/*
* HKY.java
*
* Copyright (c) 2002-2016 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.substmodel.nucleotide;
import dr.evomodel.substmodel.BaseSubstitutionModel;
import dr.evomodel.substmodel.EigenDecomposition;
import dr.evomodel.substmodel.FrequencyModel;
import dr.inference.model.Parameter;
import dr.inference.model.Statistic;
import dr.evolution.datatype.Nucleotides;
import dr.math.matrixAlgebra.Vector;
import dr.util.Author;
import dr.util.Citable;
import dr.util.Citation;
import java.util.Collections;
import java.util.List;
/**
* Hasegawa-Kishino-Yano model of nucleotide evolution
*
* @author Alexei Drummond
* @author Andrew Rambaut
* @author Marc A. Suchard
*/
public class HKY extends BaseSubstitutionModel implements Citable {
private Parameter kappaParameter = null;
/**
* A constructor which allows a more programmatic approach with
* fixed kappa.
* @param kappa
* @param freqModel
*/
public HKY(double kappa, FrequencyModel freqModel) {
this(new Parameter.Default(kappa), freqModel);
}
/**
* Constructor
* @param kappaParameter
* @param freqModel
*/
public HKY(Parameter kappaParameter, FrequencyModel freqModel) {
super("HKY", Nucleotides.INSTANCE, freqModel);
this.kappaParameter = kappaParameter;
addVariable(kappaParameter);
kappaParameter.addBounds(new Parameter.DefaultBounds(Double.POSITIVE_INFINITY, 0.0, 1));
addStatistic(tsTvStatistic);
}
/**
* set kappa
* @param kappa
*/
public void setKappa(double kappa) {
kappaParameter.setParameterValue(0, kappa);
updateMatrix = true;
}
/**
* @return kappa
*/
public final double getKappa() {
return kappaParameter.getParameterValue(0);
}
/**
* set ts/tv
* @param tsTv
*/
public void setTsTv(double tsTv) {
double freqA = freqModel.getFrequency(0);
double freqC = freqModel.getFrequency(1);
double freqG = freqModel.getFrequency(2);
double freqT = freqModel.getFrequency(3);
double freqR = freqA + freqG;
double freqY = freqC + freqT;
setKappa((tsTv * freqR * freqY) / (freqA * freqG + freqC * freqT));
}
/**
* @return tsTv
*/
public double getTsTv() {
double freqA = freqModel.getFrequency(0);
double freqC = freqModel.getFrequency(1);
double freqG = freqModel.getFrequency(2);
double freqT = freqModel.getFrequency(3);
double freqR = freqA + freqG;
double freqY = freqC + freqT;
double tsTv = (getKappa() * (freqA * freqG + freqC * freqT)) / (freqR * freqY);
return tsTv;
}
protected void frequenciesChanged() {
}
protected void ratesChanged() {
}
protected void setupRelativeRates(double[] rates) {
double kappa = kappaParameter.getParameterValue(0);
rates[0] = 1.0;
rates[1] = kappa;
rates[2] = 1.0;
rates[3] = 1.0;
rates[4] = kappa;
rates[5] = 1.0;
}
public synchronized EigenDecomposition getEigenDecomposition() {
if (eigenDecomposition == null) {
double[] evec = new double[stateCount * stateCount];
double[] ivec = new double[stateCount * stateCount];
double[] eval = new double[stateCount];
eigenDecomposition = new EigenDecomposition(evec, ivec, eval);
ivec[2 * stateCount + 1] = 1; // left eigenvectors 3 = (0,1,0,-1); 4 = (1,0,-1,0)
ivec[2 * stateCount + 3] = -1;
ivec[3 * stateCount + 0] = 1;
ivec[3 * stateCount + 2] = -1;
evec[0 * stateCount + 0] = 1; // right eigenvector 1 = (1,1,1,1)'
evec[1 * stateCount + 0] = 1;
evec[2 * stateCount + 0] = 1;
evec[3 * stateCount + 0] = 1;
}
if (updateMatrix) {
double[] evec = eigenDecomposition.getEigenVectors();
double[] ivec = eigenDecomposition.getInverseEigenVectors();
double[] pi = freqModel.getFrequencies();
double piR = pi[0] + pi[2];
double piY = pi[1] + pi[3];
// left eigenvector #1
ivec[0 * stateCount + 0] = pi[0]; // or, evec[0] = pi;
ivec[0 * stateCount + 1] = pi[1];
ivec[0 * stateCount + 2] = pi[2];
ivec[0 * stateCount + 3] = pi[3];
// left eigenvector #2
ivec[1 * stateCount + 0] = pi[0]*piY;
ivec[1 * stateCount + 1] = -pi[1]*piR;
ivec[1 * stateCount + 2] = pi[2]*piY;
ivec[1 * stateCount + 3] = -pi[3]*piR;
// right eigenvector #2
evec[0 * stateCount + 1] = 1.0/piR;
evec[1 * stateCount + 1] = -1.0/piY;
evec[2 * stateCount + 1] = 1.0/piR;
evec[3 * stateCount + 1] = -1.0/piY;
// right eigenvector #3
evec[1 * stateCount + 2] = pi[3]/piY;
evec[3 * stateCount + 2] = -pi[1]/piY;
// right eigenvector #4
evec[0 * stateCount + 3] = pi[2]/piR;
evec[2 * stateCount + 3] = -pi[0]/piR;
// eigenvectors
double[] eval = eigenDecomposition.getEigenValues();
final double kappa = getKappa();
final double beta = -1.0 / (2.0 * (piR * piY + kappa * (pi[0] * pi[2] + pi[1] * pi[3])));
final double A_R = 1.0 + piR * (kappa - 1);
final double A_Y = 1.0 + piY * (kappa - 1);
eval[1] = beta;
eval[2] = beta*A_Y;
eval[3] = beta*A_R;
updateMatrix = false;
}
return eigenDecomposition;
}
//
// Private stuff
//
private Statistic tsTvStatistic = new Statistic.Abstract() {
public String getStatisticName() {
return "tsTv";
}
public int getDimension() {
return 1;
}
public double getStatisticValue(int dim) {
return getTsTv();
}
};
@Override
public Citation.Category getCategory() {
return Citation.Category.SUBSTITUTION_MODELS;
}
@Override
public String getDescription() {
return "HKY nucleotide substitution model";
}
@Override
public List<Citation> getCitations() {
return Collections.singletonList(CITATION);
}
public static Citation CITATION = new Citation(
new Author[]{
new Author("M", "Hasegawa"),
new Author("H", "Kishino"),
new Author("T", "Yano")
},
"Dating the human-ape splitting by a molecular clock of mitochondrial DNA",
1985,
"J. Mol. Evol.",
22,
160, 174,
Citation.Status.PUBLISHED
);
public static void main(String[] args) {
// double kappa = 2.0;
// double[] pi = new double[]{0.15,0.30,0.20,0.35};
// double time = 0.1;
double kappa = 1.0;
double[] pi = new double[]{0.25,0.25,0.25,0.25};
double time = 0.1;
FrequencyModel freqModel= new FrequencyModel(Nucleotides.INSTANCE, pi);
HKY hky = new HKY(kappa,freqModel);
EigenDecomposition decomp = hky.getEigenDecomposition();
// Matrix evec = new Matrix(decomp.getEigenVectors());
// Matrix ivec = new Matrix(decomp.getInverseEigenVectors());
// System.out.println("Evec =\n"+evec);
// System.out.println("Ivec =\n"+ivec);
Vector eval = new Vector(decomp.getEigenValues());
System.out.println("Eval = "+eval);
double[] probs = new double[16];
hky.getTransitionProbabilities(time,probs);
System.out.println("new probs = "+new Vector(probs));
// check against old implementation
dr.oldevomodel.substmodel.FrequencyModel oldFreq = new dr.oldevomodel.substmodel.FrequencyModel(Nucleotides.INSTANCE,pi);
dr.oldevomodel.substmodel.HKY oldHKY = new dr.oldevomodel.substmodel.HKY(kappa,oldFreq);
oldHKY.setKappa(kappa);
oldHKY.getTransitionProbabilities(time,probs);
System.out.println("old probs = "+new Vector(probs));
}
}