org.biojava.bio.dist
Class AbstractDistribution

java.lang.Object
  |
  +--org.biojava.utils.AbstractChangeable
        |
        +--org.biojava.bio.dist.AbstractDistribution

All Implemented Interfaces:

Changeable, Distribution, java.io.Serializable

Direct Known Subclasses:

AbstractOrderNDistribution, SimpleDistribution, UniformDistribution

public abstract class AbstractDistribution

extends AbstractChangeable

implements Distribution, java.io.Serializable

An abstract implementation of Distribution.

You will need to over-ride getWeight() for a simple implementation. You may also wish to over-ride the other methods if the default implementation is not suitable.

Note that, in this implementation, the setWeight implementation throws an exception. The registerWithTrainer method registers an IgnoreCountsTrainer. To make an AbstractDistribution subclass trainable, both these methods must be overridden.

Author:

Matthew Pocock, Thomas Down, Mark Schreiber (serialization support), Greg Cox

See Also:

Serialized Form

Nested Class Summary

Nested classes inherited from class org.biojava.bio.dist.Distribution

Distribution.NullModelForwarder

Field Summary

protected Distribution.NullModelForwarder	nullModelForwarder
protected java.util.Map	symbolIndices

Fields inherited from interface org.biojava.bio.dist.Distribution

NULL_MODEL, WEIGHTS

Constructor Summary

AbstractDistribution()

Method Summary

protected ChangeSupport	getChangeSupport(ChangeType ct) Called to retrieve the ChangeSupport for this object
double	getWeight(Symbol sym) Retrieve the weight for this distribution.
protected abstract double	getWeightImpl(AtomicSymbol sym)
void	registerWithTrainer(DistributionTrainerContext dtc) Register an IgnoreCountsTrainer instance as the trainer for this distribution.
Symbol	sampleSymbol() Sample a symbol from this state's probability distribution.
void	setNullModel(Distribution nullModel) Set the null model Distribution that this Distribution recognizes.
protected abstract void	setNullModelImpl(Distribution nullModel)
void	setWeight(Symbol sym, double weight) Set the weight of a given symbol in this distribution.
protected abstract void	setWeightImpl(AtomicSymbol sym, double weight)

Methods inherited from class org.biojava.utils.AbstractChangeable

addChangeListener, addChangeListener, generateChangeSupport, hasListeners, isUnchanging, removeChangeListener, removeChangeListener

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.biojava.bio.dist.Distribution

getAlphabet, getNullModel

Methods inherited from interface org.biojava.utils.Changeable

addChangeListener, addChangeListener, isUnchanging, removeChangeListener, removeChangeListener

Field Detail

nullModelForwarder

protected transient Distribution.NullModelForwarder nullModelForwarder

symbolIndices

protected java.util.Map symbolIndices

Constructor Detail

AbstractDistribution

public AbstractDistribution()

Method Detail

getChangeSupport

protected ChangeSupport getChangeSupport(ChangeType ct)

Description copied from class: AbstractChangeable

Called to retrieve the ChangeSupport for this object

Overrides:

getChangeSupport in class AbstractChangeable

setWeightImpl

protected abstract void setWeightImpl(AtomicSymbol sym,
                                      double weight)
                               throws IllegalSymbolException,
                                      ChangeVetoException

IllegalSymbolException

ChangeVetoException

setWeight

public final void setWeight(Symbol sym,
                            double weight)
                     throws IllegalSymbolException,
                            ChangeVetoException

Set the weight of a given symbol in this distribution.

This implementation informs all listeners of the change, and then calls setWeightImpl to make the actual change. Sub-classes should over-ride setWeightImpl to implement the actual storage of the weights.

Specified by:

setWeight in interface Distribution

Parameters:

sym - the Symbol emitted

weight - the probability of emitting that symbol

Throws:

IllegalSymbolException - if s is not from this state's alphabet, or if it is an ambiguity symbol and the implementation can't handle this case

ChangeVetoException - if this state does not allow weights to be tampered with, or if one of the listeners vetoed this change

setNullModelImpl

protected abstract void setNullModelImpl(Distribution nullModel)
                                  throws IllegalAlphabetException,
                                         ChangeVetoException

IllegalAlphabetException

ChangeVetoException

setNullModel

public final void setNullModel(Distribution nullModel)
                        throws IllegalAlphabetException,
                               ChangeVetoException

Description copied from interface: Distribution

Set the null model Distribution that this Distribution recognizes.

Specified by:

setNullModel in interface Distribution

Parameters:

nullModel - the new null model Distribution

Throws:

IllegalAlphabetException - if the null model has the wrong alphabet

ChangeVetoException - if this Distirbution doesn't support setting the null model, or if one of its listeners objects

getWeight

public final double getWeight(Symbol sym)
                       throws IllegalSymbolException

Retrieve the weight for this distribution.

Performs the standard munge to handle ambiguity symbols. The actual weights for each attomic symbol should be calculated by the getWeightImpl functions.

Specified by:

getWeight in interface Distribution

Parameters:

sym - the Symbol emitted

Returns:

the probability that one of the symbols matching amb was emitted

Throws:

IllegalSymbolException - if for any reason the symbols within amb are not recognized by this state

getWeightImpl

protected abstract double getWeightImpl(AtomicSymbol sym)
                                 throws IllegalSymbolException

IllegalSymbolException

sampleSymbol

public Symbol sampleSymbol()

Description copied from interface: Distribution

Sample a symbol from this state's probability distribution.

Specified by:

sampleSymbol in interface Distribution

Returns:

the symbol sampled

registerWithTrainer

public void registerWithTrainer(DistributionTrainerContext dtc)

Register an IgnoreCountsTrainer instance as the trainer for this distribution. Override this if you wish to implement a trainable distribution.

Specified by:

registerWithTrainer in interface Distribution

Parameters:

dtc - the DistributionTrainerContext with witch to register a trainer