shared_ptr<VectorSiteContainer> SiteContainerBuilder::construct_sorted_alignment(VectorSiteContainer *sites,
bool ascending) {
VectorSequenceContainer *tmp = new VectorSequenceContainer(sites->getAlphabet());
vector<string> names = sites->getSequencesNames();
if (ascending) sort(names.begin(), names.end());
else sort(names.begin(), names.end(), greater<string>());
for (string &name : names) {
tmp->addSequence(sites->getSequence(name));
}
auto ret = make_shared<VectorSiteContainer>(*tmp);
delete tmp;
return ret;
}
void GenBank::appendFromStream(istream & input, VectorSequenceContainer & vsc) const throw (Exception)
{
if (!input) {
throw IOException ("GenBank::read: fail to open file");
}
string temp, name, sequence = ""; // Initialization
// Main loop : for all file lines
while (!input.eof())
{
getline(input, temp, '\n'); // Copy current line in temporary string
if(temp.size() >= 9 && temp.substr(0,9) == "ACCESSION")
{
name = TextTools::removeSurroundingWhiteSpaces(temp.substr(10));
StringTokenizer st(name, " ");
name = st.nextToken();
//cout << name << endl;
}
if(temp.size() >=6 && temp.substr(0,6) == "ORIGIN")
{
sequence = "";
getline(input, temp, '\n'); // Copy current line in temporary string
while(!input.eof() && temp.size() > 2 && temp.substr(0,2) != "//")
{
sequence += TextTools::removeWhiteSpaces(temp.substr(10));
getline(input, temp, '\n'); // Copy current line in temporary string
}
if(name == "") throw Exception("GenBank::read(). Sequence with no ACCESSION number!");
Sequence * seq = new Sequence(name, sequence, vsc.getAlphabet());
vsc.addSequence(* seq);
name = "";
}
}
}
int main(int args, char** argv)
{
cout << "******************************************************************" << endl;
cout << "* Bio++ Sequence Manipulator, version 2.3.0. *" << endl;
cout << "* Author: J. Dutheil Last Modif. 25/11/14 *" << endl;
cout << "******************************************************************" << endl;
cout << endl;
if (args == 1)
{
help();
return 0;
}
try {
BppApplication bppseqman(args, argv, "BppSeqMan");
bppseqman.startTimer();
// Get alphabet
Alphabet* alphabet = SequenceApplicationTools::getAlphabet(bppseqman.getParams(), "", false, true, true);
unique_ptr<GeneticCode> gCode;
CodonAlphabet* codonAlphabet = dynamic_cast<CodonAlphabet*>(alphabet);
// Get sequences:
bool aligned = ApplicationTools::getBooleanParameter("input.alignment", bppseqman.getParams(), false, "", true, 1);
OrderedSequenceContainer* sequences = 0;
if (aligned) {
VectorSiteContainer* allSites = SequenceApplicationTools::getSiteContainer(alphabet, bppseqman.getParams());
sequences = SequenceApplicationTools::getSitesToAnalyse(*allSites, bppseqman.getParams(), "", true, false);
delete allSites;
} else {
SequenceContainer* tmp = SequenceApplicationTools::getSequenceContainer(alphabet, bppseqman.getParams(), "", true, true);
sequences = new VectorSequenceContainer(*tmp);
delete tmp;
}
ApplicationTools::displayResult("Number of sequences", sequences->getNumberOfSequences());
// Perform manipulations
vector<string> actions = ApplicationTools::getVectorParameter<string>("sequence.manip", bppseqman.getParams(), ',', "", "", false, 1);
for (size_t a = 0; a < actions.size(); a++)
{
string cmdName;
map<string, string> cmdArgs;
KeyvalTools::parseProcedure(actions[a], cmdName, cmdArgs);
ApplicationTools::displayResult("Performing action", cmdName);
// +-----------------+
// | Complementation |
// +-----------------+
if (cmdName == "Complement")
{
OrderedSequenceContainer* sc = 0;
if (aligned) sc = new VectorSiteContainer(sequences->getAlphabet());
else sc = new VectorSequenceContainer(sequences->getAlphabet());
for (size_t i = 0; i < sequences->getNumberOfSequences(); i++)
{
Sequence* seq = SequenceTools::getComplement(sequences->getSequence(i));
sc->addSequence(*seq, false);
delete seq;
}
delete sequences;
sequences = sc;
}
// +------------------------+
// | (Reverse)Transcription |
// +------------------------+
else if (cmdName == "Transcript")
{
if (sequences->getAlphabet()->getAlphabetType() == AlphabetTools::DNA_ALPHABET.getAlphabetType())
{
OrderedSequenceContainer* sc = 0;
if (aligned) sc = new VectorSiteContainer(&AlphabetTools::RNA_ALPHABET);
else sc = new VectorSequenceContainer(&AlphabetTools::RNA_ALPHABET);
for (unsigned int i = 0; i < sequences->getNumberOfSequences(); i++)
{
Sequence* seq = SequenceTools::transcript(sequences->getSequence(i));
sc->addSequence(*seq, false);
delete seq;
}
delete sequences;
sequences = sc;
}
else if (sequences->getAlphabet()->getAlphabetType() == AlphabetTools::RNA_ALPHABET.getAlphabetType())
{
OrderedSequenceContainer* sc = 0;
if (aligned) sc = new VectorSiteContainer(&AlphabetTools::DNA_ALPHABET);
else sc = new VectorSequenceContainer(&AlphabetTools::DNA_ALPHABET);
for (unsigned int i = 0; i < sequences->getNumberOfSequences(); i++)
{
Sequence* seq = SequenceTools::reverseTranscript(sequences->getSequence(i));
sc->addSequence(*seq, false);
delete seq;
}
delete sequences;
sequences = sc;
}
else throw Exception("Transcription error: input alphabet must be of type 'nucleic'.");
}
// +-------------------------------+
// | Switching nucleotide alphabet |
// +-------------------------------+
else if (cmdName == "Switch")
{
const Alphabet* alpha = 0;
if (sequences->getAlphabet()->getAlphabetType() == AlphabetTools::DNA_ALPHABET.getAlphabetType())
{
alpha = &AlphabetTools::RNA_ALPHABET;
}
else if (sequences->getAlphabet()->getAlphabetType() == AlphabetTools::RNA_ALPHABET.getAlphabetType())
{
alpha = &AlphabetTools::DNA_ALPHABET;
}
else throw Exception("Cannot switch alphabet type, alphabet is not of type 'nucleic'.");
OrderedSequenceContainer* sc = 0;
if (aligned) sc = new VectorSiteContainer(alpha);
else sc = new VectorSequenceContainer(alpha);
for (size_t i = 0; i < sequences->getNumberOfSequences(); i++)
{
const Sequence* old = &sequences->getSequence(i);
vector<int> content(old->size());
for (size_t j = 0; j < old->size(); ++j)
content[j] = (*old)[j];
Sequence* seq = new BasicSequence(old->getName(), content, old->getComments(), alpha);
sc->addSequence(*seq, false);
delete seq;
}
delete sequences;
sequences = sc;
}
// +-------------+
// | Translation |
// +-------------+
else if (cmdName == "Translate")
{
if (!AlphabetTools::isCodonAlphabet(sequences->getAlphabet()))
throw Exception("Error in translation: alphabet is not of type 'codon'.");
if (cmdArgs["code"] != "")
throw Exception("ERROR: 'code' argument is deprecated. The genetic code to use for translation is now set by the top-level argument 'genetic_code'.");
if (!gCode.get()) {
string codeDesc = ApplicationTools::getStringParameter("genetic_code", bppseqman.getParams(), "Standard", "", true, 1);
ApplicationTools::displayResult("Genetic Code", codeDesc);
gCode.reset(SequenceApplicationTools::getGeneticCode(codonAlphabet->getNucleicAlphabet(), codeDesc));
}
OrderedSequenceContainer* sc = 0;
if (aligned) sc = new VectorSiteContainer(&AlphabetTools::PROTEIN_ALPHABET);
else sc = new VectorSequenceContainer(&AlphabetTools::PROTEIN_ALPHABET);
for (size_t i = 0; i < sequences->getNumberOfSequences(); ++i)
{
Sequence* seq = gCode->translate(sequences->getSequence(i));
sc->addSequence(*seq, false);
delete seq;
}
delete sequences;
sequences = sc;
}
// +-------------+
// | Remove gaps |
// +-------------+
else if (cmdName == "RemoveGaps")
{
VectorSequenceContainer* sc = new VectorSequenceContainer(sequences->getAlphabet());
for (size_t i = 0; i < sequences->getNumberOfSequences(); i++)
{
unique_ptr<Sequence> seq(sequences->getSequence(i).clone());
SequenceTools::removeGaps(*seq);
sc->addSequence(*seq);
}
delete sequences;
sequences = sc;
aligned = false;
}
// +---------------------------+
// | Change gaps to unresolved |
// +---------------------------+
else if (cmdName == "GapToUnknown")
{
OrderedSequenceContainer* sc = 0;
if (aligned) sc = new VectorSiteContainer(sequences->getAlphabet());
else sc = new VectorSequenceContainer(sequences->getAlphabet());
for (size_t i = 0; i < sequences->getNumberOfSequences(); i++)
{
Sequence* seq = new BasicSequence(sequences->getSequence(i));
SymbolListTools::changeGapsToUnknownCharacters(*seq);
sc->addSequence(*seq, false);
delete seq;
}
delete sequences;
sequences = sc;
}
// +---------------------------+
// | Change unresolved to gaps |
// +---------------------------+
else if (cmdName == "UnknownToGap")
{
OrderedSequenceContainer* sc = 0;
if (aligned) sc = new VectorSiteContainer(sequences->getAlphabet());
else sc = new VectorSequenceContainer(sequences->getAlphabet());
for (size_t i = 0; i < sequences->getNumberOfSequences(); i++)
{
Sequence* seq = new BasicSequence(sequences->getSequence(i));
SymbolListTools::changeUnresolvedCharactersToGaps(*seq);
sc->addSequence(*seq, false);
delete seq;
}
delete sequences;
sequences = sc;
}
// +--------------+
// | Remove stops |
// +--------------+
else if (cmdName == "RemoveStops")
{
if (!gCode.get()) {
string codeDesc = ApplicationTools::getStringParameter("genetic_code", bppseqman.getParams(), "Standard", "", true, 1);
ApplicationTools::displayResult("Genetic Code", codeDesc);
gCode.reset(SequenceApplicationTools::getGeneticCode(codonAlphabet->getNucleicAlphabet(), codeDesc));
}
SiteContainer* sites = dynamic_cast<SiteContainer*>(sequences);
if (!sites)
{
VectorSequenceContainer* sc = new VectorSequenceContainer(sequences->getAlphabet());
for (size_t i = 0; i < sequences->getNumberOfSequences(); ++i)
{
unique_ptr<Sequence> seq(sequences->getSequence(i).clone());
SequenceTools::removeStops(*seq, *gCode);
sc->addSequence(*seq);
}
delete sequences;
sequences = sc;
} else {
VectorSiteContainer* sc = new VectorSiteContainer(sequences->getAlphabet());
for (size_t i = 0; i < sequences->getNumberOfSequences(); ++i)
{
unique_ptr<Sequence> seq(sequences->getSequence(i).clone());
SequenceTools::replaceStopsWithGaps(*seq, *gCode);
sc->addSequence(*seq);
}
delete sequences;
sequences = sc;
}
}
// +--------------+
// | Remove stops |
// +--------------+
else if (cmdName == "RemoveColumnsWithStops")
{
SiteContainer* sites = dynamic_cast<SiteContainer*>(sequences);
if (!sites)
{
throw Exception("'RemoveColumnsWithStops' can only be used on alignment. You may consider using the 'CoerceToAlignment' command.");
}
if (!gCode.get()) {
string codeDesc = ApplicationTools::getStringParameter("genetic_code", bppseqman.getParams(), "Standard", "", true, 1);
ApplicationTools::displayResult("Genetic Code", codeDesc);
gCode.reset(SequenceApplicationTools::getGeneticCode(codonAlphabet->getNucleicAlphabet(), codeDesc));
}
for (size_t i = sites->getNumberOfSites(); i > 0; i--)
{
if (CodonSiteTools::hasStop(sites->getSite(i-1), *gCode))
sites->deleteSite(i - 1);
}
}
// +---------+
// | Get CDS |
// +---------+
else if (cmdName == "GetCDS")
{
if (!gCode.get()) {
string codeDesc = ApplicationTools::getStringParameter("genetic_code", bppseqman.getParams(), "Standard", "", true, 1);
ApplicationTools::displayResult("Genetic Code", codeDesc);
gCode.reset(SequenceApplicationTools::getGeneticCode(codonAlphabet->getNucleicAlphabet(), codeDesc));
}
OrderedSequenceContainer* sc = 0;
if (aligned) sc = new VectorSiteContainer(sequences->getAlphabet());
else sc = new VectorSequenceContainer(sequences->getAlphabet());
for (size_t i = 0; i < sequences->getNumberOfSequences(); ++i)
{
BasicSequence seq = sequences->getSequence(i);
size_t len = seq.size();
SequenceTools::getCDS(seq, *gCode, false, true, true, false);
if (aligned) {
for (size_t c = seq.size(); c < len; ++c)
seq.addElement(seq.getAlphabet()->getGapCharacterCode());
}
sc->addSequence(seq, false);
}
delete sequences;
sequences = sc;
}
// +--------------------------+
// | Resolve dotted alignment |
// +--------------------------+
else if (actions[a] == "CoerceToAlignment")
{
SiteContainer* sites = dynamic_cast<SiteContainer*>(sequences);
if(! sites)
{
sites = new VectorSiteContainer(*sequences);
delete sequences;
sequences = sites;
}
aligned = true;
}
else if (actions[a] == "ResolvedDotted")
{
SiteContainer* sites = dynamic_cast<SiteContainer *>(sequences);
if (!sites)
{
throw Exception("'ResolvedDotted' can only be used on alignment. You may consider using the 'CoerceToAlignment' command.");
}
const Alphabet* alpha = 0;
string alphastr = ApplicationTools::getStringParameter("alphabet", cmdArgs, "DNA", "", false, 1);
if (alphastr == "DNA") alpha = &AlphabetTools::DNA_ALPHABET;
else if (alphastr == "RNA") alpha = &AlphabetTools::RNA_ALPHABET;
else if (alphastr == "Protein") alpha = &AlphabetTools::PROTEIN_ALPHABET;
else throw Exception("Resolved alphabet must be one of [DNA|RNA|Protein] for solving dotted alignment.");
OrderedSequenceContainer* resolvedCont = SiteContainerTools::resolveDottedAlignment(*sites, alpha);
delete sequences;
sequences = resolvedCont;
}
// +---------------------+
// | Keep complete sites |
// +---------------------+
else if (cmdName == "KeepComplete")
{
SiteContainer* sites = dynamic_cast<SiteContainer *>(sequences);
if (!sites)
{
throw Exception("'KeepComplete' can only be used on alignment. You may consider using the 'CoerceToAlignment' command.");
}
string maxGapOption = ApplicationTools::getStringParameter("maxGapAllowed", cmdArgs, "100%", "", false, 1);
if (maxGapOption[maxGapOption.size()-1] == '%')
{
double gapFreq = TextTools::toDouble(maxGapOption.substr(0, maxGapOption.size()-1)) / 100.;
for (size_t i = sites->getNumberOfSites(); i > 0; i--)
{
map<int, double> freqs;
SiteTools::getFrequencies(sites->getSite(i - 1), freqs);
if (freqs[-1] > gapFreq) sites->deleteSite(i - 1);
}
}
else
{
size_t gapNum = TextTools::to<size_t>(maxGapOption);
for (size_t i = sites->getNumberOfSites(); i > 0; i--)
{
map<int, size_t> counts;
SiteTools::getCounts(sites->getSite(i - 1), counts);
counts[-1]; //Needed in case this entry does not exist in the map. This will set it to 0.
if (counts[-1] > gapNum) sites->deleteSite(i-1);
}
}
}
// +-----------------+
// | Invert sequence |
// +-----------------+
else if (cmdName == "Invert")
{
OrderedSequenceContainer* sc = 0;
if (aligned) sc = new VectorSiteContainer(sequences->getAlphabet());
else sc = new VectorSequenceContainer(sequences->getAlphabet());
for (size_t i = 0; i < sequences->getNumberOfSequences(); i++)
{
const Sequence* old = &sequences->getSequence(i);
Sequence* seq = SequenceTools::getInvert(*old);
sc->addSequence(*seq, false);
delete seq;
}
delete sequences;
sequences = sc;
}
// +------------------+
// | GetCodonPosition |
// +------------------+
else if (cmdName == "GetCodonPosition")
{
unsigned int pos = ApplicationTools::getParameter<unsigned int>("position", cmdArgs, 3, "", false, 1);
OrderedSequenceContainer* sc = dynamic_cast<OrderedSequenceContainer*>(SequenceContainerTools::getCodonPosition(*sequences, pos - 1));
delete sequences;
if (aligned) {
sequences = new VectorSiteContainer(*sc);
delete sc;
} else {
sequences = sc;
}
}
// +-----------------+
// | FilterFromTree |
// +-----------------+
else if (cmdName == "FilterFromTree")
{
unique_ptr<Tree> tree(PhylogeneticsApplicationTools::getTree(cmdArgs, ""));
vector<string> names = tree->getLeavesNames();
OrderedSequenceContainer* reorderedSequences = 0;
if (aligned) {
reorderedSequences = new VectorSiteContainer(sequences->getAlphabet());
} else {
reorderedSequences = new VectorSequenceContainer(sequences->getAlphabet());
}
for (size_t i = 0; i < names.size(); ++i) {
reorderedSequences->addSequence(sequences->getSequence(names[i]), false);
}
delete sequences;
sequences = reorderedSequences;
}
// +----------------------+
// | RemoveEmptySequences |
// +----------------------+
else if (cmdName == "RemoveEmptySequences")
{
OrderedSequenceContainer* sc = 0;
if (aligned) sc = new VectorSiteContainer(sequences->getAlphabet());
else sc = new VectorSequenceContainer(sequences->getAlphabet());
for (size_t i = 0; i < sequences->getNumberOfSequences(); ++i)
{
if (SequenceTools::getNumberOfSites(sequences->getSequence(i))!=0)
sc->addSequence(sequences->getSequence(i), false);
}
delete sequences;
sequences = sc;
}
else throw Exception("Unknown action: " + cmdName);
}
// Write sequences
ApplicationTools::displayBooleanResult("Final sequences are aligned", aligned);
if (aligned)
{
SequenceApplicationTools::writeAlignmentFile(*dynamic_cast<SiteContainer*>(sequences), bppseqman.getParams(), "", true, 1);
}
else
{
SequenceApplicationTools::writeSequenceFile(*sequences, bppseqman.getParams(), "", true, 1);
}
delete alphabet;
delete sequences;
bppseqman.done();
} catch(exception & e) {
cout << e.what() << endl;
return 1;
}
return 0;
}
void Mase::appendFromStream(istream & input, VectorSequenceContainer & vsc) const throw (Exception)
{
if (!input) { throw IOException ("Mase::read : fail to open file"); }
// Initialization
Comments seqComments, fileComments;
string temp, name, sequence = "";
bool comments = false;
// Get current general comments is VectorSequenceContainer
fileComments = vsc.getGeneralComments();
// Main loop : for all file lines
while(!input.eof())
{
getline(input, temp, '\n'); // Copy current line in temporary string
// If first character is ;
if(temp[0] == ';')
{
// If second character is also ;
if(temp[1] == ';')
{
// File comments isolation
temp.erase(0,2); // Characters ;; deletion
if(temp != "") fileComments.push_back(temp);
}
else
{
// If a name and a sequence were founded
if((name != "") && (sequence != ""))
{
// New sequence creation, and addition in existing VectorSequenceContainer
vsc.addSequence(Sequence(name, sequence, seqComments, vsc.getAlphabet()), _checkNames);
name = "";
sequence = "";
seqComments.clear();
}
// Sequence commentaries isolation
temp.erase(temp.begin()); // Character ; deletion
if(temp != "") seqComments.push_back(temp);
comments = true;
}
}
else
{
// If sequence commentaries were just isolated
if(comments)
{
// Sequence name isolation
name = temp;
comments = false;
}
else sequence += temp; // Sequence isolation
}
}
// Addition of the last sequence in file
if((name != "") && (sequence != ""))
{
vsc.addSequence(Sequence(name, sequence, seqComments, vsc.getAlphabet()), _checkNames);
}
// Set new general comments in VectorSequenceContainer (old + new comments)
vsc.setGeneralComments(fileComments);
}