountFileName = getOutputFileName("count", variables);
ofstream out, outGroup;
m->openOutputFile(outputFileName, out);
bool wroteSomething = false;
//if groupfile is given then use it
GroupMap* groupMap;
CountTable ct;
if (groupfile != "") {
groupMap = new GroupMap(groupfile); groupMap->readMap();
SharedUtil util;
vector<string> namesGroups = groupMap->getNamesOfGroups();
util.setGroups(Groups, namesGroups);
m->openOutputFile(outputGroupFileName, outGroup);
}else if (countfile != "") {
ct.readTable(countfile, true, false);
if (ct.hasGroupInfo()) {
vector<string> namesGroups = ct.getNamesOfGroups();
SharedUtil util;
util.setGroups(Groups, namesGroups);
}
}
if (list != NULL) {
vector<string> binLabels = list->getLabels();
vector<string> newLabels;
//make a new list vector
ListVector newList;
newList.setLabel(list->getLabel());
//for each bin
for (int i = 0; i < list->getNumBins(); i++) {
if (m->control_pressed) { if (groupfile != "") { delete groupMap; outGroup.close(); m->mothurRemove(outputGroupFileName); } out.close(); m->mothurRemove(outputFileName); return 0; }
//parse out names that are in accnos file
string binnames = list->get(i);
vector<string> names;
string saveBinNames = binnames;
m->splitAtComma(binnames, names);
int binsize = names.size();
vector<string> newGroupFile;
if (groupfile != "") {
vector<string> newNames;
saveBinNames = "";
for(int k = 0; k < names.size(); k++) {
string group = groupMap->getGroup(names[k]);
if (m->inUsersGroups(group, Groups)) {
newGroupFile.push_back(names[k] + "\t" + group);
newNames.push_back(names[k]);
saveBinNames += names[k] + ",";
}
}
names = newNames; binsize = names.size();
saveBinNames = saveBinNames.substr(0, saveBinNames.length()-1);
}else if (countfile != "") {
saveBinNames = "";
binsize = 0;
for(int k = 0; k < names.size(); k++) {
if (ct.hasGroupInfo()) {
vector<string> thisSeqsGroups = ct.getGroups(names[k]);
int thisSeqsCount = 0;
for (int n = 0; n < thisSeqsGroups.size(); n++) {
if (m->inUsersGroups(thisSeqsGroups[n], Groups)) {
thisSeqsCount += ct.getGroupCount(names[k], thisSeqsGroups[n]);
}
}
binsize += thisSeqsCount;
//if you don't have any seqs from the groups the user wants, then remove you.
if (thisSeqsCount == 0) { newGroupFile.push_back(names[k]); }
else { saveBinNames += names[k] + ","; }
}else {
binsize += ct.getNumSeqs(names[k]);
saveBinNames += names[k] + ",";
}
}
saveBinNames = saveBinNames.substr(0, saveBinNames.length()-1);
}
if (binsize > nseqs) { //keep bin
newList.push_back(saveBinNames);
newLabels.push_back(binLabels[i]);
if (groupfile != "") { for(int k = 0; k < newGroupFile.size(); k++) { outGroup << newGroupFile[k] << endl; } }
else if (countfile != "") { for(int k = 0; k < newGroupFile.size(); k++) { ct.remove(newGroupFile[k]); } }
}else { if (countfile != "") { for(int k = 0; k < names.size(); k++) { ct.remove(names[k]); } } }
}
//print new listvector
if (newList.getNumBins() != 0) {
wroteSomething = true;
newList.setLabels(newLabels);
newList.printHeaders(out);
newList.print(out);
}
}
out.close();
if (groupfile != "") { outGroup.close(); outputTypes["group"].push_back(outputGroupFileName); outputNames.push_back(outputGroupFileName); }
if (countfile != "") {
if (ct.hasGroupInfo()) {
vector<string> allGroups = ct.getNamesOfGroups();
for (int i = 0; i < allGroups.size(); i++) {
if (!m->inUsersGroups(allGroups[i], Groups)) { ct.removeGroup(allGroups[i]); }
}
}
ct.printTable(outputCountFileName);
outputTypes["count"].push_back(outputCountFileName); outputNames.push_back(outputCountFileName);
}
if (wroteSomething == false) { m->mothurOut("Your file contains only rare sequences."); m->mothurOutEndLine(); }
outputTypes["list"].push_back(outputFileName); outputNames.push_back(outputFileName);
return 0;
}
catch(exception& e) {
m->errorOut(e, "RemoveRareCommand", "processList");
exit(1);
}
}
EstOutput Unweighted::getValues(Tree* t, int p, string o) {
try {
processors = p;
outputDir = o;
CountTable* ct = t->getCountTable();
//if the users enters no groups then give them the score of all groups
int numGroups = m->getNumGroups();
//calculate number of comparsions
int numComp = 0;
vector< vector<string> > namesOfGroupCombos;
for (int r=0; r<numGroups; r++) {
for (int l = 0; l < r; l++) {
numComp++;
vector<string> groups; groups.push_back((m->getGroups())[r]); groups.push_back((m->getGroups())[l]);
namesOfGroupCombos.push_back(groups);
}
}
if (numComp != 1) {
vector<string> groups;
if (numGroups == 0) {
//get score for all users groups
for (int i = 0; i < (ct->getNamesOfGroups()).size(); i++) {
if ((ct->getNamesOfGroups())[i] != "xxx") {
groups.push_back((ct->getNamesOfGroups())[i]);
}
}
namesOfGroupCombos.push_back(groups);
}else {
for (int i = 0; i < m->getNumGroups(); i++) {
groups.push_back((m->getGroups())[i]);
}
namesOfGroupCombos.push_back(groups);
}
}
lines.clear();
int remainingPairs = namesOfGroupCombos.size();
int startIndex = 0;
for (int remainingProcessors = processors; remainingProcessors > 0; remainingProcessors--) {
int numPairs = remainingPairs; //case for last processor
if (remainingProcessors != 1) { numPairs = ceil(remainingPairs / remainingProcessors); }
lines.push_back(linePair(startIndex, numPairs)); //startIndex, numPairs
startIndex = startIndex + numPairs;
remainingPairs = remainingPairs - numPairs;
}
data = createProcesses(t, namesOfGroupCombos, ct);
lines.clear();
return data;
}
catch(exception& e) {
m->errorOut(e, "Unweighted", "getValues");
exit(1);
}
}