edu.rit.compbio.seq

Class DefaultAlignmentStats

java.lang.Object

edu.rit.compbio.seq.DefaultAlignmentStats

All Implemented Interfaces:

AlignmentStats

public class DefaultAlignmentStats
extends java.lang.Object
implements AlignmentStats

Class DefaultAlignmentStats provides an object that computes statistics of an Alignment. Methods are provided to compute the raw score, the bit score, and the E-value. The formulas for these statistics assume that the alignment:

• is for two protein sequences,
• was calculated by the Smith-Waterman local alignment algorithm,
• using the BLOSUM-62 protein substitution matrix,
• using a gap existence penalty of −11 and a gap extension penalty of −1,
• matching a query sequence against a database of subject sequences, where the sum of the subject sequence lengths is supplied as a constructor parameter.

The formulas for the bit score and E-value are:

S' = (λ S − ln K)/(ln 2)
E = K m n exp(−λ S) where S is the raw score, m is the query sequence length, n is the total subject sequence length, and the parameters are K = 0.035, λ = 0.252.

These formulas were taken from:

• http://www.ncbi.nlm.nih.gov/BLAST/tutorial/Altschul-1.html
• http://www.ncbi.nlm.nih.gov/BLAST/tutorial/Altschul-3.html

Constructor Summary

Constructors

Constructor and Description
DefaultAlignmentStats(long n) Construct a new default alignment statistics object.

Method Summary

Modifier and Type	Method and Description
double	bitScore(Alignment alignment) Returns the bit score for the given alignment.
double	eValue(Alignment alignment) Returns the E-value (expect value) for the given alignment.
void	print(java.io.PrintStream out) Print information about this alignment statistics object on the given print stream.
double	rawScore(Alignment alignment) Returns the raw score for the given alignment.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

DefaultAlignmentStats

public DefaultAlignmentStats(long n)

Construct a new default alignment statistics object.

Parameters:

n - Sum of the lengths of the subject sequences in the database.

Method Detail

rawScore

public double rawScore(Alignment alignment)

Returns the raw score for the given alignment. A larger raw score signifies a greater degree of similarity between the query sequence and subject sequence that were aligned.

Specified by:

rawScore in interface AlignmentStats

Parameters:

alignment - Alignment.

Returns:

Raw score.

bitScore

public double bitScore(Alignment alignment)

Returns the bit score for the given alignment. A larger bit score signifies a greater degree of similarity between the query sequence and subject sequence that were aligned.

The bit score is the raw score normalized to units of "bits." Bit scores for different alignment procedures may be compared, whereas raw (unnormalized) scores for different alignment procedures may not be compared.

Specified by:

bitScore in interface AlignmentStats

Parameters:

alignment - Alignment.

Returns:

Bit score.

eValue

public double eValue(Alignment alignment)

Returns the E-value (expect value) for the given alignment. A smaller E-value signifies a more statistically significant degree of similarity between the query sequence and subject sequence that were aligned.

The E-value is the expected number of alignments with a score greater than or equal to the alignment's score when a randomly-chosen query of the same length as the query that produced the alignment is matched against the database.

Specified by:

eValue in interface AlignmentStats

Parameters:

alignment - Alignment.

Returns:

Bit score.

print

public void print(java.io.PrintStream out)

Print information about this alignment statistics object on the given print stream.

Specified by:

print in interface AlignmentStats

Parameters:

out - Print stream.