void writeCharactersBlockToStream(
const NxsCharactersBlock & cb,
ostream & outf,
const std::vector<std::string> & taxaNames,
MultiFormatReader::DataFormatType f,
long interleaveLen)
{
const unsigned nt = taxaNames.size();
const unsigned nc = cb.GetNChar();
unsigned nCharsToWrite;
unsigned seqStartColumn = 0;
if (IsRelaxedPhylipType(f))
{
for (unsigned i = 0; i < nt; ++i)
{
const std::string & name = taxaNames[i];
if (name.length() > seqStartColumn)
seqStartColumn = name.length();
}
seqStartColumn += 1;
}
if (IsPhylipType(f) || IsRelaxedPhylipType(f))
{
std::string sep;
outf << nt << ' ' << nc << '\n';
if (IsInterleaveType(f) && interleaveLen > 0)
{
std::vector<std::string> storedSeqs;
storedSeqs.reserve(nt);
std::string *sp;
nCharsToWrite = (nc > (unsigned) interleaveLen ? (unsigned)interleaveLen : nc);
for (unsigned i = 0; i < nt; ++i)
{
const std::string & name = taxaNames[i];
storedSeqs.push_back(cb.GetMatrixRowAsStr(i));
sp = &(storedSeqs[i]);
if (IsRelaxedPhylipType(f))
{
sep.clear();
sep.append(seqStartColumn - name.length(), ' ');
}
outf << name << sep << sp->substr(0, nCharsToWrite) << '\n';
}
for (unsigned currIndex = (unsigned)interleaveLen; currIndex < nc; currIndex += (unsigned)interleaveLen)
{
outf << '\n';
nCharsToWrite = ((nc - currIndex) > (unsigned)interleaveLen ? (unsigned)interleaveLen : (nc - currIndex));
for (unsigned i = 0; i < nt; ++i)
{
sp = &(storedSeqs[i]);
outf << sp->substr(currIndex, nCharsToWrite) << '\n';
}
}
}
else
{
if (interleaveLen > 0)
{
// not interleaved, but wrapping at interleaveLen
for (unsigned i = 0; i < nt; ++i)
{
nCharsToWrite = (nc > (unsigned)interleaveLen ? (unsigned)interleaveLen : nc);
const std::string & name = taxaNames[i];
std::string seq = cb.GetMatrixRowAsStr(i);
if (IsRelaxedPhylipType(f))
{
sep.clear();
sep.append(seqStartColumn - name.length(), ' ');
}
outf << name << sep << seq.substr(0, nCharsToWrite) << '\n';
for (unsigned currIndex = (unsigned)interleaveLen; currIndex < nc; currIndex += (unsigned)interleaveLen)
{
nCharsToWrite = ((nc - currIndex) > (unsigned)interleaveLen ? (unsigned)interleaveLen : (nc - currIndex));
outf << seq.substr(currIndex, nCharsToWrite) << '\n';
}
}
}
else
{
// not interleaved, and not wrapping
for (unsigned i = 0; i < nt; ++i)
{
const std::string & name = taxaNames[i];
std::string seq = cb.GetMatrixRowAsStr(i);
if (IsRelaxedPhylipType(f))
{
sep.clear();
sep.append(seqStartColumn - name.length(), ' ');
}
outf << name << sep << seq << '\n';
}
}
}
}
else if (IsFastaType(f))
{
if (interleaveLen < 1)
interleaveLen = 60; // default FASTA line length
for (unsigned i = 0; i < nt; ++i)
{
nCharsToWrite = (nc > (unsigned)interleaveLen ? (unsigned)interleaveLen : nc);
const std::string & name = taxaNames[i];
std::string seq = cb.GetMatrixRowAsStr(i);
outf << '>' << name << '\n' << seq.substr(0, nCharsToWrite) << '\n';
for (unsigned currIndex = (unsigned)interleaveLen; currIndex < nc; currIndex += (unsigned)interleaveLen)
{
nCharsToWrite = ((nc - currIndex) > (unsigned)interleaveLen ? (unsigned)interleaveLen : (nc - currIndex));
outf << seq.substr(currIndex, nCharsToWrite) << '\n';
}
}
}
else
{
throw NxsException("writeCharactersBlockToStream requested for unsupported format");
}
}
void AminoacidData::CreateMatrixFromNCL(GarliReader &reader){
NxsCharactersBlock *charblock;
int num=0, numNuc = -1;
do{
charblock = reader.GetCharactersBlock(num);
if(charblock->GetDataType() == NxsCharactersBlock::protein){
if(numNuc < 0) numNuc = num;
else{
throw ErrorException("Multiple characters/data blocks containing protein data found in Nexus datafile!\n\tEither combine the blocks or comment one out.");
}
}
else outman.UserMessage("Ignoring non-protein characters block from Nexus datafile");
num++;
}while(num < reader.NumCharBlocks());
if(numNuc < 0) throw ErrorException("No characters/data blocks containing protein data found in Nexus datafile!");
charblock = reader.GetCharactersBlock(numNuc);
if(charblock->GetNumActiveChar() < charblock->GetNChar()){
outman.UserMessageNoCR("Excluded characters:\n\t");
for(int c=0;c<charblock->GetNCharTotal();c++)
if(charblock->IsExcluded(c)) outman.UserMessageNoCR("%d ", c+1);
outman.UserMessage("");
}
// vector<unsigned> reducedToOrigCharMap = charblock->GetOrigIndexVector();
NxsTaxaBlock *taxablock = reader.GetTaxaBlock();
int numOrigTaxa = charblock->GetNTax();
int numActiveTaxa = charblock->GetNumActiveTaxa();
int numOrigChar = charblock->GetNChar();
int numActiveChar = charblock->GetNumActiveChar();
//int num_chars = reducedToOrigCharMap.size();
//cout << num_chars << endl;
NewMatrix( numActiveTaxa, numActiveChar );
// read in the data, including taxon names
int i=0;
for( int origTaxIndex = 0; origTaxIndex < numOrigTaxa; origTaxIndex++ ) {
if(charblock->IsActiveTaxon(origTaxIndex)){
//internally, blanks in taxon names will be stored as underscores
NxsString tlabel = taxablock->GetTaxonLabel(origTaxIndex);
tlabel.BlanksToUnderscores();
SetTaxonLabel( i, tlabel.c_str());
int j = 0;
bool firstAmbig = true;
for( int origIndex = 0; origIndex < numOrigChar; origIndex++ ) {
if(charblock->IsActiveChar(origIndex)){
unsigned char datum = '\0';
if(charblock->IsGapState(origTaxIndex, origIndex) == true) datum = 20;
else if(charblock->IsMissingState(origTaxIndex, origIndex) == true) datum = 20;
else{
int nstates = charblock->GetNumStates(origTaxIndex, origIndex);
//assert(nstates == 1);
//need to deal with the possibility of multiple states represented in matrix
//just convert to full ambiguity
if(nstates == 1)
datum = CharToDatum(charblock->GetState(origTaxIndex, origIndex, 0));
else{
if(firstAmbig){
outman.UserMessageNoCR("Partially ambiguous characters of taxon %s converted to full ambiguity:\n\t", TaxonLabel(origTaxIndex));
firstAmbig = false;
}
outman.UserMessageNoCR("%d ", origIndex+1);
datum = CharToDatum('?');
}
}
SetMatrix( i, j++, datum );
}
}
if(firstAmbig == false) outman.UserMessage("");
i++;
}
}
}
void DataMatrix::CreateMatrixFromNCL(GarliReader &reader){
NxsCharactersBlock *charblock = reader.GetCharactersBlock();
// vector<unsigned> reducedToOrigCharMap = charblock->GetOrigIndexVector();
NxsTaxaBlock *taxablock = reader.GetTaxaBlock();
int numOrigTaxa = charblock->GetNTax();
int numActiveTaxa = charblock->GetNumActiveTaxa();
int numOrigChar = charblock->GetNChar();
int numActiveChar = charblock->GetNumActiveChar();
//int num_chars = reducedToOrigCharMap.size();
//cout << num_chars << endl;
NewMatrix( numActiveTaxa, numActiveChar );
// read in the data, including taxon names
int i=0;
if(modSpec.IsAminoAcid() == false){
for( int origTaxIndex = 0; origTaxIndex < numOrigTaxa; origTaxIndex++ ) {
if(charblock->IsActiveTaxon(origTaxIndex)){
SetTaxonLabel( i, taxablock->GetTaxonLabel(origTaxIndex).c_str());
int j = 0;
for( int origIndex = 0; origIndex < numOrigChar; origIndex++ ) {
if(charblock->IsActiveChar(origIndex)){
unsigned char datum = '\0';
if(charblock->IsGapState(origTaxIndex, origIndex) == true) datum = 15;
else if(charblock->IsMissingState(origTaxIndex, origIndex) == true) datum = 15;
else{
int nstates = charblock->GetNumStates(origTaxIndex, origIndex);
for(int s=0;s<nstates;s++){
datum += CharToBitwiseRepresentation(charblock->GetState(origTaxIndex, origIndex, s));
}
}
SetMatrix( i, j++, datum );
}
}
i++;
}
}
}
else{
for( int origTaxIndex = 0; origTaxIndex < numOrigTaxa; origTaxIndex++ ) {
if(charblock->IsActiveTaxon(origTaxIndex)){
SetTaxonLabel( i, taxablock->GetTaxonLabel(origTaxIndex).c_str());
int j = 0;
for( int origIndex = 0; origIndex < numOrigChar; origIndex++ ) {
if(charblock->IsActiveChar(origIndex)){
unsigned char datum = '\0';
if(charblock->IsGapState(origTaxIndex, origIndex) == true) datum = 20;
else if(charblock->IsMissingState(origTaxIndex, origIndex) == true) datum = 20;
else{
int nstates = charblock->GetNumStates(origTaxIndex, origIndex);
assert(nstates == 1);
datum = CharToAminoAcidNumber(charblock->GetState(origTaxIndex, origIndex, 0));
/* for(int s=0;s<nstates;s++){
datum += CharToBitwiseRepresentation(charblock->GetState(origTaxIndex, origIndex, s));
}
*/ }
SetMatrix( i, j++, datum );
}
}
i++;
}
}
}
}
void NucleotideData::CreateMatrixFromNCL(GarliReader &reader){
NxsCharactersBlock *charblock;
int num=0, numNuc = -1;
do{
charblock = reader.GetCharactersBlock(num);
if(charblock->GetDataType() == NxsCharactersBlock::nucleotide ||
charblock->GetDataType() == NxsCharactersBlock::dna ||
charblock->GetDataType() == NxsCharactersBlock::rna){
if(numNuc < 0) numNuc = num;
else{
throw ErrorException("Multiple characters/data blocks containing nucleotide data found in Nexus datafile!\n\tEither combine the blocks or comment one out.");
}
}
else outman.UserMessage("Ignoring non-nucleotide characters block from Nexus datafile");
num++;
}while(num < reader.NumCharBlocks());
if(numNuc < 0) throw ErrorException("No characters/data blocks containing nucleotide data found in Nexus datafile!");
charblock = reader.GetCharactersBlock(numNuc);
if(charblock->GetNumActiveChar() < charblock->GetNChar()){
outman.UserMessageNoCR("Excluded characters:\n\t");
for(int c=0;c<charblock->GetNCharTotal();c++)
if(charblock->IsExcluded(c)) outman.UserMessageNoCR("%d ", c+1);
outman.UserMessage("");
}
// vector<unsigned> reducedToOrigCharMap = charblock->GetOrigIndexVector();
NxsTaxaBlock *taxablock = reader.GetTaxaBlock();
int numOrigTaxa = charblock->GetNTax();
int numActiveTaxa = charblock->GetNumActiveTaxa();
int numOrigChar = charblock->GetNChar();
int numActiveChar = charblock->GetNumActiveChar();
//int num_chars = reducedToOrigCharMap.size();
//cout << num_chars << endl;
NewMatrix( numActiveTaxa, numActiveChar );
// read in the data, including taxon names
int i=0;
for( int origTaxIndex = 0; origTaxIndex < numOrigTaxa; origTaxIndex++ ) {
if(charblock->IsActiveTaxon(origTaxIndex)){
//internally, blanks in taxon names will be stored as underscores
NxsString tlabel = taxablock->GetTaxonLabel(origTaxIndex);
tlabel.BlanksToUnderscores();
SetTaxonLabel( i, tlabel.c_str());
int j = 0;
for( int origIndex = 0; origIndex < numOrigChar; origIndex++ ) {
if(charblock->IsActiveChar(origIndex)){
unsigned char datum = '\0';
if(charblock->IsGapState(origTaxIndex, origIndex) == true) datum = 15;
else if(charblock->IsMissingState(origTaxIndex, origIndex) == true) datum = 15;
else{
int nstates = charblock->GetNumStates(origTaxIndex, origIndex);
for(int s=0;s<nstates;s++){
datum += CharToBitwiseRepresentation(charblock->GetState(origTaxIndex, origIndex, s));
}
}
SetMatrix( i, j++, datum );
}
}
i++;
}
}
}
