org.biojava.bio.dp
Class SimpleMarkovModel

java.lang.Object
  extended by org.biojava.utils.AbstractChangeable
      extended by org.biojava.bio.dp.SimpleMarkovModel
All Implemented Interfaces:
Serializable, MarkovModel, Changeable
Direct Known Subclasses:
FullHmmerProfileHMM, ProfileHMM

public class SimpleMarkovModel
extends AbstractChangeable
implements MarkovModel, Serializable

Author:
Matthew Pocock, Thomas Down, Samiul Hasan, Keith James
See Also:
Serialized Form

Field Summary
static long serialVersionUID
           
 
Fields inherited from interface org.biojava.bio.dp.MarkovModel
ARCHITECTURE, PARAMETER
 
Constructor Summary
SimpleMarkovModel(int heads, Alphabet emissionAlpha)
          Deprecated.  
SimpleMarkovModel(int heads, Alphabet emissionAlpha, String name)
           
 
Method Summary
 void addState(State toAdd)
          Adds a state to the model.
 int[] advance()
          The maximum advance for this model.
 boolean containsTransition(State from, State to)
          Returns wether a transition exists or not.
 void createTransition(State from, State to)
          Makes a transition between two states legal.
 void destroyTransition(State from, State to)
          Breaks a transition between two states legal.
 Alphabet emissionAlphabet()
          Alphabet that is emitted by the emission states.
protected  ChangeSupport getChangeSupport(ChangeType ct)
          Called to retrieve the ChangeSupport for this object.
 Distribution getWeights(State source)
          Get a probability Distribution over the transition from 'source'.
 int heads()
          The number of heads on this model.
 MagicalState magicalState()
          The MagicalState for this model.
 void removeState(State toGo)
          Remove a state from the model.
 void setWeights(State source, Distribution dist)
          Use this methods to customize the transition probabilities.
 FiniteAlphabet stateAlphabet()
          FiniteAlphabet of the states.
 String toString()
           
 FiniteAlphabet transitionsFrom(State from)
          Returns the FiniteAlphabet of all states that have a transition from 'source'.
 FiniteAlphabet transitionsTo(State to)
          Returns the FiniteAlphabet of all states that have a transition to 'dest'.
 
Methods inherited from class org.biojava.utils.AbstractChangeable
addChangeListener, addChangeListener, generateChangeSupport, hasListeners, hasListeners, isUnchanging, removeChangeListener, removeChangeListener
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait
 
Methods inherited from interface org.biojava.utils.Changeable
addChangeListener, addChangeListener, isUnchanging, removeChangeListener, removeChangeListener
 

Field Detail

serialVersionUID

public static final long serialVersionUID
See Also:
Constant Field Values
Constructor Detail

SimpleMarkovModel

public SimpleMarkovModel(int heads,
                         Alphabet emissionAlpha,
                         String name)

SimpleMarkovModel

public SimpleMarkovModel(int heads,
                         Alphabet emissionAlpha)
Deprecated. 

Method Detail

getChangeSupport

protected ChangeSupport getChangeSupport(ChangeType ct)
Description copied from class: AbstractChangeable
Called to retrieve the ChangeSupport for this object.

Your implementation of this method should have the following structure:

 ChangeSupport cs = super.getChangeSupport(ct);

 if(someForwarder == null && ct.isMatching(SomeInterface.SomeChangeType)) {
   someForwarder = new ChangeForwarder(...

   this.stateVariable.addChangeListener(someForwarder, VariableInterface.AChange);
 }

 return cs;
 
It is usual for the forwarding listeners (someForwarder in this example) to be transient and lazily instantiated. Be sure to register & unregister the forwarder in the code that does the ChangeEvent handling in setter methods.

Overrides:
getChangeSupport in class AbstractChangeable

emissionAlphabet

public Alphabet emissionAlphabet()
Description copied from interface: MarkovModel
Alphabet that is emitted by the emission states.

Specified by:
emissionAlphabet in interface MarkovModel

stateAlphabet

public FiniteAlphabet stateAlphabet()
Description copied from interface: MarkovModel
FiniteAlphabet of the states.

We are modeling a finite-state-machine, so there will be a finite set of states.

The MagicalState returned by getMagicalState is always contained within this as the start/end state.

Specified by:
stateAlphabet in interface MarkovModel
Returns:
the alphabet over states

heads

public int heads()
Description copied from interface: MarkovModel
The number of heads on this model.

Each head consumes a single SymbolList. A single-head model just consumes/ emits a single sequence. A two-head model performs alignment between two sequences (e.g. smith-waterman). Models with more heads do more interesting things.

heads() should equal advance().length.

Specified by:
heads in interface MarkovModel
Returns:
the number of heads in this model.

magicalState

public MagicalState magicalState()
Description copied from interface: MarkovModel
The MagicalState for this model.

Specified by:
magicalState in interface MarkovModel

advance

public int[] advance()
Description copied from interface: MarkovModel
The maximum advance for this model.

Each head consumes a single SymbolList. However, the states that advance through that SymbolList may emit more than one symbol at a time. This array give the maximum advance in each direction.

Be sure to return a new array each time this is called. This protects the internal state of the object from someone modifying the advance array. Be sure to update this as/when states modify their advance arrays and as/when states are added or removed

Specified by:
advance in interface MarkovModel
Returns:
the maximum advance of all states for all heads

getWeights

public Distribution getWeights(State source)
                        throws IllegalSymbolException
Description copied from interface: MarkovModel
Get a probability Distribution over the transition from 'source'.

Specified by:
getWeights in interface MarkovModel
Parameters:
source - the State currently occupied
Returns:
the probability Distribution over the reachable states
Throws:
IllegalSymbolException - if from is not a legal state

setWeights

public void setWeights(State source,
                       Distribution dist)
                throws IllegalSymbolException,
                       IllegalAlphabetException,
                       ChangeVetoException
Use this methods to customize the transition probabilities.

By default, the distribution P(destination | source) is a totally free distribution. This allows the different probabilities to vary. If you wish to change this behaviour (for example, to make one set of transition probabilities equal to another), then use this method to replace the Distribution with one of your own.

Specified by:
setWeights in interface MarkovModel
Parameters:
source - source State
dist - the new Distribution over the transition probabilites from source
Throws:
IllegalSymbolException - if source is not a member of this model
IllegalAlphabetException - if dist is not a distribution over the states returned by model.transitionsFrom(source)
ChangeVetoException - if a listener vetoed this change

createTransition

public void createTransition(State from,
                             State to)
                      throws IllegalSymbolException,
                             ChangeVetoException
Description copied from interface: MarkovModel
Makes a transition between two states legal.

This should inform each TransitionListener that a transition is to be created using preCreateTransition, and if none of the listeners fire a ChangeVetoException, it should create the transition, and then inform each TransitionListener with postCreateTransition.

Specified by:
createTransition in interface MarkovModel
Parameters:
from - the State currently occupied
to - the State to move to
Throws:
IllegalSymbolException - if either from or to are not legal states
ChangeVetoException - if creating the transition is vetoed

destroyTransition

public void destroyTransition(State from,
                              State to)
                       throws IllegalSymbolException,
                              ChangeVetoException
Description copied from interface: MarkovModel
Breaks a transition between two states legal.

This should inform each TransitionListener that a transition is to be broken using preDestroyTransition, and if none of the listeners fire a ChangeVetoException, it should break the transition, and then inform each TransitionListener with postDestroyTransition.

Specified by:
destroyTransition in interface MarkovModel
Parameters:
from - the State currently occupied
to - the State to move to
Throws:
IllegalSymbolException - if either from or to are not legal states
ChangeVetoException - if breaking the transition is vetoed

containsTransition

public boolean containsTransition(State from,
                                  State to)
                           throws IllegalSymbolException
Description copied from interface: MarkovModel
Returns wether a transition exists or not.

Specified by:
containsTransition in interface MarkovModel
Parameters:
from - the transitin source
to - the transition destination
Returns:
true/false depending on wether this model has the transition
Throws:
IllegalSymbolException - if either from or to are not states in this model

transitionsFrom

public FiniteAlphabet transitionsFrom(State from)
                               throws IllegalSymbolException
Description copied from interface: MarkovModel
Returns the FiniteAlphabet of all states that have a transition from 'source'.

Specified by:
transitionsFrom in interface MarkovModel
Parameters:
from - the source State
Returns:
a FiniteAlphabet of State objects that can reach from 'source'
Throws:
IllegalSymbolException

transitionsTo

public FiniteAlphabet transitionsTo(State to)
                             throws IllegalSymbolException
Description copied from interface: MarkovModel
Returns the FiniteAlphabet of all states that have a transition to 'dest'.

Specified by:
transitionsTo in interface MarkovModel
Parameters:
to - the destination state
Returns:
a FiniteAlphabet of State objects that can reach 'dest'
Throws:
IllegalSymbolException

addState

public void addState(State toAdd)
              throws IllegalSymbolException,
                     ChangeVetoException
Description copied from interface: MarkovModel
Adds a state to the model.

Specified by:
addState in interface MarkovModel
Parameters:
toAdd - the state to add
Throws:
IllegalSymbolException - if the state is not valid or is a MagicalState
ChangeVetoException - if either the model does not allow states to be added, or the change was vetoed

removeState

public void removeState(State toGo)
                 throws IllegalSymbolException,
                        IllegalTransitionException,
                        ChangeVetoException
Description copied from interface: MarkovModel
Remove a state from the model.

States should not be removed untill they are involved in no transitions. This is to avoid producing corrupted models by accident.

Specified by:
removeState in interface MarkovModel
Parameters:
toGo - the state to remove
Throws:
IllegalSymbolException - if the symbol is not part of this model or a MagicalState
IllegalTransitionException - if the state is currently involved in any transitions
ChangeVetoException - if either the model does not allow states to be removed, or the change was vetoed

toString

public String toString()
Overrides:
toString in class Object