/*
* ModelAveragingSpeciationLikelihood.java
*
* Copyright (c) 2002-2015 Alexei Drummond, Andrew Rambaut and Marc Suchard
*
* This file is part of BEAST.
* See the NOTICE file distributed with this work for additional
* information regarding copyright ownership and licensing.
*
* BEAST is free software; you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as
* published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* BEAST is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with BEAST; if not, write to the
* Free Software Foundation, Inc., 51 Franklin St, Fifth Floor,
* Boston, MA 02110-1301 USA
*/
package dr.evomodel.speciation;
import dr.evolution.tree.Tree;
import dr.evolution.util.Units;
import dr.evomodelxml.speciation.SpeciationLikelihoodParser;
import dr.inference.model.*;
import java.util.List;
/**
* @author Alexei Drummond
*/
public class ModelAveragingSpeciationLikelihood extends AbstractModelLikelihood implements Units {
// PUBLIC STUFF
/**
* @param trees the tree
* @param speciationModels the model of speciation
* @param id a unique identifier for this likelihood
*/
public ModelAveragingSpeciationLikelihood(List<Tree> trees, List<MaskableSpeciationModel> speciationModels,
Variable<Integer> indexVariable, Variable<Double> maxIndexVariable, String id) {
this(SpeciationLikelihoodParser.SPECIATION_LIKELIHOOD, trees, speciationModels, indexVariable, maxIndexVariable);
setId(id);
}
public ModelAveragingSpeciationLikelihood(String name, List<Tree> trees, List<MaskableSpeciationModel> speciationModels,
Variable<Integer> indexVariable, Variable<Double> maxIndexVariable) {
super(name);
this.trees = trees;
this.speciationModels = speciationModels;
if (trees.size() != speciationModels.size()) {
throw new IllegalArgumentException("The number of trees and the number of speciation models should be equal.");
}
for (Tree tree : trees) {
if (tree instanceof Model) {
addModel((Model) tree);
}
}
for (SpeciationModel speciationModel : speciationModels) {
if (speciationModel != null) {
addModel(speciationModel);
}
}
if ( (indexVariable.getSize() + 1) != trees.size()) { // integer index parameter size = real size - 1
throw new IllegalArgumentException("Index parameter must be same size as the number of trees.");
}
this.indexVariable = indexVariable;
for (int i = 0; i < indexVariable.getSize(); i++) {
indexVariable.setValue(i, i+1); // if starts all 0, the top value (i+1) of index will be missing
}
indexVariable.addBounds(new Bounds.Staircase(indexVariable));
addVariable(indexVariable);
for (int i = 0; i < maxIndexVariable.getSize(); i++) {
maxIndexVariable.setValue(i, 0.0);
}
this.maxIndexVariable = maxIndexVariable;
addVariable(maxIndexVariable);
}
// **************************************************************
// ModelListener IMPLEMENTATION
// **************************************************************
protected final void handleModelChangedEvent(Model model, Object object, int index) {
likelihoodKnown = false;
}
// **************************************************************
// VariableListener IMPLEMENTATION
// **************************************************************
protected final void handleVariableChangedEvent(Variable variable, int index, Parameter.ChangeType type) {
likelihoodKnown = false;
} // No parameters to respond to
// **************************************************************
// Model IMPLEMENTATION
// **************************************************************
/**
* Stores the precalculated state: likelihood
*/
protected final void storeState() {
storedLikelihoodKnown = likelihoodKnown;
storedLogLikelihood = logLikelihood;
}
/**
* Restores the precalculated state: computed likelihood
*/
protected final void restoreState() {
likelihoodKnown = storedLikelihoodKnown;
logLikelihood = storedLogLikelihood;
}
protected final void acceptState() {
} // nothing to do
// **************************************************************
// Likelihood IMPLEMENTATION
// **************************************************************
public final Model getModel() {
return this;
}
public final double getLogLikelihood() {
if (!likelihoodKnown) {
logLikelihood = calculateLogLikelihood();
likelihoodKnown = true;
}
return logLikelihood;
}
public final void makeDirty() {
likelihoodKnown = false;
}
/**
* Calculates the log likelihood of this set of coalescent intervals,
* given a demographic model.
*
* @return the log likelihood
*/
private double calculateLogLikelihood() {
double logL = 0;
// Rule: index k cannot be appeared unless k-1 appeared
if (!isValidate(indexVariable.getValues())) {
// output("illegal index variable", indexVariable);
return Double.NEGATIVE_INFINITY;
}
for (int i = 0; i < trees.size(); i++) {
MaskableSpeciationModel model = speciationModels.get(i);
if (i > 0) {
SpeciationModel mask = speciationModels.get(indexVariable.getValue(i-1)); // integer index parameter size = real size - 1
if (model != mask) {
model.mask(mask);
} else {
model.unmask();
}
}
logL += model.calculateTreeLogLikelihood(trees.get(i));
}
Double maxI = (double) (int) getMaxIndex(indexVariable.getValues());
maxIndexVariable.setValue(0, maxI);
return logL;
}
private boolean isValidate(Integer[] pattern) {
// Rule: index k cannot be appeared unless k-1 appeared before it appears
int[] indexFreq = new int[pattern.length];
for (int i = 0; i < pattern.length; i++) {
if (pattern[i] > 0) // not validate 0
indexFreq[pattern[i] - 1] += 1; // integer index parameter size = real size - 1
if (i > 0 && (pattern[i] - pattern[i - 1] > 1)) {
for (int f = 0; f < i; f++) {
if (indexFreq[f] < 1) return false;
}
}
}
return true;
}
private int getMaxIndex(Integer[] pattern) {
int max = 0;
for (int p : pattern) {
if (p > max) {
max = p;
}
}
return max;
}
private void output(String message, Variable<Integer> indexVariable) {
System.out.print(message + ": ");
for (int i = 0; i < indexVariable.getSize(); i++) {
System.out.print(indexVariable.getValue(i) + "\t");
}
System.out.println();
}
// **************************************************************
// Loggable IMPLEMENTATION
// **************************************************************
/**
* @return the log columns.
*/
public final dr.inference.loggers.LogColumn[] getColumns() {
String columnName = getId();
if (columnName == null) columnName = getModelName() + ".likelihood";
return new dr.inference.loggers.LogColumn[]{
new LikelihoodColumn(columnName)
};
}
private final class LikelihoodColumn extends dr.inference.loggers.NumberColumn {
public LikelihoodColumn(String label) {
super(label);
}
public double getDoubleValue() {
return getLogLikelihood();
}
}
// **************************************************************
// Units IMPLEMENTATION
// **************************************************************
/**
* Sets the units these coalescent intervals are
* measured in.
*/
public final void setUnits(Type u) {
for (SpeciationModel speciationModel : speciationModels) {
speciationModel.setUnits(u);
}
}
/**
* Returns the units these coalescent intervals are
* measured in.
*/
public final Type getUnits() {
return speciationModels.get(0).getUnits();
}
// ****************************************************************
// Private and protected stuff
// ****************************************************************
/**
* The speciation models.
*/
List<MaskableSpeciationModel> speciationModels = null;
/**
* The trees.
*/
List<Tree> trees = null;
Variable<Integer> indexVariable = null; // integer index parameter size = real size - 1
Variable<Double> maxIndexVariable = null;
private double logLikelihood;
private double storedLogLikelihood;
private boolean likelihoodKnown = false;
private boolean storedLikelihoodKnown = false;
}