/// <summary>
/// Implementation of the IMatcher interface. An any pattern matches any sequence.
/// <see cref="QUT.Bio.BioPatML.Patterns.IMatcher">IMatcher interface</see>.
/// </summary>
/// <param name="sequence">Sequence to compare with.</param>
/// <param name="position">Matching position.</param>
/// <returns>The matched item</returns>
public Match Match(Sequence sequence, int position)
{
Match match = Composition.Matched;
CurrLength = (int)(Composition.MinLength + Counter * Composition.IncLength + 0.5);
CurrLength = Math.Min(CurrLength, Composition.MaxLength);
increment = 0;
Counter++;
double sum = 0.0;
for (int len = 0; len < CurrLength; len++)
sum += Composition.Weight(sequence.GetSymbol(position + len));
if (CurrLength >= Composition.MaxLength)
{
increment = 1;
Counter = 0;
}
double similarity = sum / (CurrLength * Composition.maxWeight);
if (similarity < Composition.Threshold)
return null;
match.Set(sequence, position, CurrLength, sequence.Strand, similarity);
return match;
}
/// <summary>
/// The standard override match method that performs match base on
/// inverted match algortihm
/// </summary>
/// <param name="sequence">sequence to compare</param>
/// <param name="position">matching position</param>
/// <returns></returns>
public override Match Match(Sequence sequence, int position)
{
Init();
for (int i = 0; i < matchLen; i++)
{
remSim -= (1.0 - Compare(matchSeq.GetSymbol(matchLen - i), sequence.GetSymbol(position + i)));
if (remSim / matchLen < repeat.Threshold)
return (null);
}
repeat.Matched.Set(sequence, position, matchLen, matchSeq.Strand, remSim / matchLen);
return repeat.Matched;
}
/// <summary>
/// Implementation of the IMatcher interface. An any pattern matches any sequence.
/// <see cref="QUT.Bio.BioPatML.Patterns.IMatcher">IMatcher interface</see>.
/// </summary>
/// <param name="sequence">Sequence to compare with.</param>
/// <param name="position">Matching position.</param>
/// <returns>The matched item</returns>
public Match Match(Sequence sequence, int position)
{
Match match = Matched;
int maxLen = Composition.MaxLength;
int minLen = Composition.MinLength;
double incLen = Composition.IncLength;
double sum = 0.0;
for (int len = 1; len <= minLen; len++)
sum += Composition.Weight(sequence.GetSymbol(position + len - 1));
double bestSum = sum;
int bestLen = minLen;
for (int len = minLen + 1; len <= maxLen; len++)
{
sum += Composition.Weight(sequence.GetSymbol(position + len - 1));
double diff = len - minLen;
double rest = Math.Abs(diff - incLen * (int)(diff / incLen + 0.5));
if (sum / len >= bestSum / bestLen && rest < 0.5)
{
bestSum = sum;
bestLen = len;
}
}
double similarity = bestSum / (bestLen * Composition.MaxWeight);
if (similarity < Composition.Threshold)
return null;
match.Set(sequence, position, bestLen, sequence.Strand, similarity);
return match;
}
/// <summary>
/// Implementation of the IMatcher interface. An any pattern matches any sequence.
/// <see cref="QUT.Bio.BioPatML.Patterns.IMatcher">IMatcher interface</see>.
/// </summary>
/// <param name="sequence">Sequence to compare with.</param>
/// <param name="position">Matching position.</param>
/// <returns>A match object containning the search result</returns>
public override Match Match
(Sequence sequence, int position)
{
int length = MotifSymbols.Count;
int mismatches = 0;
int maxMismatches = (int)(length * (1 - Threshold));
for (int i = 0; i < length; i++)
if (!MotifSymbols[i].Equals(sequence.GetSymbol(position + i)))
if (++mismatches > maxMismatches)
return (null);
Matched.Set(sequence, position, length, sequence.Strand,
(Double)(length - mismatches) / length);
return (Matched);
}
/// <summary>
/// Adds all symbols contained in the given sequence to the histogram.
/// </summary>
/// <param name="sequence"> The sequence with sysmbols to add. </param>
public void Add(Sequence sequence)
{
for (int pos = 1; pos <= sequence.Length; pos++)
Add(sequence.GetSymbol(pos));
}
public void TestGet () {
Sequence seq = new Sequence( AlphabetType.AA, "artg", true );
Assert.AreEqual( 'T', seq.GetSymbol( -5 ).Letter );
Assert.AreEqual( 'T', seq.GetSymbol( -1 ).Letter );
Assert.AreEqual( 'G', seq.GetSymbol( 0 ).Letter );
Assert.AreEqual( 'A', seq.GetSymbol( 1 ).Letter );
Assert.AreEqual( 'R', seq.GetSymbol( 2 ).Letter );
Assert.AreEqual( 'T', seq.GetSymbol( 3 ).Letter );
Assert.AreEqual( 'G', seq.GetSymbol( 4 ).Letter );
Assert.AreEqual( 'A', seq.GetSymbol( 5 ).Letter );
Assert.AreEqual( 'A', seq.GetSymbol( 9 ).Letter );
}
/// <summary> Creates a paragraph representing a feature.
/// </summary>
/// <param name="letters">the sequence of characters to render</param>
/// <param name="hitStart">Start position of the hit</param>
/// <param name="hitEnd">End position of hit</param>
/// <returns></returns>
static public Paragraph GetHitDetails (
Sequence sequence,
int hitStart,
int hitEnd
) {
int sequenceLength = sequence.Length;
Paragraph resultPara = new Paragraph();
int displayStartPos = CalcStartPosition( hitStart );
int nextLineBreak = displayStartPos;
//for (int i = 0; i < letters.Length; i++)
for ( int i = displayStartPos; i < CalcEndPosition( hitEnd, sequenceLength ); i++ ) {
Run lineDesc = new Run {
FontFamily = new FontFamily( "Courier" ),
FontWeight = FontWeights.Normal,
FontSize = 14
};
// Add 3 blank spaces for every set of 10 characters
if ( i % 10 == 0 ) lineDesc.Text += ( "\t\t" );
// break lines after fixed number of characters.
if ( i == 0 ) {
//first line definitely a new line
lineDesc.Text += NEWLINE
+ "Line "
+ CompileNumber( ( i + 1 ), sequenceLength )
+ " ... "
+ CompileNumber( ( i + 1 + CHARACTERS_PER_LINE ), sequenceLength )
+ "\t\t";
resultPara.Inlines.Add( lineDesc ); //Add the lines string into paragraph first
lineDesc = new Run {
FontFamily = new FontFamily( "Courier" ),
FontWeight = FontWeights.Normal,
FontSize = 14
}; //renew out run
nextLineBreak += CHARACTERS_PER_LINE; //set the next break
}
else if ( i == nextLineBreak ) {
lineDesc.Text += NEWLINE
+ "Line "
+ CompileNumber( ( i + 1 ), sequenceLength )
+ " ... "
+ CompileNumber( ( i + 1 + CHARACTERS_PER_LINE ), sequenceLength )
+ "\t\t";
resultPara.Inlines.Add( lineDesc ); //Add the lines string into paragraph first
lineDesc = new Run {
FontFamily = new FontFamily( "Courier" ),
FontWeight = FontWeights.Normal,
FontSize = 14
}; //renew out run
nextLineBreak += CHARACTERS_PER_LINE; //set the next break
}
char currentLetter = sequence.GetSymbol( i ).Letter;
if ( i >= hitStart && i <= hitEnd ) {
lineDesc.FontWeight = FontWeights.Bold;
lineDesc.Foreground = new SolidColorBrush( Colors.Black );
}
else {
// else we set the attribute of run to be normal alphabet coloring
SetRunAttr( lineDesc, currentLetter );
}
lineDesc.Text += currentLetter + " "; // otherwise append the character
resultPara.Inlines.Add( lineDesc );
}
return resultPara;
}