void FeatureInvestigator::printBayesian (
Persistance::TextWriter& writer, Feature& feature,
const PositionVector& positions)
{
const int motifLength = feature.assignment ().length ();
CPositionIterator it (positions.begin (), positions.end ());
for (; it.hasNext () ; it.next()) {
const SeqPosition& position = *(*it);
StrBuffer buf;
USELESS (int middleSection = )
position.getSeedString (buf, motifLength, _outputLength, '?');
writer << '(';
int length = buf.length();
for (int i=0 ; i<length ; i++) {
writer << buf [i] << ' ';
}
writer << ')';
writer.writeln ();
}
}
void FeatureInvestigator::printSeed (Persistance::TextTableReport::Output& out,
Feature& feature,
const PositionVector& positions)
{
int currentField = Persistance::TextTableReport::firstFieldIndex;
Persistance::TextTableReport::Data data (_seedFormat);
TextWriter writer (NULL);
//
// print all scores
Scores::Score_ptr score = feature.score ();
while (score) {
writer.setStream (data.getOutputStream(currentField++));
writer << (- score->log10Score ());
score = score->next ();
}
//
// print the assignment
writer.setStream (data.getOutputStream(currentField++));
writer << _parameters.langauge ().format (feature.assignment ());
//
// print score params if available
writer.setStream (data.getOutputStream(currentField++));
feature.score ()->writeAsText (writer);
/*
if (feature.scoreParameters ()) {
writer << '[';
feature.scoreFunction ().writeAsText (writer, feature.scoreParameters ());
writer << ']';
}*/
//
// print projection details if available
writer.setStream (data.getOutputStream(currentField++));
if (feature.projection ()) {
writer << _parameters.langauge ().format (*feature.projection ());
}
/*
//
// print how many seeds counted for bonf correction, if available
writer.setStream (data.getOutputStream(currentField++));
if (feature.numSeedsSearched () > 0) {
writer << feature.numSeedsSearched();
}
*/
writer.setStream (NULL);
out << data;
}
void FeatureInvestigator::createPSSM (PositionWeightType positionWeightType,
Feature& feature_i,
const PositionVector& positions,
PSSM& outPSSM)
{
const int seed_length = feature_i.assignment ().length ();
outPSSM = PSSM (
positionWeightType,
_parameters.langauge ().code (),
seed_length,
_outputLength,
positions,
*_parameters.wf ());
}
//
// adds the positive positions of the feature to the position vector
void FeatureInvestigator::addPositions (
Feature& feature,
PositionVector& outPositives,
PositionVector& outNegatives)
{
//
// use the positions in the cluster if they are available
if (!feature.cluster ().hasPositions ()) {
//
// it is assumed that if it has any positions, then
// it contains all relevant positions
Preprocessor::NodeCluster motifNodes;
_parameters.preprocessor ().add2Cluster ( motifNodes,
feature.assignment ());
motifNodes.add2SeqClusterPositions (feature.cluster ());
}
SeqCluster::AddPositions pos(outPositives);
SeqCluster::AddPositions neg(outNegatives);
feature.cluster ().performDivided(*_parameters.wf (), pos, neg);
}
void FeatureInvestigator::printMotifPosition (
Persistance::TextTableReport::Output& out,
Feature& feature,
const SeqPosition& position)
{
Persistance::TextTableReport::Data data (_motifPositionFormat);
const int motifLength = feature.assignment ().length ();
StrBuffer buf(_outputLength);
int middleSection =
position.getSeedString (
buf,
motifLength,
_outputLength
);
int fieldIndex = Persistance::TextTableReport::firstFieldIndex;
Persistance::TextWriter writer (data.getOutputStream (fieldIndex++));
writer << buf.substring (0, middleSection) << ' '
<< buf.substring (middleSection, middleSection + motifLength) << ' '
<< buf.substring (middleSection + motifLength);
//
// write seq id
writer.setStream(data.getOutputStream(fieldIndex++));
writer << position.sequence ()->id ();
//
// write seq name
writer.setStream(data.getOutputStream(fieldIndex++));
writer << position.sequence ()->name ();
//
// output sequence weight
writer.setStream(data.getOutputStream(fieldIndex++));
writer << '[' << _parameters.wf ()->weight (position.sequence ()->id ()) << ']';
//
// output the position information:
// we write the position from the TSS, that is:
// the upstream offset from the gene
// the promoter is assumed to be on the same strand as the gene
// and read in the same direction as the gene.
// this means that the 'reverse' strand is the strand that DOES not
// have the gene.
//
//
// 98765|432|10
// TTTTT|TGC|TT <--
// --> AAAAA|ACG|AA
// 01234|567|89
//
// CGT-positions: 2-->5
// CGT-tssPositions (): 5-->8 = (10 - 2 - 3) --> (10 - 2)
writer.setStream(data.getOutputStream(fieldIndex++));
int tssPosition = position.tssPosition (motifLength);
writer << tssPosition;
writer.setStream(data.getOutputStream(fieldIndex++));
writer << tssPosition + (position.strand ()? motifLength : - motifLength);
//
// print +/- if it is on normal/reverse strand
writer.setStream(data.getOutputStream(fieldIndex++));
writer << (position.strand ()? '+' : '-');
//
// print the sequence length
writer.setStream(data.getOutputStream(fieldIndex++));
writer << position.sequence ()->length ();
writer.setStream(NULL);
out << data;
}
