//**********************************************************************************************************************
int GetSeqsCommand::readCount(){
try {
string thisOutputDir = outputDir;
if (outputDir == "") { thisOutputDir += m->hasPath(countfile); }
map<string, string> variables;
variables["[filename]"] = thisOutputDir + m->getRootName(m->getSimpleName(countfile));
variables["[extension]"] = m->getExtension(countfile);
string outputFileName = getOutputFileName("count", variables);
ofstream out;
m->openOutputFile(outputFileName, out);
ifstream in;
m->openInputFile(countfile, in);
bool wroteSomething = false;
int selectedCount = 0;
string headers = m->getline(in); m->gobble(in);
out << headers << endl;
string test = headers; vector<string> pieces = m->splitWhiteSpace(test);
string name, rest; int thisTotal; rest = "";
while (!in.eof()) {
if (m->control_pressed) { in.close(); out.close(); m->mothurRemove(outputFileName); return 0; }
in >> name; m->gobble(in);
in >> thisTotal; m->gobble(in);
if (pieces.size() > 2) { rest = m->getline(in); m->gobble(in); }
if (m->debug) { m->mothurOut("[DEBUG]: " + name + '\t' + rest + "\n"); }
if (names.count(name) != 0) {
out << name << '\t' << thisTotal << '\t' << rest << endl;
wroteSomething = true;
selectedCount+= thisTotal;
}
}
in.close();
out.close();
//check for groups that have been eliminated
CountTable ct;
if (ct.testGroups(outputFileName)) {
ct.readTable(outputFileName, true, false);
ct.printTable(outputFileName);
}
if (wroteSomething == false) { m->mothurOut("Your file does not contain any sequence from the .accnos file."); m->mothurOutEndLine(); }
outputTypes["count"].push_back(outputFileName); outputNames.push_back(outputFileName);
m->mothurOut("Selected " + toString(selectedCount) + " sequences from your count file."); m->mothurOutEndLine();
return 0;
}
catch(exception& e) {
m->errorOut(e, "GetSeqsCommand", "readCount");
exit(1);
}
}
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 ListSeqsCommand::readCount(){
try {
CountTable ct;
ct.readTable(countfile, false, false);
if (m->control_pressed) { return 0; }
names = ct.getNamesOfSeqs();
return 0;
}
catch(exception& e) {
m->errorOut(e, "ListSeqsCommand", "readCount");
exit(1);
}
}
//**********************************************************************************************************************
int SharedCommand::createSharedFromCount() {
try {
//getting output filename
string filename = countfile;
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);
ofstream out; bool printHeaders = true;
util.openOutputFile(filename, out);
CountTable ct; ct.readTable(countfile, true, false);
map<string, string> seqNameToOtuName;
SharedRAbundVectors* lookup = ct.getShared(Groups, seqNameToOtuName);
lookup->setLabels(*labels.begin());
lookup->print(out, printHeaders);
out.close();
delete lookup;
string mapFilename = getOutputFileName("map",variables);
outputNames.push_back(mapFilename); outputTypes["map"].push_back(mapFilename);
ofstream outMap;
util.openOutputFile(mapFilename, outMap);
for (map<string, string>::iterator it = seqNameToOtuName.begin(); it != seqNameToOtuName.end(); it++) { outMap << it->first << '\t' << it->second << endl; }
outMap.close();
return 0;
}
catch(exception& e) {
m->errorOut(e, "SharedCommand", "createSharedFromCount");
exit(1);
}
}
int GetRAbundCommand::createRabund(CountTable& ct, ListVector*& list, RAbundVector*& rabund){
try {
rabund->setLabel(list->getLabel());
for(int i = 0; i < list->getNumBins(); i++) {
if (m->control_pressed) { return 0; }
vector<string> binNames;
string bin = list->get(i);
m->splitAtComma(bin, binNames);
int total = 0;
for (int j = 0; j < binNames.size(); j++) {
total += ct.getNumSeqs(binNames[j]);
}
rabund->push_back(total);
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "GetRAbundCommand", "createRabund");
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::createProcesses(Tree* t, vector< vector<string> > namesOfGroupCombos, CountTable* ct) {
try {
int process = 1;
vector<int> processIDS;
bool recalc = false;
EstOutput results;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
//loop through and create all the processes you want
while (process != processors) {
pid_t pid = fork();
if (pid > 0) {
processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
process++;
}else if (pid == 0){
EstOutput myresults;
myresults = driver(t, namesOfGroupCombos, lines[process].start, lines[process].num, ct);
if (m->control_pressed) { exit(0); }
//pass numSeqs to parent
ofstream out;
string tempFile = outputDir + m->mothurGetpid(process) + ".unweighted.results.temp";
m->openOutputFile(tempFile, out);
out << myresults.size() << endl;
for (int i = 0; i < myresults.size(); i++) { out << myresults[i] << '\t'; } out << endl;
out.close();
exit(0);
}else {
m->mothurOut("[ERROR]: unable to spawn the number of processes you requested, reducing number to " + toString(process) + "\n"); processors = process;
for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
//wait to die
for (int i=0;i<processIDS.size();i++) {
int temp = processIDS[i];
wait(&temp);
}
m->control_pressed = false;
for (int i=0;i<processIDS.size();i++) {
m->mothurRemove(outputDir + (toString(processIDS[i]) + ".unweighted.results.temp"));
}
recalc = true;
break;
}
}
if (recalc) {
//test line, also set recalc to true.
//for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); } for (int i=0;i<processIDS.size();i++) { int temp = processIDS[i]; wait(&temp); } m->control_pressed = false; for (int i=0;i<processIDS.size();i++) {m->mothurRemove(outputDir + (toString(processIDS[i]) + ".unweighted.results.temp"));}processors=3; m->mothurOut("[ERROR]: unable to spawn the number of processes you requested, reducing number to " + toString(processors) + "\n");
//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;
}
results.clear();
processIDS.resize(0);
process = 1;
//loop through and create all the processes you want
while (process != processors) {
pid_t pid = fork();
if (pid > 0) {
processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
process++;
}else if (pid == 0){
EstOutput myresults;
myresults = driver(t, namesOfGroupCombos, lines[process].start, lines[process].num, ct);
if (m->control_pressed) { exit(0); }
//pass numSeqs to parent
ofstream out;
string tempFile = outputDir + m->mothurGetpid(process) + ".unweighted.results.temp";
m->openOutputFile(tempFile, out);
out << myresults.size() << endl;
for (int i = 0; i < myresults.size(); i++) { out << myresults[i] << '\t'; } out << endl;
out.close();
exit(0);
}else {
m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine();
for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
exit(0);
}
}
}
results = driver(t, namesOfGroupCombos, lines[0].start, lines[0].num, ct);
//force parent to wait until all the processes are done
for (int i=0;i<(processors-1);i++) {
int temp = processIDS[i];
wait(&temp);
}
if (m->control_pressed) { return results; }
//get data created by processes
for (int i=0;i<(processors-1);i++) {
ifstream in;
string s = outputDir + toString(processIDS[i]) + ".unweighted.results.temp";
m->openInputFile(s, in);
//get quantiles
if (!in.eof()) {
int num;
in >> num; m->gobble(in);
if (m->control_pressed) { break; }
double w;
for (int j = 0; j < num; j++) {
in >> w;
results.push_back(w);
}
m->gobble(in);
}
in.close();
m->mothurRemove(s);
}
#else
//fill in functions
vector<unweightedData*> pDataArray;
DWORD dwThreadIdArray[processors-1];
HANDLE hThreadArray[processors-1];
vector<CountTable*> cts;
vector<Tree*> trees;
//Create processor worker threads.
for( int i=1; i<processors; i++ ){
CountTable* copyCount = new CountTable();
copyCount->copy(ct);
Tree* copyTree = new Tree(copyCount);
copyTree->getCopy(t);
cts.push_back(copyCount);
trees.push_back(copyTree);
unweightedData* tempweighted = new unweightedData(m, lines[i].start, lines[i].num, namesOfGroupCombos, copyTree, copyCount, includeRoot);
pDataArray.push_back(tempweighted);
processIDS.push_back(i);
hThreadArray[i-1] = CreateThread(NULL, 0, MyUnWeightedThreadFunction, pDataArray[i-1], 0, &dwThreadIdArray[i-1]);
}
results = driver(t, namesOfGroupCombos, lines[0].start, lines[0].num, ct);
//Wait until all threads have terminated.
WaitForMultipleObjects(processors-1, hThreadArray, TRUE, INFINITE);
//Close all thread handles and free memory allocations.
for(int i=0; i < pDataArray.size(); i++){
for (int j = 0; j < pDataArray[i]->results.size(); j++) { results.push_back(pDataArray[i]->results[j]); }
delete cts[i];
delete trees[i];
CloseHandle(hThreadArray[i]);
delete pDataArray[i];
}
#endif
return results;
}
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);
}
}
int SeqSummaryCommand::execute(){
try{
if (abort == true) { if (calledHelp) { return 0; } return 2; }
//set current fasta to fastafile
m->setFastaFile(fastafile);
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastafile));
string summaryFile = getOutputFileName("summary",variables);
int numSeqs = 0;
vector<int> startPosition;
vector<int> endPosition;
vector<int> seqLength;
vector<int> ambigBases;
vector<int> longHomoPolymer;
if (namefile != "") { nameMap = m->readNames(namefile); }
else if (countfile != "") {
CountTable ct;
ct.readTable(countfile, false, false);
nameMap = ct.getNameMap();
}
if (m->control_pressed) { return 0; }
#ifdef USE_MPI
int pid, numSeqsPerProcessor;
int tag = 2001;
int startTag = 1; int endTag = 2; int lengthTag = 3; int baseTag = 4; int lhomoTag = 5;
int outMode=MPI_MODE_CREATE|MPI_MODE_WRONLY;
vector<unsigned long long> MPIPos;
MPI_Status status;
MPI_Status statusOut;
MPI_File inMPI;
MPI_File outMPI;
MPI_Comm_size(MPI_COMM_WORLD, &processors);
MPI_Comm_rank(MPI_COMM_WORLD, &pid);
char tempFileName[1024];
strcpy(tempFileName, fastafile.c_str());
char sumFileName[1024];
strcpy(sumFileName, summaryFile.c_str());
MPI_File_open(MPI_COMM_WORLD, tempFileName, MPI_MODE_RDONLY, MPI_INFO_NULL, &inMPI); //comm, filename, mode, info, filepointer
MPI_File_open(MPI_COMM_WORLD, sumFileName, outMode, MPI_INFO_NULL, &outMPI);
if (m->control_pressed) { MPI_File_close(&inMPI); MPI_File_close(&outMPI); return 0; }
if (pid == 0) { //you are the root process
//print header
string outputString = "seqname\tstart\tend\tnbases\tambigs\tpolymer\tnumSeqs\n";
int length = outputString.length();
char* buf2 = new char[length];
memcpy(buf2, outputString.c_str(), length);
MPI_File_write_shared(outMPI, buf2, length, MPI_CHAR, &statusOut);
delete buf2;
MPIPos = m->setFilePosFasta(fastafile, numSeqs); //fills MPIPos, returns numSeqs
for(int i = 1; i < processors; i++) {
MPI_Send(&numSeqs, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
MPI_Send(&MPIPos[0], (numSeqs+1), MPI_LONG, i, tag, MPI_COMM_WORLD);
}
//figure out how many sequences you have to do
numSeqsPerProcessor = numSeqs / processors;
int startIndex = pid * numSeqsPerProcessor;
if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
//do your part
MPICreateSummary(startIndex, numSeqsPerProcessor, startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, inMPI, outMPI, MPIPos);
}else { //i am the child process
MPI_Recv(&numSeqs, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &status);
MPIPos.resize(numSeqs+1);
MPI_Recv(&MPIPos[0], (numSeqs+1), MPI_LONG, 0, tag, MPI_COMM_WORLD, &status);
//figure out how many sequences you have to align
numSeqsPerProcessor = numSeqs / processors;
int startIndex = pid * numSeqsPerProcessor;
if(pid == (processors - 1)){ numSeqsPerProcessor = numSeqs - pid * numSeqsPerProcessor; }
//do your part
MPICreateSummary(startIndex, numSeqsPerProcessor, startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, inMPI, outMPI, MPIPos);
}
MPI_File_close(&inMPI);
MPI_File_close(&outMPI);
MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
if (pid == 0) {
//get the info from the child processes
for(int i = 1; i < processors; i++) {
int size;
MPI_Recv(&size, 1, MPI_INT, i, tag, MPI_COMM_WORLD, &status);
vector<int> temp; temp.resize(size+1);
for(int j = 0; j < 5; j++) {
MPI_Recv(&temp[0], (size+1), MPI_INT, i, 2001, MPI_COMM_WORLD, &status);
int receiveTag = temp[temp.size()-1]; //child process added a int to the end to indicate what count this is for
if (receiveTag == startTag) {
for (int k = 0; k < size; k++) { startPosition.push_back(temp[k]); }
}else if (receiveTag == endTag) {
for (int k = 0; k < size; k++) { endPosition.push_back(temp[k]); }
}else if (receiveTag == lengthTag) {
for (int k = 0; k < size; k++) { seqLength.push_back(temp[k]); }
}else if (receiveTag == baseTag) {
for (int k = 0; k < size; k++) { ambigBases.push_back(temp[k]); }
}else if (receiveTag == lhomoTag) {
for (int k = 0; k < size; k++) { longHomoPolymer.push_back(temp[k]); }
}
}
}
}else{
//send my counts
int size = startPosition.size();
MPI_Send(&size, 1, MPI_INT, 0, tag, MPI_COMM_WORLD);
startPosition.push_back(startTag);
int ierr = MPI_Send(&(startPosition[0]), (size+1), MPI_INT, 0, 2001, MPI_COMM_WORLD);
endPosition.push_back(endTag);
ierr = MPI_Send (&(endPosition[0]), (size+1), MPI_INT, 0, 2001, MPI_COMM_WORLD);
seqLength.push_back(lengthTag);
ierr = MPI_Send(&(seqLength[0]), (size+1), MPI_INT, 0, 2001, MPI_COMM_WORLD);
ambigBases.push_back(baseTag);
ierr = MPI_Send(&(ambigBases[0]), (size+1), MPI_INT, 0, 2001, MPI_COMM_WORLD);
longHomoPolymer.push_back(lhomoTag);
ierr = MPI_Send(&(longHomoPolymer[0]), (size+1), MPI_INT, 0, 2001, MPI_COMM_WORLD);
}
MPI_Barrier(MPI_COMM_WORLD); //make everyone wait - just in case
#else
vector<unsigned long long> positions;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
positions = m->divideFile(fastafile, processors);
for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); }
#else
positions = m->setFilePosFasta(fastafile, numSeqs);
if (positions.size() < processors) { processors = positions.size(); }
//figure out how many sequences you have to process
int numSeqsPerProcessor = numSeqs / processors;
for (int i = 0; i < processors; i++) {
int startIndex = i * numSeqsPerProcessor;
if(i == (processors - 1)){ numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; }
lines.push_back(new linePair(positions[startIndex], numSeqsPerProcessor));
}
#endif
if(processors == 1){
numSeqs = driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile, lines[0]);
}else{
numSeqs = createProcessesCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile);
}
if (m->control_pressed) { return 0; }
#endif
#ifdef USE_MPI
if (pid == 0) {
#endif
sort(startPosition.begin(), startPosition.end());
sort(endPosition.begin(), endPosition.end());
sort(seqLength.begin(), seqLength.end());
sort(ambigBases.begin(), ambigBases.end());
sort(longHomoPolymer.begin(), longHomoPolymer.end());
int size = startPosition.size();
//find means
unsigned long long meanStartPosition, meanEndPosition, meanSeqLength, meanAmbigBases, meanLongHomoPolymer;
meanStartPosition = 0; meanEndPosition = 0; meanSeqLength = 0; meanAmbigBases = 0; meanLongHomoPolymer = 0;
for (int i = 0; i < size; i++) {
meanStartPosition += startPosition[i];
meanEndPosition += endPosition[i];
meanSeqLength += seqLength[i];
meanAmbigBases += ambigBases[i];
meanLongHomoPolymer += longHomoPolymer[i];
}
double meanstartPosition, meanendPosition, meanseqLength, meanambigBases, meanlongHomoPolymer;
meanstartPosition = meanStartPosition / (double) size; meanendPosition = meanEndPosition /(double) size; meanlongHomoPolymer = meanLongHomoPolymer / (double) size; meanseqLength = meanSeqLength / (double) size; meanambigBases = meanAmbigBases /(double) size;
int ptile0_25 = int(size * 0.025);
int ptile25 = int(size * 0.250);
int ptile50 = int(size * 0.500);
int ptile75 = int(size * 0.750);
int ptile97_5 = int(size * 0.975);
int ptile100 = size - 1;
//to compensate for blank sequences that would result in startPosition and endPostion equalling -1
if (startPosition[0] == -1) { startPosition[0] = 0; }
if (endPosition[0] == -1) { endPosition[0] = 0; }
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
m->mothurOutEndLine();
m->mothurOut("\t\tStart\tEnd\tNBases\tAmbigs\tPolymer\tNumSeqs"); m->mothurOutEndLine();
m->mothurOut("Minimum:\t" + toString(startPosition[0]) + "\t" + toString(endPosition[0]) + "\t" + toString(seqLength[0]) + "\t" + toString(ambigBases[0]) + "\t" + toString(longHomoPolymer[0]) + "\t" + toString(1)); m->mothurOutEndLine();
m->mothurOut("2.5%-tile:\t" + toString(startPosition[ptile0_25]) + "\t" + toString(endPosition[ptile0_25]) + "\t" + toString(seqLength[ptile0_25]) + "\t" + toString(ambigBases[ptile0_25]) + "\t"+ toString(longHomoPolymer[ptile0_25]) + "\t" + toString(ptile0_25+1)); m->mothurOutEndLine();
m->mothurOut("25%-tile:\t" + toString(startPosition[ptile25]) + "\t" + toString(endPosition[ptile25]) + "\t" + toString(seqLength[ptile25]) + "\t" + toString(ambigBases[ptile25]) + "\t" + toString(longHomoPolymer[ptile25]) + "\t" + toString(ptile25+1)); m->mothurOutEndLine();
m->mothurOut("Median: \t" + toString(startPosition[ptile50]) + "\t" + toString(endPosition[ptile50]) + "\t" + toString(seqLength[ptile50]) + "\t" + toString(ambigBases[ptile50]) + "\t" + toString(longHomoPolymer[ptile50]) + "\t" + toString(ptile50+1)); m->mothurOutEndLine();
m->mothurOut("75%-tile:\t" + toString(startPosition[ptile75]) + "\t" + toString(endPosition[ptile75]) + "\t" + toString(seqLength[ptile75]) + "\t" + toString(ambigBases[ptile75]) + "\t" + toString(longHomoPolymer[ptile75]) + "\t" + toString(ptile75+1)); m->mothurOutEndLine();
m->mothurOut("97.5%-tile:\t" + toString(startPosition[ptile97_5]) + "\t" + toString(endPosition[ptile97_5]) + "\t" + toString(seqLength[ptile97_5]) + "\t" + toString(ambigBases[ptile97_5]) + "\t" + toString(longHomoPolymer[ptile97_5]) + "\t" + toString(ptile97_5+1)); m->mothurOutEndLine();
m->mothurOut("Maximum:\t" + toString(startPosition[ptile100]) + "\t" + toString(endPosition[ptile100]) + "\t" + toString(seqLength[ptile100]) + "\t" + toString(ambigBases[ptile100]) + "\t" + toString(longHomoPolymer[ptile100]) + "\t" + toString(ptile100+1)); m->mothurOutEndLine();
m->mothurOut("Mean:\t" + toString(meanstartPosition) + "\t" + toString(meanendPosition) + "\t" + toString(meanseqLength) + "\t" + toString(meanambigBases) + "\t" + toString(meanlongHomoPolymer)); m->mothurOutEndLine();
if ((namefile == "") && (countfile == "")) { m->mothurOut("# of Seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); }
else { m->mothurOut("# of unique seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); m->mothurOut("total # of seqs:\t" + toString(startPosition.size())); m->mothurOutEndLine(); }
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
m->mothurOut(summaryFile); m->mothurOutEndLine(); outputNames.push_back(summaryFile); outputTypes["summary"].push_back(summaryFile);
m->mothurOutEndLine();
#ifdef USE_MPI
}
#endif
//set fasta file as new current fastafile
string current = "";
itTypes = outputTypes.find("summary");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSummaryFile(current); }
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "SeqSummaryCommand", "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);
}
}
EstOutput Parsimony::createProcesses(Tree* t, vector< vector<string> > namesOfGroupCombos, CountTable* ct) {
try {
int process = 1;
vector<int> processIDS;
EstOutput results;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
//loop through and create all the processes you want
while (process != processors) {
int pid = fork();
if (pid > 0) {
processIDS.push_back(pid); //create map from line number to pid so you can append files in correct order later
process++;
}else if (pid == 0){
EstOutput myresults;
myresults = driver(t, namesOfGroupCombos, lines[process].start, lines[process].num, ct);
if (m->control_pressed) { exit(0); }
//pass numSeqs to parent
ofstream out;
string tempFile = outputDir + toString(getpid()) + ".pars.results.temp";
m->openOutputFile(tempFile, out);
out << myresults.size() << endl;
for (int i = 0; i < myresults.size(); i++) { out << myresults[i] << '\t'; } out << endl;
out.close();
exit(0);
}else {
m->mothurOut("[ERROR]: unable to spawn the necessary processes."); m->mothurOutEndLine();
for (int i = 0; i < processIDS.size(); i++) { kill (processIDS[i], SIGINT); }
exit(0);
}
}
results = driver(t, namesOfGroupCombos, lines[0].start, lines[0].num, ct);
//force parent to wait until all the processes are done
for (int i=0;i<processIDS.size();i++) {
int temp = processIDS[i];
wait(&temp);
}
if (m->control_pressed) { return results; }
//get data created by processes
for (int i=0;i<processIDS.size();i++) {
ifstream in;
string s = outputDir + toString(processIDS[i]) + ".pars.results.temp";
m->openInputFile(s, in);
//get scores
if (!in.eof()) {
int num;
in >> num; m->gobble(in);
if (m->control_pressed) { break; }
double w;
for (int j = 0; j < num; j++) {
in >> w;
results.push_back(w);
}
m->gobble(in);
}
in.close();
m->mothurRemove(s);
}
#else
//fill in functions
vector<parsData*> pDataArray;
DWORD dwThreadIdArray[processors-1];
HANDLE hThreadArray[processors-1];
vector<CountTable*> cts;
vector<Tree*> trees;
//Create processor worker threads.
for( int i=1; i<processors; i++ ){
CountTable* copyCount = new CountTable();
copyCount->copy(ct);
Tree* copyTree = new Tree(copyCount);
copyTree->getCopy(t);
cts.push_back(copyCount);
trees.push_back(copyTree);
parsData* temppars = new parsData(m, lines[i].start, lines[i].num, namesOfGroupCombos, copyTree, copyCount);
pDataArray.push_back(temppars);
processIDS.push_back(i);
hThreadArray[i-1] = CreateThread(NULL, 0, MyParsimonyThreadFunction, pDataArray[i-1], 0, &dwThreadIdArray[i-1]);
}
results = driver(t, namesOfGroupCombos, lines[0].start, lines[0].num, ct);
//Wait until all threads have terminated.
WaitForMultipleObjects(processors-1, hThreadArray, TRUE, INFINITE);
//Close all thread handles and free memory allocations.
for(int i=0; i < pDataArray.size(); i++){
for (int j = 0; j < pDataArray[i]->results.size(); j++) { results.push_back(pDataArray[i]->results[j]); }
delete cts[i];
delete trees[i];
CloseHandle(hThreadArray[i]);
delete pDataArray[i];
}
#endif
return results;
}
//**********************************************************************************************************************
ClassifyOtuCommand::ClassifyOtuCommand(string option) {
try{
abort = false; calledHelp = false;
allLines = 1;
labels.clear();
//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();
OptionParser parser(option);
map<string, string> parameters = parser.getParameters();
ValidParameters validParameter;
map<string, string>::iterator it;
//check to make sure all parameters are valid for command
for (it = parameters.begin(); it != parameters.end(); it++) {
if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
}
//initialize outputTypes
vector<string> tempOutNames;
outputTypes["constaxonomy"] = tempOutNames;
outputTypes["taxsummary"] = tempOutNames;
//if the user changes the input directory command factory will send this info to us in the output parameter
string inputDir = validParameter.validFile(parameters, "inputdir", false);
if (inputDir == "not found"){ inputDir = ""; }
else {
string path;
it = parameters.find("list");
//user has given a template file
if(it != parameters.end()){
path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["list"] = inputDir + it->second; }
}
it = parameters.find("name");
//user has given a template file
if(it != parameters.end()){
path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["name"] = inputDir + it->second; }
}
it = parameters.find("taxonomy");
//user has given a template file
if(it != parameters.end()){
path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["taxonomy"] = inputDir + it->second; }
}
it = parameters.find("reftaxonomy");
//user has given a template file
if(it != parameters.end()){
path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["reftaxonomy"] = inputDir + it->second; }
}
it = parameters.find("group");
//user has given a template file
if(it != parameters.end()){
path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["group"] = inputDir + it->second; }
}
it = parameters.find("count");
//user has given a template file
if(it != parameters.end()){
path = m->hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["count"] = inputDir + it->second; }
}
}
//if the user changes the output directory command factory will send this info to us in the output parameter
outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = ""; }
//check for required parameters
listfile = validParameter.validFile(parameters, "list", true);
if (listfile == "not found") {
//if there is a current list file, use it
listfile = m->getListFile();
if (listfile != "") { m->mothurOut("Using " + listfile + " as input file for the list parameter."); m->mothurOutEndLine(); }
else { m->mothurOut("You have no current listfile and the list parameter is required."); m->mothurOutEndLine(); abort = true; }
}
else if (listfile == "not open") { abort = true; }
else { m->setListFile(listfile); }
taxfile = validParameter.validFile(parameters, "taxonomy", true);
if (taxfile == "not found") { //if there is a current list file, use it
taxfile = m->getTaxonomyFile();
if (taxfile != "") { m->mothurOut("Using " + taxfile + " as input file for the taxonomy parameter."); m->mothurOutEndLine(); }
else { m->mothurOut("You have no current taxonomy file and the taxonomy parameter is required."); m->mothurOutEndLine(); abort = true; }
}
else if (taxfile == "not open") { abort = true; }
else { m->setTaxonomyFile(taxfile); }
refTaxonomy = validParameter.validFile(parameters, "reftaxonomy", true);
if (refTaxonomy == "not found") { refTaxonomy = ""; m->mothurOut("reftaxonomy is not required, but if given will keep the rankIDs in the summary file static."); m->mothurOutEndLine(); }
else if (refTaxonomy == "not open") { abort = true; }
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") { abort = true; }
else if (groupfile == "not found") { groupfile = ""; }
else { m->setGroupFile(groupfile); }
countfile = validParameter.validFile(parameters, "count", true);
if (countfile == "not open") { countfile = ""; abort = true; }
else if (countfile == "not found") { countfile = ""; }
else { m->setCountTableFile(countfile); }
if ((namefile != "") && (countfile != "")) {
m->mothurOut("[ERROR]: you may only use one of the following: name or count."); m->mothurOutEndLine(); abort = true;
}
if ((groupfile != "") && (countfile != "")) {
m->mothurOut("[ERROR]: you may only use one of the following: group or count."); m->mothurOutEndLine(); abort=true;
}
//check for optional parameter and set defaults
// ...at some point should added some additional type checking...
label = validParameter.validFile(parameters, "label", false);
if (label == "not found") { label = ""; allLines = 1; }
else {
if(label != "all") { m->splitAtDash(label, labels); allLines = 0; }
else { allLines = 1; }
}
basis = validParameter.validFile(parameters, "basis", false);
if (basis == "not found") { basis = "otu"; }
if ((basis != "otu") && (basis != "sequence")) { m->mothurOut("Invalid option for basis. basis options are otu and sequence, using otu."); m->mothurOutEndLine(); }
string temp = validParameter.validFile(parameters, "cutoff", false); if (temp == "not found") { temp = "51"; }
m->mothurConvert(temp, cutoff);
temp = validParameter.validFile(parameters, "probs", false); if (temp == "not found"){ temp = "true"; }
probs = m->isTrue(temp);
temp = validParameter.validFile(parameters, "persample", false); if (temp == "not found"){ temp = "f"; }
persample = m->isTrue(temp);
if ((groupfile == "") && (countfile == "")) { if (persample) { m->mothurOut("persample is only valid with a group file, or count file with group information. Setting persample=f.\n"); persample = false; }
}
if (countfile != "") {
CountTable cts;
if (!cts.testGroups(countfile)) {
if (persample) { m->mothurOut("persample is only valid with a group file, or count file with group information. Setting persample=f.\n"); persample = false; }
}
}
if ((cutoff < 51) || (cutoff > 100)) { m->mothurOut("cutoff must be above 50, and no greater than 100."); m->mothurOutEndLine(); abort = true; }
if (countfile == "") {
if (namefile == ""){
vector<string> files; files.push_back(taxfile);
parser.getNameFile(files);
}
}
}
}
catch(exception& e) {
m->errorOut(e, "ClassifyOtuCommand", "ClassifyOtuCommand");
exit(1);
}
}
int SeqSummaryCommand::execute(){
try{
if (abort == true) { if (calledHelp) { return 0; } return 2; }
int start = time(NULL);
//set current fasta to fastafile
m->setFastaFile(fastafile);
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastafile));
string summaryFile = getOutputFileName("summary",variables);
long long numSeqs = 0;
long long size = 0;
long long numUniques = 0;
map<int, long long> startPosition;
map<int, long long> endPosition;
map<int, long long> seqLength;
map<int, long long> ambigBases;
map<int, long long> longHomoPolymer;
if (namefile != "") { nameMap = m->readNames(namefile); numUniques = nameMap.size(); }
else if (countfile != "") {
CountTable ct;
ct.readTable(countfile, false, false);
nameMap = ct.getNameMap();
size = ct.getNumSeqs();
numUniques = ct.getNumUniqueSeqs();
}
if (m->control_pressed) { return 0; }
vector<unsigned long long> positions;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
positions = m->divideFile(fastafile, processors);
for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); }
#else
positions = m->setFilePosFasta(fastafile, numSeqs);
if (numSeqs < processors) { processors = numSeqs; }
//figure out how many sequences you have to process
int numSeqsPerProcessor = numSeqs / processors;
for (int i = 0; i < processors; i++) {
int startIndex = i * numSeqsPerProcessor;
if(i == (processors - 1)){ numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; }
lines.push_back(new linePair(positions[startIndex], numSeqsPerProcessor));
}
#endif
if(processors == 1){
numSeqs = driverCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile, lines[0]);
}else{
numSeqs = createProcessesCreateSummary(startPosition, endPosition, seqLength, ambigBases, longHomoPolymer, fastafile, summaryFile);
}
if (m->control_pressed) { return 0; }
//set size
if (countfile != "") {}//already set
else if (namefile == "") { size = numSeqs; }
else { for (map<int, long long>::iterator it = startPosition.begin(); it != startPosition.end(); it++) { size += it->second; } }
if ((namefile != "") || (countfile != "")) {
string type = "count";
if (namefile != "") { type = "name"; }
if (numSeqs != numUniques) { // do fasta and name/count files match
m->mothurOut("[ERROR]: Your " + type + " file contains " + toString(numUniques) + " unique sequences, but your fasta file contains " + toString(numSeqs) + ". File mismatch detected, quitting command.\n"); m->control_pressed = true;
}
}
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
long long ptile0_25 = 1+(long long)(size * 0.025); //number of sequences at 2.5%
long long ptile25 = 1+(long long)(size * 0.250); //number of sequences at 25%
long long ptile50 = 1+(long long)(size * 0.500);
long long ptile75 = 1+(long long)(size * 0.750);
long long ptile97_5 = 1+(long long)(size * 0.975);
long long ptile100 = (long long)(size);
vector<int> starts; starts.resize(7,0); vector<int> ends; ends.resize(7,0); vector<int> ambigs; ambigs.resize(7,0); vector<int> lengths; lengths.resize(7,0); vector<int> homops; homops.resize(7,0);
//find means
long long meanStartPosition, meanEndPosition, meanSeqLength, meanAmbigBases, meanLongHomoPolymer;
meanStartPosition = 0; meanEndPosition = 0; meanSeqLength = 0; meanAmbigBases = 0; meanLongHomoPolymer = 0;
//minimum
if ((startPosition.begin())->first == -1) { starts[0] = 0; }
else {starts[0] = (startPosition.begin())->first; }
long long totalSoFar = 0;
//set all values to min
starts[1] = starts[0]; starts[2] = starts[0]; starts[3] = starts[0]; starts[4] = starts[0]; starts[5] = starts[0];
int lastValue = 0;
for (map<int, long long>::iterator it = startPosition.begin(); it != startPosition.end(); it++) {
int value = it->first; if (value == -1) { value = 0; }
meanStartPosition += (value*it->second);
totalSoFar += it->second;
if (((totalSoFar <= ptile0_25) && (totalSoFar > 1)) || ((lastValue < ptile0_25) && (totalSoFar > ptile0_25))){ starts[1] = value; } //save value
if (((totalSoFar <= ptile25) && (totalSoFar > ptile0_25)) || ((lastValue < ptile25) && (totalSoFar > ptile25))) { starts[2] = value; } //save value
if (((totalSoFar <= ptile50) && (totalSoFar > ptile25)) || ((lastValue < ptile50) && (totalSoFar > ptile50))) { starts[3] = value; } //save value
if (((totalSoFar <= ptile75) && (totalSoFar > ptile50)) || ((lastValue < ptile75) && (totalSoFar > ptile75))) { starts[4] = value; } //save value
if (((totalSoFar <= ptile97_5) && (totalSoFar > ptile75)) || ((lastValue < ptile97_5) && (totalSoFar > ptile97_5))) { starts[5] = value; } //save value
if ((totalSoFar <= ptile100) && (totalSoFar > ptile97_5)) { starts[6] = value; } //save value
lastValue = totalSoFar;
}
starts[6] = (startPosition.rbegin())->first;
if ((endPosition.begin())->first == -1) { ends[0] = 0; }
else {ends[0] = (endPosition.begin())->first; }
totalSoFar = 0;
//set all values to min
ends[1] = ends[0]; ends[2] = ends[0]; ends[3] = ends[0]; ends[4] = ends[0]; ends[5] = ends[0];
lastValue = 0;
for (map<int, long long>::iterator it = endPosition.begin(); it != endPosition.end(); it++) {
int value = it->first; if (value == -1) { value = 0; }
meanEndPosition += (value*it->second);
totalSoFar += it->second;
if (((totalSoFar <= ptile0_25) && (totalSoFar > 1)) || ((lastValue < ptile0_25) && (totalSoFar > ptile0_25))){ ends[1] = value; } //save value
if (((totalSoFar <= ptile25) && (totalSoFar > ptile0_25)) || ((lastValue < ptile25) && (totalSoFar > ptile25))) { ends[2] = value; } //save value
if (((totalSoFar <= ptile50) && (totalSoFar > ptile25)) || ((lastValue < ptile50) && (totalSoFar > ptile50))) { ends[3] = value; } //save value
if (((totalSoFar <= ptile75) && (totalSoFar > ptile50)) || ((lastValue < ptile75) && (totalSoFar > ptile75))) { ends[4] = value; } //save value
if (((totalSoFar <= ptile97_5) && (totalSoFar > ptile75)) || ((lastValue < ptile97_5) && (totalSoFar > ptile97_5))) { ends[5] = value; } //save value
if ((totalSoFar <= ptile100) && (totalSoFar > ptile97_5)) { ends[6] = value; } //save value
lastValue = totalSoFar;
}
ends[6] = (endPosition.rbegin())->first;
if ((seqLength.begin())->first == -1) { lengths[0] = 0; }
else {lengths[0] = (seqLength.begin())->first; }
//set all values to min
lengths[1] = lengths[0]; lengths[2] = lengths[0]; lengths[3] = lengths[0]; lengths[4] = lengths[0]; lengths[5] = lengths[0];
totalSoFar = 0;
lastValue = 0;
for (map<int, long long>::iterator it = seqLength.begin(); it != seqLength.end(); it++) {
int value = it->first;
meanSeqLength += (value*it->second);
totalSoFar += it->second;
if (((totalSoFar <= ptile0_25) && (totalSoFar > 1)) || ((lastValue < ptile0_25) && (totalSoFar > ptile0_25))){ lengths[1] = value; } //save value
if (((totalSoFar <= ptile25) && (totalSoFar > ptile0_25)) || ((lastValue < ptile25) && (totalSoFar > ptile25))) { lengths[2] = value; } //save value
if (((totalSoFar <= ptile50) && (totalSoFar > ptile25)) || ((lastValue < ptile50) && (totalSoFar > ptile50))) { lengths[3] = value; } //save value
if (((totalSoFar <= ptile75) && (totalSoFar > ptile50)) || ((lastValue < ptile75) && (totalSoFar > ptile75))) { lengths[4] = value; } //save value
if (((totalSoFar <= ptile97_5) && (totalSoFar > ptile75)) || ((lastValue < ptile97_5) && (totalSoFar > ptile97_5))) { lengths[5] = value; } //save value
if ((totalSoFar <= ptile100) && (totalSoFar > ptile97_5)) { lengths[6] = value; } //save value
lastValue = totalSoFar;
}
lengths[6] = (seqLength.rbegin())->first;
if ((ambigBases.begin())->first == -1) { ambigs[0] = 0; }
else {ambigs[0] = (ambigBases.begin())->first; }
//set all values to min
ambigs[1] = ambigs[0]; ambigs[2] = ambigs[0]; ambigs[3] = ambigs[0]; ambigs[4] = ambigs[0]; ambigs[5] = ambigs[0];
totalSoFar = 0;
lastValue = 0;
for (map<int, long long>::iterator it = ambigBases.begin(); it != ambigBases.end(); it++) {
int value = it->first;
meanAmbigBases += (value*it->second);
totalSoFar += it->second;
if (((totalSoFar <= ptile0_25) && (totalSoFar > 1)) || ((lastValue < ptile0_25) && (totalSoFar > ptile0_25))){ ambigs[1] = value; } //save value
if (((totalSoFar <= ptile25) && (totalSoFar > ptile0_25)) || ((lastValue < ptile25) && (totalSoFar > ptile25))) { ambigs[2] = value; } //save value
if (((totalSoFar <= ptile50) && (totalSoFar > ptile25)) || ((lastValue < ptile50) && (totalSoFar > ptile50))) { ambigs[3] = value; } //save value
if (((totalSoFar <= ptile75) && (totalSoFar > ptile50)) || ((lastValue < ptile75) && (totalSoFar > ptile75))) { ambigs[4] = value; } //save value
if (((totalSoFar <= ptile97_5) && (totalSoFar > ptile75)) || ((lastValue < ptile97_5) && (totalSoFar > ptile97_5))) { ambigs[5] = value; } //save value
if ((totalSoFar <= ptile100) && (totalSoFar > ptile97_5)) { ambigs[6] = value; } //save value
lastValue = totalSoFar;
}
ambigs[6] = (ambigBases.rbegin())->first;
if ((longHomoPolymer.begin())->first == -1) { homops[0] = 0; }
else {homops[0] = (longHomoPolymer.begin())->first; }
//set all values to min
homops[1] = homops[0]; homops[2] = homops[0]; homops[3] = homops[0]; homops[4] = homops[0]; homops[5] = homops[0];
totalSoFar = 0;
lastValue = 0;
for (map<int, long long>::iterator it = longHomoPolymer.begin(); it != longHomoPolymer.end(); it++) {
int value = it->first;
meanLongHomoPolymer += (it->first*it->second);
totalSoFar += it->second;
if (((totalSoFar <= ptile0_25) && (totalSoFar > 1)) || ((lastValue < ptile0_25) && (totalSoFar > ptile0_25))){ homops[1] = value; } //save value
if (((totalSoFar <= ptile25) && (totalSoFar > ptile0_25)) || ((lastValue < ptile25) && (totalSoFar > ptile25))) { homops[2] = value; } //save value
if (((totalSoFar <= ptile50) && (totalSoFar > ptile25)) || ((lastValue < ptile50) && (totalSoFar > ptile50))) { homops[3] = value; } //save value
if (((totalSoFar <= ptile75) && (totalSoFar > ptile50)) || ((lastValue < ptile75) && (totalSoFar > ptile75))) { homops[4] = value; } //save value
if (((totalSoFar <= ptile97_5) && (totalSoFar > ptile75)) || ((lastValue < ptile97_5) && (totalSoFar > ptile97_5))) { homops[5] = value; } //save value
if ((totalSoFar <= ptile100) && (totalSoFar > ptile97_5)) { homops[6] = value; } //save value
lastValue = totalSoFar;
}
homops[6] = (longHomoPolymer.rbegin())->first;
double meanstartPosition, meanendPosition, meanseqLength, meanambigBases, meanlongHomoPolymer;
meanstartPosition = meanStartPosition / (double) size; meanendPosition = meanEndPosition /(double) size; meanlongHomoPolymer = meanLongHomoPolymer / (double) size; meanseqLength = meanSeqLength / (double) size; meanambigBases = meanAmbigBases /(double) size;
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
m->mothurOutEndLine();
m->mothurOut("\t\tStart\tEnd\tNBases\tAmbigs\tPolymer\tNumSeqs"); m->mothurOutEndLine();
m->mothurOut("Minimum:\t" + toString(starts[0]) + "\t" + toString(ends[0]) + "\t" + toString(lengths[0]) + "\t" + toString(ambigs[0]) + "\t" + toString(homops[0]) + "\t" + toString(1)); m->mothurOutEndLine();
m->mothurOut("2.5%-tile:\t" + toString(starts[1]) + "\t" + toString(ends[1]) + "\t" + toString(lengths[1]) + "\t" + toString(ambigs[1]) + "\t" + toString(homops[1]) + "\t" + toString(ptile0_25)); m->mothurOutEndLine();
m->mothurOut("25%-tile:\t" + toString(starts[2]) + "\t" + toString(ends[2]) + "\t" + toString(lengths[2]) + "\t" + toString(ambigs[2]) + "\t" + toString(homops[2]) + "\t" + toString(ptile25)); m->mothurOutEndLine();
m->mothurOut("Median: \t" + toString(starts[3]) + "\t" + toString(ends[3]) + "\t" + toString(lengths[3]) + "\t" + toString(ambigs[3]) + "\t" + toString(homops[3]) + "\t" + toString(ptile50)); m->mothurOutEndLine();
m->mothurOut("75%-tile:\t" + toString(starts[4]) + "\t" + toString(ends[4]) + "\t" + toString(lengths[4]) + "\t" + toString(ambigs[4]) + "\t" + toString(homops[4]) + "\t" + toString(ptile75)); m->mothurOutEndLine();
m->mothurOut("97.5%-tile:\t" + toString(starts[5]) + "\t" + toString(ends[5]) + "\t" + toString(lengths[5]) + "\t" + toString(ambigs[5]) + "\t" + toString(homops[5]) + "\t" + toString(ptile97_5)); m->mothurOutEndLine();
m->mothurOut("Maximum:\t" + toString(starts[6]) + "\t" + toString(ends[6]) + "\t" + toString(lengths[6]) + "\t" + toString(ambigs[6]) + "\t" + toString(homops[6]) + "\t" + toString(ptile100)); m->mothurOutEndLine();
m->mothurOut("Mean:\t" + toString(meanstartPosition) + "\t" + toString(meanendPosition) + "\t" + toString(meanseqLength) + "\t" + toString(meanambigBases) + "\t" + toString(meanlongHomoPolymer)); m->mothurOutEndLine();
if ((namefile == "") && (countfile == "")) { m->mothurOut("# of Seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); }
else { m->mothurOut("# of unique seqs:\t" + toString(numSeqs)); m->mothurOutEndLine(); m->mothurOut("total # of seqs:\t" + toString(size)); m->mothurOutEndLine(); }
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
m->mothurOut(summaryFile); m->mothurOutEndLine(); outputNames.push_back(summaryFile); outputTypes["summary"].push_back(summaryFile);
m->mothurOutEndLine();
if ((namefile == "") && (countfile == "")) { m->mothurOut("It took " + toString(time(NULL) - start) + " secs to summarize " + toString(numSeqs) + " sequences.\n"); }
else{ m->mothurOut("It took " + toString(time(NULL) - start) + " secs to summarize " + toString(size) + " sequences.\n"); }
//set fasta file as new current fastafile
string current = "";
itTypes = outputTypes.find("summary");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSummaryFile(current); }
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "SeqSummaryCommand", "execute");
exit(1);
}
}
int DeconvoluteCommand::execute() {
try {
if (abort) { if (calledHelp) { return 0; } return 2; }
//prepare filenames and open files
map<string, string> variables;
variables["[filename]"] = outputDir + util.getRootName(util.getSimpleName(fastafile));
string outNameFile = getOutputFileName("name", variables);
string outCountFile = getOutputFileName("count", variables);
variables["[extension]"] = util.getExtension(fastafile);
string outFastaFile = getOutputFileName("fasta", variables);
map<string, string> nameMap;
map<string, string>::iterator itNames;
if (namefile != "") {
util.readNames(namefile, nameMap);
if (namefile == outNameFile){
//prepare filenames and open files
map<string, string> mvariables;
mvariables["[filename]"] = outputDir + util.getRootName(util.getSimpleName(fastafile));
mvariables["[tag]"] = "unique";
outNameFile = getOutputFileName("name", mvariables);
}
}
CountTable ct;
if (countfile != "") {
ct.readTable(countfile, true, false);
if (countfile == outCountFile){
//prepare filenames and open files
map<string, string> mvariables;
mvariables["[filename]"] = outputDir + util.getRootName(util.getSimpleName(fastafile));
mvariables["[tag]"] = "unique";
outCountFile = getOutputFileName("count", mvariables); }
}
if (m->getControl_pressed()) { return 0; }
ifstream in;
util.openInputFile(fastafile, in);
ofstream outFasta;
util.openOutputFile(outFastaFile, outFasta);
map<string, string> sequenceStrings; //sequenceString -> list of names. "atgc...." -> seq1,seq2,seq3.
map<string, string>::iterator itStrings;
set<string> nameInFastaFile; //for sanity checking
set<string>::iterator itname;
vector<string> nameFileOrder;
CountTable newCt;
int count = 0;
while (!in.eof()) {
if (m->getControl_pressed()) { in.close(); outFasta.close(); util.mothurRemove(outFastaFile); return 0; }
Sequence seq(in);
if (seq.getName() != "") {
//sanity checks
itname = nameInFastaFile.find(seq.getName());
if (itname == nameInFastaFile.end()) { nameInFastaFile.insert(seq.getName()); }
else { m->mothurOut("[ERROR]: You already have a sequence named " + seq.getName() + " in your fasta file, sequence names must be unique, please correct."); m->mothurOutEndLine(); }
itStrings = sequenceStrings.find(seq.getAligned());
if (itStrings == sequenceStrings.end()) { //this is a new unique sequence
//output to unique fasta file
seq.printSequence(outFasta);
if (namefile != "") {
itNames = nameMap.find(seq.getName());
if (itNames == nameMap.end()) { //namefile and fastafile do not match
m->mothurOut("[ERROR]: " + seq.getName() + " is in your fasta file, and not in your namefile, please correct."); m->mothurOutEndLine();
}else {
if (format == "name") { sequenceStrings[seq.getAligned()] = itNames->second; nameFileOrder.push_back(seq.getAligned());
}else { newCt.push_back(seq.getName(), util.getNumNames(itNames->second)); sequenceStrings[seq.getAligned()] = seq.getName(); nameFileOrder.push_back(seq.getAligned()); }
}
}else if (countfile != "") {
if (format == "name") {
int numSeqs = ct.getNumSeqs(seq.getName());
string expandedName = seq.getName()+"_0";
for (int i = 1; i < numSeqs; i++) { expandedName += "," + seq.getName() + "_" + toString(i); }
sequenceStrings[seq.getAligned()] = expandedName; nameFileOrder.push_back(seq.getAligned());
}else {
ct.getNumSeqs(seq.getName()); //checks to make sure seq is in table
sequenceStrings[seq.getAligned()] = seq.getName(); nameFileOrder.push_back(seq.getAligned());
}
}else {
if (format == "name") { sequenceStrings[seq.getAligned()] = seq.getName(); nameFileOrder.push_back(seq.getAligned()); }
else { newCt.push_back(seq.getName()); sequenceStrings[seq.getAligned()] = seq.getName(); nameFileOrder.push_back(seq.getAligned()); }
}
}else { //this is a dup
if (namefile != "") {
itNames = nameMap.find(seq.getName());
if (itNames == nameMap.end()) { //namefile and fastafile do not match
m->mothurOut("[ERROR]: " + seq.getName() + " is in your fasta file, and not in your namefile, please correct."); m->mothurOutEndLine();
}else {
if (format == "name") { sequenceStrings[seq.getAligned()] += "," + itNames->second; }
else { int currentReps = newCt.getNumSeqs(itStrings->second); newCt.setNumSeqs(itStrings->second, currentReps+(util.getNumNames(itNames->second))); }
}
}else if (countfile != "") {
if (format == "name") {
int numSeqs = ct.getNumSeqs(seq.getName());
string expandedName = seq.getName()+"_0";
for (int i = 1; i < numSeqs; i++) { expandedName += "," + seq.getName() + "_" + toString(i); }
sequenceStrings[seq.getAligned()] += "," + expandedName;
}else {
int num = ct.getNumSeqs(seq.getName()); //checks to make sure seq is in table
if (num != 0) { //its in the table
ct.mergeCounts(itStrings->second, seq.getName()); //merges counts and saves in uniques name
}
}
}else {
if (format == "name") { sequenceStrings[seq.getAligned()] += "," + seq.getName(); }
else { int currentReps = newCt.getNumSeqs(itStrings->second); newCt.setNumSeqs(itStrings->second, currentReps+1); }
}
}
count++;
}
util.gobble(in);
if(count % 1000 == 0) { m->mothurOutJustToScreen(toString(count) + "\t" + toString(sequenceStrings.size()) + "\n"); }
}
if(count % 1000 != 0) { m->mothurOut(toString(count) + "\t" + toString(sequenceStrings.size())); m->mothurOutEndLine(); }
in.close();
outFasta.close();
if (m->getControl_pressed()) { util.mothurRemove(outFastaFile); return 0; }
//print new names file
ofstream outNames;
if (format == "name") { util.openOutputFile(outNameFile, outNames); outputNames.push_back(outNameFile); outputTypes["name"].push_back(outNameFile); }
else { util.openOutputFile(outCountFile, outNames); outputTypes["count"].push_back(outCountFile); outputNames.push_back(outCountFile); }
if ((countfile != "") && (format == "count")) { ct.printHeaders(outNames); }
else if ((countfile == "") && (format == "count")) { newCt.printHeaders(outNames); }
for (int i = 0; i < nameFileOrder.size(); i++) {
if (m->getControl_pressed()) { outputTypes.clear(); util.mothurRemove(outFastaFile); outNames.close(); for (int j = 0; j < outputNames.size(); j++) { util.mothurRemove(outputNames[j]); } return 0; }
itStrings = sequenceStrings.find(nameFileOrder[i]);
if (itStrings != sequenceStrings.end()) {
if (format == "name") {
//get rep name
int pos = (itStrings->second).find_first_of(',');
if (pos == string::npos) { // only reps itself
outNames << itStrings->second << '\t' << itStrings->second << endl;
}else {
outNames << (itStrings->second).substr(0, pos) << '\t' << itStrings->second << endl;
}
}else {
if (countfile != "") { ct.printSeq(outNames, itStrings->second); }
else if (format == "count") { newCt.printSeq(outNames, itStrings->second); }
}
}else{ m->mothurOut("[ERROR]: mismatch in namefile print."); m->mothurOutEndLine(); m->setControl_pressed(true); }
}
outNames.close();
if (m->getControl_pressed()) { outputTypes.clear(); util.mothurRemove(outFastaFile); for (int j = 0; j < outputNames.size(); j++) { util.mothurRemove(outputNames[j]); } return 0; }
m->mothurOut("\nOutput File Names: \n");
outputNames.push_back(outFastaFile); outputTypes["fasta"].push_back(outFastaFile);
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i] +"\n"); } m->mothurOutEndLine();
//set fasta file as new current fastafile
string currentName = "";
itTypes = outputTypes.find("fasta");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setFastaFile(currentName); }
}
itTypes = outputTypes.find("name");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setNameFile(currentName); }
}
itTypes = outputTypes.find("count");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setCountFile(currentName); }
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "DeconvoluteCommand", "execute");
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);
}
}
EstOutput Weighted::getValues(Tree* t, string groupA, string groupB) {
try {
data.clear(); //clear out old values
CountTable* ct = t->getCountTable();
if (m->control_pressed) { return data; }
//initialize weighted score
WScore[(groupA+groupB)] = 0.0;
double D = 0.0;
set<int> validBranches;
vector<string> groups; groups.push_back(groupA); groups.push_back(groupB);
//adding the wieghted sums from group i
for (int j = 0; j < t->groupNodeInfo[groups[0]].size(); j++) { //the leaf nodes that have seqs from group i
map<string, int>::iterator it = t->tree[t->groupNodeInfo[groups[0]][j]].pcount.find(groups[0]);
int numSeqsInGroupI = it->second;
double sum = getLengthToRoot(t, t->groupNodeInfo[groups[0]][j], groups[0], groups[1]);
double weightedSum = ((numSeqsInGroupI * sum) / (double)ct->getGroupCount(groups[0]));
D += weightedSum;
}
//adding the wieghted sums from group l
for (int j = 0; j < t->groupNodeInfo[groups[1]].size(); j++) { //the leaf nodes that have seqs from group l
map<string, int>::iterator it = t->tree[t->groupNodeInfo[groups[1]][j]].pcount.find(groups[1]);
int numSeqsInGroupL = it->second;
double sum = getLengthToRoot(t, t->groupNodeInfo[groups[1]][j], groups[0], groups[1]);
double weightedSum = ((numSeqsInGroupL * sum) / (double)ct->getGroupCount(groups[1]));
D += weightedSum;
}
//calculate u for the group comb
for(int i=0;i<t->getNumNodes();i++){
if (m->control_pressed) { return data; }
double u;
//int pcountSize = 0;
//does this node have descendants from groupA
it = t->tree[i].pcount.find(groupA);
//if it does u = # of its descendants with a certain group / total number in tree with a certain group
if (it != t->tree[i].pcount.end()) {
u = (double) t->tree[i].pcount[groupA] / (double) ct->getGroupCount(groupA);
}else { u = 0.00; }
//does this node have descendants from group l
it = t->tree[i].pcount.find(groupB);
//if it does subtract their percentage from u
if (it != t->tree[i].pcount.end()) {
u -= (double) t->tree[i].pcount[groupB] / (double) ct->getGroupCount(groupB);
}
if (includeRoot) {
if (t->tree[i].getBranchLength() != -1) {
u = abs(u * t->tree[i].getBranchLength());
WScore[(groupA+groupB)] += u;
}
}else{
//if this is not the root then add it
if (rootForGrouping[groups].count(i) == 0) {
if (t->tree[i].getBranchLength() != -1) {
u = abs(u * t->tree[i].getBranchLength());
WScore[(groupA+groupB)] += u;
}
}
}
}
/********************************************************/
//calculate weighted score for the group combination
double UN;
UN = (WScore[(groupA+groupB)] / D);
if (isnan(UN) || isinf(UN)) { UN = 0; }
data.push_back(UN);
return data;
}
catch(exception& e) {
m->errorOut(e, "Weighted", "getValues");
exit(1);
}
}
int ClusterCommand::execute(){
try {
if (abort == true) { if (calledHelp) { return 0; } return 2; }
//phylip file given and cutoff not given - use cluster.classic because it uses less memory and is faster
if ((format == "phylip") && (cutoff > 10.0)) {
m->mothurOutEndLine(); m->mothurOut("You are using a phylip file and no cutoff. I will run cluster.classic to save memory and time."); m->mothurOutEndLine();
//run unique.seqs for deconvolute results
string inputString = "phylip=" + distfile;
if (namefile != "") { inputString += ", name=" + namefile; }
else if (countfile != "") { inputString += ", count=" + countfile; }
inputString += ", precision=" + toString(precision);
inputString += ", method=" + method;
if (hard) { inputString += ", hard=T"; }
else { inputString += ", hard=F"; }
if (sim) { inputString += ", sim=T"; }
else { inputString += ", sim=F"; }
m->mothurOutEndLine();
m->mothurOut("/------------------------------------------------------------/"); m->mothurOutEndLine();
m->mothurOut("Running command: cluster.classic(" + inputString + ")"); m->mothurOutEndLine();
Command* clusterClassicCommand = new ClusterDoturCommand(inputString);
clusterClassicCommand->execute();
delete clusterClassicCommand;
m->mothurOut("/------------------------------------------------------------/"); m->mothurOutEndLine();
return 0;
}
ReadMatrix* read;
if (format == "column") { read = new ReadColumnMatrix(columnfile, sim); } //sim indicates whether its a similarity matrix
else if (format == "phylip") { read = new ReadPhylipMatrix(phylipfile, sim); }
read->setCutoff(cutoff);
NameAssignment* nameMap = NULL;
CountTable* ct = NULL;
map<string, int> counts;
if(namefile != ""){
nameMap = new NameAssignment(namefile);
nameMap->readMap();
read->read(nameMap);
}else if (countfile != "") {
ct = new CountTable();
ct->readTable(countfile, false, false);
read->read(ct);
counts = ct->getNameMap();
}else { read->read(nameMap); }
list = read->getListVector();
matrix = read->getDMatrix();
if(countfile != "") {
rabund = new RAbundVector();
createRabund(ct, list, rabund); //creates an rabund that includes the counts for the unique list
delete ct;
}else { rabund = new RAbundVector(list->getRAbundVector()); }
delete read;
if (m->control_pressed) { //clean up
delete list; delete matrix; delete rabund; if(countfile == ""){rabundFile.close(); sabundFile.close(); m->mothurRemove((fileroot+ tag + ".rabund")); m->mothurRemove((fileroot+ tag + ".sabund")); }
listFile.close(); m->mothurRemove((fileroot+ tag + ".list")); outputTypes.clear(); return 0;
}
//create cluster
if (method == "furthest") { cluster = new CompleteLinkage(rabund, list, matrix, cutoff, method, adjust); }
else if(method == "nearest"){ cluster = new SingleLinkage(rabund, list, matrix, cutoff, method, adjust); }
else if(method == "average"){ cluster = new AverageLinkage(rabund, list, matrix, cutoff, method, adjust); }
else if(method == "weighted"){ cluster = new WeightedLinkage(rabund, list, matrix, cutoff, method, adjust); }
tag = cluster->getTag();
if (outputDir == "") { outputDir += m->hasPath(distfile); }
fileroot = outputDir + m->getRootName(m->getSimpleName(distfile));
map<string, string> variables;
variables["[filename]"] = fileroot;
variables["[clustertag]"] = tag;
string sabundFileName = getOutputFileName("sabund", variables);
string rabundFileName = getOutputFileName("rabund", variables);
if (countfile != "") { variables["[tag2]"] = "unique_list"; }
string listFileName = getOutputFileName("list", variables);
if (countfile == "") {
m->openOutputFile(sabundFileName, sabundFile);
m->openOutputFile(rabundFileName, rabundFile);
outputNames.push_back(sabundFileName); outputTypes["sabund"].push_back(sabundFileName);
outputNames.push_back(rabundFileName); outputTypes["rabund"].push_back(rabundFileName);
}
m->openOutputFile(listFileName, listFile);
outputNames.push_back(listFileName); outputTypes["list"].push_back(listFileName);
list->printHeaders(listFile);
time_t estart = time(NULL);
float previousDist = 0.00000;
float rndPreviousDist = 0.00000;
oldRAbund = *rabund;
oldList = *list;
print_start = true;
start = time(NULL);
loops = 0;
double saveCutoff = cutoff;
while (matrix->getSmallDist() < cutoff && matrix->getNNodes() > 0){
if (m->control_pressed) { //clean up
delete list; delete matrix; delete rabund; delete cluster;
if(countfile == "") {rabundFile.close(); sabundFile.close(); m->mothurRemove((fileroot+ tag + ".rabund")); m->mothurRemove((fileroot+ tag + ".sabund")); }
listFile.close(); m->mothurRemove((fileroot+ tag + ".list")); outputTypes.clear(); return 0;
}
if (print_start && m->isTrue(timing)) {
m->mothurOut("Clustering (" + tag + ") dist " + toString(matrix->getSmallDist()) + "/"
+ toString(m->roundDist(matrix->getSmallDist(), precision))
+ "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")");
cout.flush();
print_start = false;
}
loops++;
cluster->update(cutoff);
float dist = matrix->getSmallDist();
float rndDist;
if (hard) {
rndDist = m->ceilDist(dist, precision);
}else{
rndDist = m->roundDist(dist, precision);
}
if(previousDist <= 0.0000 && dist != previousDist){
printData("unique", counts);
}
else if(rndDist != rndPreviousDist){
printData(toString(rndPreviousDist, length-1), counts);
}
previousDist = dist;
rndPreviousDist = rndDist;
oldRAbund = *rabund;
oldList = *list;
}
if (print_start && m->isTrue(timing)) {
m->mothurOut("Clustering (" + tag + ") for distance " + toString(previousDist) + "/" + toString(rndPreviousDist)
+ "\t(precision: " + toString(precision) + ", Nodes: " + toString(matrix->getNNodes()) + ")");
cout.flush();
print_start = false;
}
if(previousDist <= 0.0000){
printData("unique", counts);
}
else if(rndPreviousDist<cutoff){
printData(toString(rndPreviousDist, length-1), counts);
}
delete matrix;
delete list;
delete rabund;
delete cluster;
if (countfile == "") {
sabundFile.close();
rabundFile.close();
}
listFile.close();
if (saveCutoff != cutoff) {
if (hard) { saveCutoff = m->ceilDist(saveCutoff, precision); }
else { saveCutoff = m->roundDist(saveCutoff, precision); }
m->mothurOut("changed cutoff to " + toString(cutoff)); m->mothurOutEndLine();
}
//set list file as new current listfile
string current = "";
itTypes = outputTypes.find("list");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setListFile(current); }
}
//set rabund file as new current rabundfile
itTypes = outputTypes.find("rabund");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setRabundFile(current); }
}
//set sabund file as new current sabundfile
itTypes = outputTypes.find("sabund");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSabundFile(current); }
}
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();
//if (m->isTrue(timing)) {
m->mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster"); m->mothurOutEndLine();
//}
return 0;
}
catch(exception& e) {
m->errorOut(e, "ClusterCommand", "execute");
exit(1);
}
}
int ClassifySeqsCommand::execute(){
try {
if (abort) { if (calledHelp) { return 0; } return 2; }
string outputMethodTag = method;
if(method == "wang"){ classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, util.getRandomNumber(), flip, writeShortcuts, current->getVersion()); }
else if(method == "knn"){ classify = new Knn(taxonomyFileName, templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch, numWanted, util.getRandomNumber(), current->getVersion()); }
else if(method == "zap"){
outputMethodTag = search + "_" + outputMethodTag;
if (search == "kmer") { classify = new KmerTree(templateFileName, taxonomyFileName, kmerSize, cutoff); }
else { classify = new AlignTree(templateFileName, taxonomyFileName, cutoff); }
}
else {
m->mothurOut(search + " is not a valid method option. I will run the command using wang.");
m->mothurOutEndLine();
classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, util.getRandomNumber(), flip, writeShortcuts, current->getVersion());
}
if (m->getControl_pressed()) { delete classify; return 0; }
m->mothurOut("Classifying sequences from " + fastafile + " ...\n" );
string baseTName = util.getSimpleName(taxonomyFileName);
//set rippedTaxName to
string RippedTaxName = "";
bool foundDot = false;
for (int i = baseTName.length()-1; i >= 0; i--) {
if (foundDot && (baseTName[i] != '.')) { RippedTaxName = baseTName[i] + RippedTaxName; }
else if (foundDot && (baseTName[i] == '.')) { break; }
else if (!foundDot && (baseTName[i] == '.')) { foundDot = true; }
}
if (outputDir == "") { outputDir += util.hasPath(fastafile); }
map<string, string> variables;
variables["[filename]"] = outputDir + util.getRootName(util.getSimpleName(fastafile));
variables["[tag]"] = RippedTaxName;
variables["[tag2]"] = outputMethodTag;
string newTaxonomyFile = getOutputFileName("taxonomy", variables);
string newaccnosFile = getOutputFileName("accnos", variables);
string tempTaxonomyFile = outputDir + util.getRootName(util.getSimpleName(fastafile)) + "taxonomy.temp";
string taxSummary = getOutputFileName("taxsummary", variables);
if ((method == "knn") && (search == "distance")) {
string DistName = getOutputFileName("matchdist", variables);
classify->setDistName(DistName); outputNames.push_back(DistName); outputTypes["matchdist"].push_back(DistName);
}
outputNames.push_back(newTaxonomyFile); outputTypes["taxonomy"].push_back(newTaxonomyFile);
outputNames.push_back(taxSummary); outputTypes["taxsummary"].push_back(taxSummary);
long start = time(NULL);
int numFastaSeqs = createProcesses(newTaxonomyFile, tempTaxonomyFile, newaccnosFile, fastafile);
if (!util.isBlank(newaccnosFile)) { m->mothurOut("\n[WARNING]: mothur reversed some your sequences for a better classification. If you would like to take a closer look, please check " + newaccnosFile + " for the list of the sequences.\n");
outputNames.push_back(newaccnosFile); outputTypes["accnos"].push_back(newaccnosFile);
}else { util.mothurRemove(newaccnosFile); }
m->mothurOut("\nIt took " + toString(time(NULL) - start) + " secs to classify " + toString(numFastaSeqs) + " sequences.\n\n");
start = time(NULL);
//read namefile
map<string, vector<string> > nameMap;
map<string, vector<string> >::iterator itNames;
if(namefile != "") {
m->mothurOut("Reading " + namefile + "..."); cout.flush();
nameMap.clear(); //remove old names
util.readNames(namefile, nameMap);
m->mothurOut(" Done.\n");
}
//output taxonomy with the unclassified bins added
ifstream inTax;
util.openInputFile(newTaxonomyFile, inTax);
ofstream outTax;
string unclass = newTaxonomyFile + ".unclass.temp";
util.openOutputFile(unclass, outTax);
//get maxLevel from phylotree so you know how many 'unclassified's to add
int maxLevel = classify->getMaxLevel();
//read taxfile - this reading and rewriting is done to preserve the confidence scores.
string name, taxon;
GroupMap* groupMap = NULL;
CountTable* ct = NULL;
PhyloSummary* taxaSum;
if (hasCount) {
ct = new CountTable();
ct->readTable(countfile, true, false);
taxaSum = new PhyloSummary(ct, relabund, printlevel);
}else {
if (groupfile != "") { groupMap = new GroupMap(groupfile); groupMap->readMap(); }
taxaSum = new PhyloSummary(groupMap, relabund, printlevel);
}
while (!inTax.eof()) {
if (m->getControl_pressed()) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } delete taxaSum; for (int i = 0; i < outputNames.size(); i++) { util.mothurRemove(outputNames[i]); } delete classify; return 0; }
inTax >> name; util.gobble(inTax);
taxon = util.getline(inTax); util.gobble(inTax);
string newTax = util.addUnclassifieds(taxon, maxLevel, probs);
outTax << name << '\t' << newTax << endl;
if (namefile != "") {
itNames = nameMap.find(name);
if (itNames == nameMap.end()) {
m->mothurOut(name + " is not in your name file please correct.\n"); exit(1);
}else{
//add it as many times as there are identical seqs
for (int i = 0; i < itNames->second.size(); i++) { taxaSum->addSeqToTree(itNames->second[i], newTax); }
itNames->second.clear();
nameMap.erase(itNames->first);
}
}else { taxaSum->addSeqToTree(name, newTax); }
}
inTax.close();
outTax.close();
util.mothurRemove(newTaxonomyFile);
util.renameFile(unclass, newTaxonomyFile);
if (m->getControl_pressed()) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } for (int i = 0; i < outputNames.size(); i++) { util.mothurRemove(outputNames[i]); } delete classify; return 0; }
//print summary file
ofstream outTaxTree;
util.openOutputFile(taxSummary, outTaxTree);
taxaSum->print(outTaxTree, output);
outTaxTree.close();
if (ct != NULL) { delete ct; }
if (groupMap != NULL) { delete groupMap; } delete taxaSum;
util.mothurRemove(tempTaxonomyFile);
delete classify;
m->mothurOut("\nIt took " + toString(time(NULL) - start) + " secs to create the summary file for " + toString(numFastaSeqs) + " sequences.\n\n");
m->mothurOut("\nOutput File Names: \n");
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
m->mothurOutEndLine();
//set taxonomy file as new current taxonomyfile
string currentName = "";
itTypes = outputTypes.find("taxonomy");
if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setTaxonomyFile(currentName); } }
currentName = "";
itTypes = outputTypes.find("accnos");
if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setAccnosFile(currentName); } }
return 0;
}
catch(exception& e) {
m->errorOut(e, "ClassifySeqsCommand", "execute");
exit(1);
}
}
//**********************************************************************************************************************
int GetRAbundCommand::processList(ofstream& out){
try {
CountTable ct;
ct.readTable(countfile, false, false);
InputData input(inputfile, format);
ListVector* list = input.getListVector();
string lastLabel = list->getLabel();
//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;
if (m->control_pressed) { delete list; return 0; }
while((list != NULL) && ((allLines == 1) || (userLabels.size() != 0))) {
if(allLines == 1 || labels.count(list->getLabel()) == 1){
m->mothurOut(list->getLabel()); m->mothurOutEndLine();
if (m->control_pressed) { delete list; return 0; }
RAbundVector* rabund = new RAbundVector();
createRabund(ct, list, rabund);
if(sorted) { rabund->print(out); }
else { rabund->nonSortedPrint(out); }
delete rabund;
processedLabels.insert(list->getLabel());
userLabels.erase(list->getLabel());
}
if ((m->anyLabelsToProcess(list->getLabel(), userLabels, "") == true) && (processedLabels.count(lastLabel) != 1)) {
string saveLabel = list->getLabel();
delete list;
list = input.getListVector(lastLabel);
m->mothurOut(list->getLabel()); m->mothurOutEndLine();
if (m->control_pressed) { delete list; return 0; }
RAbundVector* rabund = new RAbundVector();
createRabund(ct, list, rabund);
if(sorted) { rabund->print(out); }
else { rabund->nonSortedPrint(out); }
delete rabund;
processedLabels.insert(list->getLabel());
userLabels.erase(list->getLabel());
//restore real lastlabel to save below
list->setLabel(saveLabel);
}
lastLabel = list->getLabel();
delete list;
list = input.getListVector();
}
//output error messages about any remaining user labels
set<string>::iterator it;
bool needToRun = false;
for (it = userLabels.begin(); it != userLabels.end(); it++) {
m->mothurOut("Your file does not include the label " + *it);
if (processedLabels.count(lastLabel) != 1) {
m->mothurOut(". I will use " + lastLabel + "."); m->mothurOutEndLine();
needToRun = true;
}else {
m->mothurOut(". Please refer to " + lastLabel + "."); m->mothurOutEndLine();
}
}
//run last label if you need to
if (needToRun == true) {
if (list != NULL) { delete list; }
list = input.getListVector(lastLabel);
m->mothurOut(list->getLabel()); m->mothurOutEndLine();
if (m->control_pressed) { delete list; return 0; }
RAbundVector* rabund = new RAbundVector();
createRabund(ct, list, rabund);
if(sorted) { rabund->print(out); }
else { rabund->nonSortedPrint(out); }
delete rabund;
delete list;
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "GetRAbundCommand", "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);
}
}
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 OptiBlastMatrix::readBlast(){
try {
Utils util;
map<string, long long> nameAssignment;
if (namefile != "") { util.readNames(namefile, nameAssignment); }
else if (countfile != "") {
CountTable ct; ct.readTable(countfile, false, true);
map<string, int> temp = ct.getNameMap();
for (map<string, int>::iterator it = temp.begin(); it!= temp.end(); it++) { nameAssignment[it->first] = it->second; }
}
else { readBlastNames(nameAssignment); }
int count = 0;
for (map<string, long long>::iterator it = nameAssignment.begin(); it!= nameAssignment.end(); it++) {
it->second = count; count++;
nameMap.push_back(it->first);
overlapNameMap.push_back(it->first);
}
m->mothurOut("Reading Blast File... "); cout.flush();
string firstName, secondName, eScore, currentRow; currentRow = "";
string repeatName = "";
float distance, thisoverlap, refScore;
float percentId;
float numBases, mismatch, gap, startQuery, endQuery, startRef, endRef, score, lengthThisSeq;
map<string, float> thisRowsBlastScores;
///////////////////// Read to eliminate singletons ///////////////////////
ifstream fileHandle;
util.openInputFile(distFile, fileHandle);
map<int, int> singletonIndexSwap;
map<int, int> blastSingletonIndexSwap;
vector<bool> singleton; singleton.resize(nameAssignment.size(), true);
vector<bool> overlapSingleton; overlapSingleton.resize(nameAssignment.size(), true);
vector< map<string,float> > dists; dists.resize(nameAssignment.size());
if (!fileHandle.eof()) {
//read in line from file
fileHandle >> firstName >> secondName >> percentId >> numBases >> mismatch >> gap >> startQuery >> endQuery >> startRef >> endRef >> eScore >> score;
util.gobble(fileHandle);
currentRow = firstName;
lengthThisSeq = numBases;
repeatName = firstName + secondName;
if (firstName == secondName) { refScore = score; }
else{
thisRowsBlastScores[secondName] = score;
//calc overlap score
thisoverlap = 1.0 - (percentId * (lengthThisSeq - startQuery) / endRef / 100.0 - penalty);
//if there is a valid overlap, add it
if ((startRef <= length) && ((endQuery+length) >= lengthThisSeq) && (thisoverlap <= cutoff)) {
//convert name to number
map<string,long long>::iterator itA = nameAssignment.find(firstName);
map<string,long long>::iterator itB = nameAssignment.find(secondName);
if(itA == nameAssignment.end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); }
if(itB == nameAssignment.end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); }
int indexA = (itA->second);
int indexB = (itB->second);
overlapSingleton[indexA] = false;
overlapSingleton[indexB] = false;
blastSingletonIndexSwap[indexA] = indexA;
blastSingletonIndexSwap[indexB] = indexB;
}
}
}else { m->mothurOut("Error in your blast file, cannot read."); m->mothurOutEndLine(); exit(1); }
while(fileHandle){ //let's assume it's a triangular matrix...
if (m->getControl_pressed()) { fileHandle.close(); return 0; }
//read in line from file
fileHandle >> firstName >> secondName >> percentId >> numBases >> mismatch >> gap >> startQuery >> endQuery >> startRef >> endRef >> eScore >> score;
util.gobble(fileHandle);
string temp = firstName + secondName; //to check if this file has repeat lines, ie. is this a blast instead of a blscreen file
//if this is a new pairing
if (temp != repeatName) {
repeatName = temp;
if (currentRow == firstName) {
if (firstName == secondName) { refScore = score; }
else{
//save score
thisRowsBlastScores[secondName] = score;
//calc overlap score
thisoverlap = 1.0 - (percentId * (lengthThisSeq - startQuery) / endRef / 100.0 - penalty);
//if there is a valid overlap, add it
if ((startRef <= length) && ((endQuery+length) >= lengthThisSeq) && (thisoverlap <= cutoff)) {
//convert name to number
map<string,long long>::iterator itA = nameAssignment.find(firstName);
map<string,long long>::iterator itB = nameAssignment.find(secondName);
if(itA == nameAssignment.end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); }
if(itB == nameAssignment.end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); }
int indexA = (itA->second);
int indexB = (itB->second);
overlapSingleton[indexA] = false;
overlapSingleton[indexB] = false;
blastSingletonIndexSwap[indexA] = indexA;
blastSingletonIndexSwap[indexB] = indexB;
}
} //end else
}else { //end row
//convert blast scores to distance and add cell to sparse matrix if we can
map<string, float>::iterator it;
map<string, float>::iterator itDist;
for(it=thisRowsBlastScores.begin(); it!=thisRowsBlastScores.end(); it++) {
distance = 1.0 - (it->second / refScore);
//do we already have the distance calculated for b->a
map<string,long long>::iterator itA = nameAssignment.find(currentRow);
map<string,long long>::iterator itB = nameAssignment.find(it->first);
itDist = dists[itB->second].find(itA->first);
//if we have it then compare
if (itDist != dists[itB->second].end()) {
//if you want the minimum blast score ratio, then pick max distance
if(minWanted) { distance = max(itDist->second, distance); }
else{ distance = min(itDist->second, distance); }
//is this distance below cutoff
if (distance <= cutoff) {
int indexA = (itA->second);
int indexB = (itB->second);
singleton[indexA] = false;
singleton[indexB] = false;
singletonIndexSwap[indexA] = indexA;
singletonIndexSwap[indexB] = indexB;
}
//not going to need this again
dists[itB->second].erase(itDist);
}else { //save this value until we get the other ratio
dists[itA->second][it->first] = distance;
}
}
//clear out last rows info
thisRowsBlastScores.clear();
currentRow = firstName;
lengthThisSeq = numBases;
//add this row to thisRowsBlastScores
if (firstName == secondName) { refScore = score; }
else{ //add this row to thisRowsBlastScores
thisRowsBlastScores[secondName] = score;
//calc overlap score
thisoverlap = 1.0 - (percentId * (lengthThisSeq - startQuery) / endRef / 100.0 - penalty);
//if there is a valid overlap, add it
if ((startRef <= length) && ((endQuery+length) >= lengthThisSeq) && (thisoverlap <= cutoff)) {
//convert name to number
map<string,long long>::iterator itA = nameAssignment.find(firstName);
map<string,long long>::iterator itB = nameAssignment.find(secondName);
if(itA == nameAssignment.end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); }
if(itB == nameAssignment.end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); }
int indexA = (itA->second);
int indexB = (itB->second);
overlapSingleton[indexA] = false;
overlapSingleton[indexB] = false;
blastSingletonIndexSwap[indexA] = indexA;
blastSingletonIndexSwap[indexB] = indexB;
}
}
}//end if current row
}//end if repeat
}
fileHandle.close();
//convert blast scores to distance and add cell to sparse matrix if we can
map<string, float>::iterator it;
map<string, float>::iterator itDist;
for(it=thisRowsBlastScores.begin(); it!=thisRowsBlastScores.end(); it++) {
distance = 1.0 - (it->second / refScore);
//do we already have the distance calculated for b->a
map<string,long long>::iterator itA = nameAssignment.find(currentRow);
map<string,long long>::iterator itB = nameAssignment.find(it->first);
itDist = dists[itB->second].find(itA->first);
//if we have it then compare
if (itDist != dists[itB->second].end()) {
//if you want the minimum blast score ratio, then pick max distance
if(minWanted) { distance = max(itDist->second, distance); }
else{ distance = min(itDist->second, distance); }
//is this distance below cutoff
if (distance <= cutoff) {
int indexA = (itA->second);
int indexB = (itB->second);
singleton[indexA] = false;
singleton[indexB] = false;
singletonIndexSwap[indexA] = indexA;
singletonIndexSwap[indexB] = indexB;
}
//not going to need this again
dists[itB->second].erase(itDist);
}else { //save this value until we get the other ratio
dists[itA->second][it->first] = distance;
}
}
//clear out info
thisRowsBlastScores.clear();
dists.clear();
//////////////////////////////////////////////////////////////////////////
int nonSingletonCount = 0;
for (int i = 0; i < singleton.size(); i++) {
if (!singleton[i]) { //if you are a singleton
singletonIndexSwap[i] = nonSingletonCount;
nonSingletonCount++;
}else { singletons.push_back(nameMap[i]); }
}
singleton.clear();
int overlapNonSingletonCount = 0;
for (int i = 0; i < overlapSingleton.size(); i++) {
if (!overlapSingleton[i]) { //if you are a singleton
blastSingletonIndexSwap[i] = overlapNonSingletonCount;
overlapNonSingletonCount++;
}
}
overlapSingleton.clear();
ifstream in;
util.openInputFile(distFile, in);
dists.resize(nameAssignment.size());
closeness.resize(nonSingletonCount);
blastOverlap.resize(overlapNonSingletonCount);
map<string, string> names;
if (namefile != "") {
util.readNames(namefile, names);
for (int i = 0; i < singletons.size(); i++) {
singletons[i] = names[singletons[i]];
}
}
m->mothurOut(" halfway ... "); cout.flush();
if (!in.eof()) {
//read in line from file
in >> firstName >> secondName >> percentId >> numBases >> mismatch >> gap >> startQuery >> endQuery >> startRef >> endRef >> eScore >> score;
util.gobble(fileHandle);
currentRow = firstName;
lengthThisSeq = numBases;
repeatName = firstName + secondName;
if (firstName == secondName) { refScore = score; }
else{
//convert name to number
map<string,long long>::iterator itA = nameAssignment.find(firstName);
map<string,long long>::iterator itB = nameAssignment.find(secondName);
if(itA == nameAssignment.end()){ m->mothurOut("AAError: Sequence '" + firstName + "' was not found in the names file, please correct\n"); exit(1); }
if(itB == nameAssignment.end()){ m->mothurOut("ABError: Sequence '" + secondName + "' was not found in the names file, please correct\n"); exit(1); }
thisRowsBlastScores[secondName] = score;
if (namefile != "") {
firstName = names[firstName]; //redundant names
secondName = names[secondName]; //redundant names
}
nameMap[singletonIndexSwap[itA->second]] = firstName;
nameMap[singletonIndexSwap[itB->second]] = secondName;
//calc overlap score
thisoverlap = 1.0 - (percentId * (lengthThisSeq - startQuery) / endRef / 100.0 - penalty);
//if there is a valid overlap, add it
if ((startRef <= length) && ((endQuery+length) >= lengthThisSeq) && (thisoverlap <= cutoff)) {
int indexA = (itA->second);
int indexB = (itB->second);
int newB = blastSingletonIndexSwap[indexB];
int newA = blastSingletonIndexSwap[indexA];
blastOverlap[newA].insert(newB);
blastOverlap[newB].insert(newA);
overlapNameMap[newA] = firstName;
overlapNameMap[newB] = secondName;
}
}
}else { m->mothurOut("Error in your blast file, cannot read."); m->mothurOutEndLine(); 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 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 ClusterDoturCommand::execute(){
try {
if (abort) { if (calledHelp) { return 0; } return 2; }
ClusterClassic* cluster = new ClusterClassic(cutoff, method, sim);
NameAssignment* nameMap = NULL;
CountTable* ct = NULL;
map<string, int> counts;
if(namefile != "") {
nameMap = new NameAssignment(namefile);
nameMap->readMap();
cluster->readPhylipFile(phylipfile, nameMap);
delete nameMap;
}else if (countfile != "") {
ct = new CountTable();
ct->readTable(countfile, false, false);
cluster->readPhylipFile(phylipfile, ct);
counts = ct->getNameMap();
delete ct;
}else {
cluster->readPhylipFile(phylipfile, nameMap);
}
tag = cluster->getTag();
if (m->getControl_pressed()) { delete cluster; return 0; }
list = cluster->getListVector();
rabund = cluster->getRAbundVector();
if (outputDir == "") { outputDir += util.hasPath(phylipfile); }
fileroot = outputDir + util.getRootName(util.getSimpleName(phylipfile));
map<string, string> variables;
variables["[filename]"] = fileroot;
variables["[clustertag]"] = tag;
string sabundFileName = getOutputFileName("sabund", variables);
string rabundFileName = getOutputFileName("rabund", variables);
//if (countfile != "") { variables["[tag2]"] = "unique_list"; }
string listFileName = getOutputFileName("list", variables);
if (countfile == "") {
util.openOutputFile(sabundFileName, sabundFile);
util.openOutputFile(rabundFileName, rabundFile);
outputNames.push_back(sabundFileName); outputTypes["sabund"].push_back(sabundFileName);
outputNames.push_back(rabundFileName); outputTypes["rabund"].push_back(rabundFileName);
}
util.openOutputFile(listFileName, listFile);
outputNames.push_back(listFileName); outputTypes["list"].push_back(listFileName);
float previousDist = 0.00000;
float rndPreviousDist = 0.00000;
oldRAbund = *rabund;
oldList = *list;
bool printHeaders = true;
int estart = time(NULL); int loop = 0;
while ((cluster->getSmallDist() <= cutoff) && (cluster->getNSeqs() > 1)){
if (m->getControl_pressed()) { delete cluster; delete list; delete rabund; if(countfile == "") {rabundFile.close(); sabundFile.close(); util.mothurRemove((fileroot+ tag + ".rabund")); util.mothurRemove((fileroot+ tag + ".sabund")); }
listFile.close(); util.mothurRemove((fileroot+ tag + ".list")); outputTypes.clear(); return 0; }
cluster->update(cutoff);
float dist = cluster->getSmallDist();
float rndDist = util.ceilDist(dist, precision);
//cout << loop << '\t' << dist << '\t' << oldList.getNumBins() << endl; loop++;
if(previousDist <= 0.0000 && dist != previousDist) { printData("unique", counts, printHeaders); }
else if(rndDist != rndPreviousDist) { printData(toString(rndPreviousDist, length-1), counts, printHeaders); }
previousDist = dist;
rndPreviousDist = rndDist;
oldRAbund = *rabund;
oldList = *list;
}
if(previousDist <= 0.0000) { printData("unique", counts, printHeaders); }
else if(rndPreviousDist<cutoff) { printData(toString(rndPreviousDist, length-1), counts, printHeaders); }
if (countfile == "") {
sabundFile.close();
rabundFile.close();
}
listFile.close();
delete cluster; delete list; delete rabund;
//set list file as new current listfile
string currentName = "";
itTypes = outputTypes.find("list");
if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setListFile(currentName); } }
//set rabund file as new current rabundfile
itTypes = outputTypes.find("rabund");
if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setRabundFile(currentName); } }
//set sabund file as new current sabundfile
itTypes = outputTypes.find("sabund");
if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { currentName = (itTypes->second)[0]; current->setSabundFile(currentName); } }
m->mothurOut("\nOutput File Names: \n");
for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i] +"\n"); } m->mothurOutEndLine();
m->mothurOut("It took " + toString(time(NULL) - estart) + " seconds to cluster"); m->mothurOutEndLine();
return 0;
}
catch(exception& e) {
m->errorOut(e, "ClusterDoturCommand", "execute");
exit(1);
}
}
//**********************************************************************************************************************
SharedCommand::SharedCommand(string option) {
try {
abort = false; calledHelp = false; pickedGroups=false;
allLines = 1;
//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();
OptionParser parser(option);
map<string, string> parameters = parser.getParameters();
ValidParameters validParameter;
map<string, string>::iterator it;
//check to make sure all parameters are valid for command
for (it = parameters.begin(); it != parameters.end(); it++) {
if (!validParameter.isValidParameter(it->first, myArray, it->second)) { abort = true; }
}
//if the user changes the input directory command factory will send this info to us in the output parameter
string inputDir = validParameter.valid(parameters, "inputdir");
if (inputDir == "not found"){ inputDir = ""; }
else {
string path;
it = parameters.find("list");
//user has given a template file
if(it != parameters.end()){
path = util.hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["list"] = inputDir + it->second; }
}
it = parameters.find("group");
//user has given a template file
if(it != parameters.end()){
path = util.hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["group"] = inputDir + it->second; }
}
it = parameters.find("count");
//user has given a template file
if(it != parameters.end()){
path = util.hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["count"] = inputDir + it->second; }
}
it = parameters.find("biom");
//user has given a template file
if(it != parameters.end()){
path = util.hasPath(it->second);
//if the user has not given a path then, add inputdir. else leave path alone.
if (path == "") { parameters["biom"] = inputDir + it->second; }
}
}
vector<string> tempOutNames;
outputTypes["shared"] = tempOutNames;
outputTypes["group"] = tempOutNames;
outputTypes["map"] = tempOutNames;
//if the user changes the output directory command factory will send this info to us in the output parameter
outputDir = validParameter.valid(parameters, "outputdir"); if (outputDir == "not found"){ outputDir = ""; }
//check for required parameters
listfile = validParameter.validFile(parameters, "list");
if (listfile == "not open") { listfile = ""; abort = true; }
else if (listfile == "not found") { listfile = ""; }
else { current->setListFile(listfile); }
biomfile = validParameter.validFile(parameters, "biom");
if (biomfile == "not open") { biomfile = ""; abort = true; }
else if (biomfile == "not found") { biomfile = ""; }
else { current->setBiomFile(biomfile); }
ordergroupfile = validParameter.validFile(parameters, "ordergroup");
if (ordergroupfile == "not open") { abort = true; }
else if (ordergroupfile == "not found") { ordergroupfile = ""; }
groupfile = validParameter.validFile(parameters, "group");
if (groupfile == "not open") { groupfile = ""; abort = true; }
else if (groupfile == "not found") { groupfile = ""; }
else { current->setGroupFile(groupfile); }
countfile = validParameter.validFile(parameters, "count");
if (countfile == "not open") { countfile = ""; abort = true; }
else if (countfile == "not found") { countfile = ""; }
else {
current->setCountFile(countfile);
CountTable temp;
if (!temp.testGroups(countfile)) {
m->mothurOut("\n[WARNING]: Your count file does not have group info, all reads will be assigned to mothurGroup.\n");
temp.readTable(countfile, false, false); //dont read groups
map<string, int> seqs = temp.getNameMap();
CountTable newCountTable;
newCountTable.addGroup("mothurGroup");
for (map<string, int>::iterator it = seqs.begin(); it != seqs.end(); it++) {
vector<int> counts; counts.push_back(it->second);
newCountTable.push_back(it->first, counts);
}
string newCountfileName = util.getRootName(countfile) + "mothurGroup" + util.getExtension(countfile);
newCountTable.printTable(newCountfileName);
current->setCountFile(newCountfileName);
countfile = newCountfileName;
outputNames.push_back(newCountfileName);
}
}
if ((biomfile == "") && (listfile == "") && (countfile == "")) { //you must provide at least one of the following
//is there are current file available for either of these?
//give priority to list, then biom, then count
listfile = current->getListFile();
if (listfile != "") { m->mothurOut("Using " + listfile + " as input file for the list parameter.\n"); }
else {
biomfile = current->getBiomFile();
if (biomfile != "") { m->mothurOut("Using " + biomfile + " as input file for the biom parameter.\n"); }
else {
countfile = current->getCountFile();
if (countfile != "") { m->mothurOut("Using " + countfile + " as input file for the count parameter.\n"); }
else {
m->mothurOut("[ERROR]: No valid current files. You must provide a list or biom or count file before you can use the make.shared command.\n"); abort = true;
}
}
}
}
else if ((biomfile != "") && (listfile != "")) { m->mothurOut("When executing a make.shared command you must enter ONLY ONE of the following: list or biom.\n"); abort = true; }
if (listfile != "") {
if ((groupfile == "") && (countfile == "")) {
groupfile = current->getGroupFile();
if (groupfile != "") { m->mothurOut("Using " + groupfile + " as input file for the group parameter.\n"); }
else {
countfile = current->getCountFile();
if (countfile != "") { m->mothurOut("Using " + countfile + " as input file for the count parameter.\n"); }
else { m->mothurOut("[ERROR]: You need to provide a groupfile or countfile if you are going to use the list format.\n"); abort = true; }
}
}
}
string groups = validParameter.valid(parameters, "groups");
if (groups == "not found") { groups = ""; }
else {
pickedGroups=true;
util.splitAtDash(groups, Groups);
if (Groups.size() != 0) { if (Groups[0]== "all") { Groups.clear(); } }
}
//check for optional parameter and set defaults
// ...at some point should added some additional type checking...
string label = validParameter.valid(parameters, "label");
if (label == "not found") { label = ""; }
else {
if(label != "all") { util.splitAtDash(label, labels); allLines = 0; }
else { allLines = 1; }
}
if ((listfile == "") && (biomfile == "") && (countfile != "")) { //building a shared file from a count file, require label
if (labels.size() == 0) {
m->mothurOut("[ERROR]: You must provide a label when converting a count file to a shared file, please correct.\n"); abort = true;
}
}
}
}
catch(exception& e) {
m->errorOut(e, "SharedCommand", "SharedCommand");
exit(1);
}
}
//***************************************************************************************************************
int SummaryQualCommand::execute(){
try{
if (abort == true) { if (calledHelp) { return 0; } return 2; }
int start = time(NULL);
int numSeqs = 0;
vector<int> position;
vector<int> averageQ;
vector< vector<int> > scores;
if (m->control_pressed) { return 0; }
if (namefile != "") { nameMap = m->readNames(namefile); }
else if (countfile != "") {
CountTable ct;
ct.readTable(countfile, false, false);
nameMap = ct.getNameMap();
}
vector<unsigned long long> positions;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
positions = m->divideFile(qualfile, processors);
for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(linePair(positions[i], positions[(i+1)])); }
#else
if (processors == 1) {
lines.push_back(linePair(0, 1000));
}else {
positions = m->setFilePosFasta(qualfile, numSeqs);
if (numSeqs < processors) { processors = numSeqs; }
//figure out how many sequences you have to process
int numSeqsPerProcessor = numSeqs / processors;
for (int i = 0; i < processors; i++) {
int startIndex = i * numSeqsPerProcessor;
if(i == (processors - 1)){ numSeqsPerProcessor = numSeqs - i * numSeqsPerProcessor; }
lines.push_back(linePair(positions[startIndex], numSeqsPerProcessor));
}
}
#endif
if(processors == 1){ numSeqs = driverCreateSummary(position, averageQ, scores, qualfile, lines[0]); }
else{ numSeqs = createProcessesCreateSummary(position, averageQ, scores, qualfile); }
if (m->control_pressed) { return 0; }
//print summary file
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(qualfile));
string summaryFile = getOutputFileName("summary",variables);
printQual(summaryFile, position, averageQ, scores);
if (m->control_pressed) { m->mothurRemove(summaryFile); return 0; }
//output results to screen
cout.setf(ios::fixed, ios::floatfield); cout.setf(ios::showpoint);
m->mothurOutEndLine();
m->mothurOut("Position\tNumSeqs\tAverageQ"); m->mothurOutEndLine();
for (int i = 0; i < position.size(); i+=100) {
float average = averageQ[i] / (float) position[i];
cout << i << '\t' << position[i] << '\t' << average;
m->mothurOutJustToLog(toString(i) + "\t" + toString(position[i]) + "\t" + toString(average)); m->mothurOutEndLine();
}
m->mothurOutEndLine();
m->mothurOut("It took " + toString(time(NULL) - start) + " secs to create the summary file for " + toString(numSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
m->mothurOutEndLine();
m->mothurOut("Output File Names: "); m->mothurOutEndLine();
m->mothurOut(summaryFile); m->mothurOutEndLine(); outputNames.push_back(summaryFile); outputTypes["summary"].push_back(summaryFile);
m->mothurOutEndLine();
return 0;
}
catch(exception& e) {
m->errorOut(e, "SummaryQualCommand", "execute");
exit(1);
}
}
int ClassifySeqsCommand::execute(){
try {
if (abort == true) { if (calledHelp) { return 0; } return 2; }
string outputMethodTag = method;
if(method == "wang"){ classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip, writeShortcuts); }
else if(method == "knn"){ classify = new Knn(taxonomyFileName, templateFileName, search, kmerSize, gapOpen, gapExtend, match, misMatch, numWanted, rand()); }
else if(method == "zap"){
outputMethodTag = search + "_" + outputMethodTag;
if (search == "kmer") { classify = new KmerTree(templateFileName, taxonomyFileName, kmerSize, cutoff); }
else { classify = new AlignTree(templateFileName, taxonomyFileName, cutoff); }
}
else {
m->mothurOut(search + " is not a valid method option. I will run the command using wang.");
m->mothurOutEndLine();
classify = new Bayesian(taxonomyFileName, templateFileName, search, kmerSize, cutoff, iters, rand(), flip, writeShortcuts);
}
if (m->control_pressed) { delete classify; return 0; }
for (int s = 0; s < fastaFileNames.size(); s++) {
m->mothurOut("Classifying sequences from " + fastaFileNames[s] + " ..." ); m->mothurOutEndLine();
string baseTName = m->getSimpleName(taxonomyFileName);
//set rippedTaxName to
string RippedTaxName = "";
bool foundDot = false;
for (int i = baseTName.length()-1; i >= 0; i--) {
if (foundDot && (baseTName[i] != '.')) { RippedTaxName = baseTName[i] + RippedTaxName; }
else if (foundDot && (baseTName[i] == '.')) { break; }
else if (!foundDot && (baseTName[i] == '.')) { foundDot = true; }
}
//if (RippedTaxName != "") { RippedTaxName += "."; }
if (outputDir == "") { outputDir += m->hasPath(fastaFileNames[s]); }
map<string, string> variables;
variables["[filename]"] = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s]));
variables["[tag]"] = RippedTaxName;
variables["[tag2]"] = outputMethodTag;
string newTaxonomyFile = getOutputFileName("taxonomy", variables);
string newaccnosFile = getOutputFileName("accnos", variables);
string tempTaxonomyFile = outputDir + m->getRootName(m->getSimpleName(fastaFileNames[s])) + "taxonomy.temp";
string taxSummary = getOutputFileName("taxsummary", variables);
if ((method == "knn") && (search == "distance")) {
string DistName = getOutputFileName("matchdist", variables);
classify->setDistName(DistName); outputNames.push_back(DistName); outputTypes["matchdist"].push_back(DistName);
}
outputNames.push_back(newTaxonomyFile); outputTypes["taxonomy"].push_back(newTaxonomyFile);
outputNames.push_back(taxSummary); outputTypes["taxsummary"].push_back(taxSummary);
int start = time(NULL);
int numFastaSeqs = 0;
for (int i = 0; i < lines.size(); i++) { delete lines[i]; } lines.clear();
vector<unsigned long long> positions;
#if defined (__APPLE__) || (__MACH__) || (linux) || (__linux) || (__linux__) || (__unix__) || (__unix)
positions = m->divideFile(fastaFileNames[s], processors);
for (int i = 0; i < (positions.size()-1); i++) { lines.push_back(new linePair(positions[i], positions[(i+1)])); }
#else
if (processors == 1) {
lines.push_back(new linePair(0, 1000));
}else {
positions = m->setFilePosFasta(fastaFileNames[s], numFastaSeqs);
if (numFastaSeqs < processors) { processors = numFastaSeqs; }
//figure out how many sequences you have to process
int numSeqsPerProcessor = numFastaSeqs / processors;
for (int i = 0; i < processors; i++) {
int startIndex = i * numSeqsPerProcessor;
if(i == (processors - 1)){ numSeqsPerProcessor = numFastaSeqs - i * numSeqsPerProcessor; }
lines.push_back(new linePair(positions[startIndex], numSeqsPerProcessor));
}
}
#endif
if(processors == 1){
numFastaSeqs = driver(lines[0], newTaxonomyFile, tempTaxonomyFile, newaccnosFile, fastaFileNames[s]);
}else{
numFastaSeqs = createProcesses(newTaxonomyFile, tempTaxonomyFile, newaccnosFile, fastaFileNames[s]);
}
if (!m->isBlank(newaccnosFile)) { m->mothurOutEndLine(); m->mothurOut("[WARNING]: mothur reversed some your sequences for a better classification. If you would like to take a closer look, please check " + newaccnosFile + " for the list of the sequences."); m->mothurOutEndLine();
outputNames.push_back(newaccnosFile); outputTypes["accnos"].push_back(newaccnosFile);
}else { m->mothurRemove(newaccnosFile); }
m->mothurOutEndLine();
m->mothurOut("It took " + toString(time(NULL) - start) + " secs to classify " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
start = time(NULL);
//read namefile
if(namefile != "") {
m->mothurOut("Reading " + namefileNames[s] + "..."); cout.flush();
nameMap.clear(); //remove old names
m->readNames(namefileNames[s], nameMap);
m->mothurOut(" Done."); m->mothurOutEndLine();
}
//output taxonomy with the unclassified bins added
ifstream inTax;
m->openInputFile(newTaxonomyFile, inTax);
ofstream outTax;
string unclass = newTaxonomyFile + ".unclass.temp";
m->openOutputFile(unclass, outTax);
//get maxLevel from phylotree so you know how many 'unclassified's to add
int maxLevel = classify->getMaxLevel();
//read taxfile - this reading and rewriting is done to preserve the confidence scores.
string name, taxon;
string group = "";
GroupMap* groupMap = NULL;
CountTable* ct = NULL;
PhyloSummary* taxaSum;
if (hasCount) {
ct = new CountTable();
ct->readTable(countfileNames[s], true, false);
taxaSum = new PhyloSummary(ct, relabund, printlevel);
}else {
if (groupfile != "") { group = groupfileNames[s]; groupMap = new GroupMap(group); groupMap->readMap(); }
taxaSum = new PhyloSummary(groupMap, relabund, printlevel);
}
while (!inTax.eof()) {
if (m->control_pressed) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } delete taxaSum; for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; }
inTax >> name >> taxon; m->gobble(inTax);
string newTax = m->addUnclassifieds(taxon, maxLevel, probs);
outTax << name << '\t' << newTax << endl;
if (namefile != "") {
itNames = nameMap.find(name);
if (itNames == nameMap.end()) {
m->mothurOut(name + " is not in your name file please correct."); m->mothurOutEndLine(); exit(1);
}else{
for (int i = 0; i < itNames->second.size(); i++) {
taxaSum->addSeqToTree(itNames->second[i], newTax); //add it as many times as there are identical seqs
}
itNames->second.clear();
nameMap.erase(itNames->first);
}
}else {
taxaSum->addSeqToTree(name, newTax);
}
}
inTax.close();
outTax.close();
m->mothurRemove(newTaxonomyFile);
rename(unclass.c_str(), newTaxonomyFile.c_str());
if (m->control_pressed) { outputTypes.clear(); if (ct != NULL) { delete ct; } if (groupMap != NULL) { delete groupMap; } for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } delete classify; return 0; }
//print summary file
ofstream outTaxTree;
m->openOutputFile(taxSummary, outTaxTree);
taxaSum->print(outTaxTree, output);
outTaxTree.close();
if (ct != NULL) { delete ct; }
if (groupMap != NULL) { delete groupMap; } delete taxaSum;
m->mothurRemove(tempTaxonomyFile);
m->mothurOutEndLine();
m->mothurOut("It took " + toString(time(NULL) - start) + " secs to create the summary file for " + toString(numFastaSeqs) + " sequences."); m->mothurOutEndLine(); m->mothurOutEndLine();
}
delete classify;
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 taxonomy file as new current taxonomyfile
string current = "";
itTypes = outputTypes.find("taxonomy");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setTaxonomyFile(current); }
}
current = "";
itTypes = outputTypes.find("accnos");
if (itTypes != outputTypes.end()) {
if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setAccnosFile(current); }
}
return 0;
}
catch(exception& e) {
m->errorOut(e, "ClassifySeqsCommand", "execute");
exit(1);
}
}