void alignSeqs(ProbabilisticModelPtr model, SequenceList seqs1, SequenceList seqs2, vector<std::string> names1, vector<std::string> names2, ostream & output, ostream & states_output)
{
if(seqs1.size() != seqs2.size()) {
cerr << "Both files must to have the same amount of sequences." << endl;
exit(-1);
}
for(int i = 0; i < (int)seqs1.size(); i++)
{
Sequence s1,s2,path;
std::stringstream aux1,aux2,aux3;
vector<Matrix> a;
model->pairDecodable()->viterbi(seqs1[i], seqs2[i], path, s1, s2, a);
aux1 << names1[i] << "_" << names2[i] << "_al1" ;
aux2 << names1[i] << "_" << names2[i] << "_al2" ;
aux3 << names1[i] << "_" << names2[i] << "_path" ;
SequenceEntry out1(model->alphabet());
SequenceEntry out2(model->alphabet());
SequenceEntry out3(model->pairDecodable()->getStateNames());
out1.setName(aux1.str());
out2.setName(aux2.str());
out3.setName(aux3.str());
out1.setSequence(s1);
out2.setSequence(s2);
out3.setSequence(path);
output << out1;
output << out2 << endl;
states_output << out3 << endl;
}
}
int main(int argc, char const *argv[])
{
SequenceList<int> iseq;
int *p = iseq.begin();
iseq.insert(p, 1);
std::cout << iseq[0] << "\n";
p = iseq.begin();
iseq.insert(p + 1, 2);
std::cout << iseq[1] << "\n";
iseq.erase(iseq.begin());
std::cout << iseq[0];
}
void SequenceSpace::nextModificationSymbol(set<string>& mod_set, set<string>::iterator iter, SequenceList& seq_list, ModificationSequence mod_seq, Modifier mod_manager)
{
// Get currently available modification sites.
ModificationSites current_sites = mod_manager.getModificationSitesBySymbol(*iter, 1);
size_t mod_num = (size_t)gs->getModificationConstraint(*iter);
//ModificationSites mod_sites;
size_t i = 0;
// Non-const version of the container.
std::vector<ModificationPosition> current_vec(current_sites.begin(), current_sites.end());
std::vector<ModificationPosition> mod_vec;
for(ModificationSites::iterator mod_iter = current_sites.begin(); mod_iter != current_sites.end(); mod_iter++)
{
mod_vec.push_back(*mod_iter);
if(++i >= mod_num)
break;
}
//ModificationSites::iterator CurIter(current_sites.begin());
do
{
// Set the status of mod_sites to "occupied".
Modifier new_manager(mod_manager);
for(auto mod_iter = mod_vec.begin(); mod_iter != mod_vec.end(); mod_iter++)
{
new_manager.modifyModificationStatus(*iter, *mod_iter, 0);
}
// Select the current combination, set the status and move to the next symbol.
ModificationSequence current_seq(mod_seq);
ModificationSites mod_sites(mod_vec.begin(), mod_vec.end());
current_seq.insert(std::make_pair(*iter, mod_sites));
set<std::string>::iterator it_copy = iter;
it_copy++;
if(it_copy == mod_set.end()) {
seq_list.push_back(current_seq);
continue;
}
this->nextModificationSymbol(mod_set, it_copy, seq_list, current_seq, new_manager);
} while(stdcomb::next_combination(current_vec.begin(), current_vec.end(), mod_vec.begin(), mod_vec.end()));
}
int main (int argc, char ** argv)
{
srand(time(NULL));
options_description desc("Allowed options");
desc.add_options()
("help,?", "produce help message")
("outputDirectory,o", value<string> (), "name of directory where output files will be written in")
("alignmentmodel,a", value<string> (), "model")
("predictionmodels,p", value<string> (), "file with a list of prediction models (one for each sequence file)")
("sequences,s", value<string> (), "file with a list of sequence files (one for each organism). Sequences that are to be aligned should be in the same position in the files")
("iterations,i", value<int> (), "number of consistency iterations")
("fasta,F", "use fasta format")
("alternate", "alternate between a round of alignment-prediction consistency, and a round of prediction-alignment consistency");
try
{
variables_map vm;
store(parse_command_line(argc, argv, desc), vm);
notify(vm);
if(vm.count("help"))
{
cout << desc << "\n";
return 1;
}
string models_file = vm["predictionmodels"].as<std::string>();
vector<ProbabilisticModelPtr> predmodels;
ifstream fin;
fin.open(models_file.c_str());
while(!fin.eof()){
ProbabilisticModelCreatorClient creator;
string model_file;
fin >> model_file;
if(model_file.compare("") == 0){
continue;
}
cerr << model_file << endl;
ProbabilisticModelPtr predmodel = creator.create(model_file);
if(predmodel == NULL)
{
cerr << "Could not create model from file " << model_file << endl;
exit(-1);
}
predmodels.push_back(predmodel);
}
fin.close();
string model_file = vm["alignmentmodel"].as<std::string>();
ProbabilisticModelCreatorClient creator;
ProbabilisticModelPtr almodel = creator.create(model_file);
if(almodel == NULL)
{
exit(-1);
}
if(vm.count("fasta") )
SequenceFormatManager::instance()->setFormat(FastaSequenceFormatPtr(new FastaSequenceFormat()));
int numit = 1;
if(vm.count("iterations"))
numit = vm["iterations"].as<int>();
bool alternate = false;
if(vm.count("alternate"))
alternate = true;
SparseMatrix::setppthresh(0.01);
AlphabetPtr alphabet = almodel->alphabet();
string outDir = vm["outputDirectory"].as<std::string>();
string seqsFile = vm["sequences"].as<std::string>();
vector<SequenceList> listOfSeqLists;
vector< vector<string> > listOfSeqNamesLists;
fin.open(seqsFile.c_str());
while(!fin.eof()){
string seqFileName;
fin >> seqFileName;
if(seqFileName.compare("") == 0 || seqFileName.compare("\n") == 0)
continue;
cerr << seqFileName << endl;
SequenceEntryList sample_set;
readSequencesFromFile(sample_set, alphabet, seqFileName.c_str());
SequenceList seqs;
std::vector<std::string> names;
for(int i = 0; i < (int)sample_set.size(); i++){
seqs.push_back(sample_set[i]->getSequence());
names.push_back(sample_set[i]->getName());
}
listOfSeqLists.push_back(seqs);
listOfSeqNamesLists.push_back(names);
}
fin.close();
for(int i = 0; i < (int)listOfSeqLists[0].size(); i++){
SequenceList seqs;
std::vector<std::string> names;
for(int j = 0; j < (int)listOfSeqLists.size(); j++){
seqs.push_back(listOfSeqLists[j][i]);
names.push_back(listOfSeqNamesLists[j][i]);
}
MultipleAlignmentPtr ma = MultipleAlignmentPtr(new MultipleAlignment());
ma->computePredictionsAndAlignment(almodel,predmodels,seqs,names,numit,alternate,outDir);
}
}
catch (boost::program_options::invalid_command_line_syntax &e)
{
cerr << "error: " << e.what() << std::endl;
cerr << desc << endl;
}
catch (boost::program_options::unknown_option &e)
{
cerr << "error: " << e.what() << std::endl;
cerr << desc << endl;
}
catch (boost::bad_any_cast & e)
{
cerr << desc << endl;
}
return 0;
}
unsigned int NRows(void) const { return sequences->size(); }
