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 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);
}
}