/*
* ComplexSubstitutionModelParser.java
*
* Copyright (C) 2002-2012 Alexei Drummond, Andrew Rambaut & Marc A. 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.app.beagle.evomodel.parsers;
import dr.app.beagle.evomodel.substmodel.ComplexSubstitutionModel;
import dr.app.beagle.evomodel.substmodel.FrequencyModel;
import dr.app.beagle.evomodel.substmodel.SVSComplexSubstitutionModel;
import dr.app.beagle.evomodel.substmodel.SubstitutionModel;
import dr.evolution.datatype.DataType;
import dr.inference.model.BayesianStochasticSearchVariableSelection;
import dr.inference.model.Parameter;
import dr.xml.*;
import java.util.logging.Logger;
/**
* @author Marc Suchard
*/
public class ComplexSubstitutionModelParser extends AbstractXMLObjectParser {
public static final String COMPLEX_SUBSTITUTION_MODEL = "complexSubstitutionModel";
public static final String SVS_COMPLEX_SUBSTITUTION_MODEL = "svsComplexSubstitutionModel";
public static final String DATA_TYPE = "dataType";
public static final String RATES = "rates";
public static final String FREQUENCIES = "frequencies";
public static final String ROOT_FREQUENCIES = "rootFrequencies";
public static final String RANDOMIZE = "randomizeIndicator";
public static final String INDICATOR = "rateIndicator";
public static final String BSSVS_TOLERANCE = "bssvsTolerance";
public static final String BSSVS_SCALAR = "bssvsScalar";
public static final int maxRandomizationTries = 100;
public String getParserName() {
return COMPLEX_SUBSTITUTION_MODEL;
}
public String[] getParserNames() {
return new String[] {COMPLEX_SUBSTITUTION_MODEL, SVS_COMPLEX_SUBSTITUTION_MODEL};
}
public Object parseXMLObject(XMLObject xo) throws XMLParseException {
Parameter ratesParameter;
XMLObject cxo;
if (xo.hasChildNamed(FREQUENCIES)) {
cxo = xo.getChild(FREQUENCIES);
} else {
cxo = xo.getChild(ROOT_FREQUENCIES);
}
FrequencyModel freqModel = (FrequencyModel) cxo.getChild(FrequencyModel.class);
DataType dataType = freqModel.getDataType();
cxo = xo.getChild(RATES);
int states = dataType.getStateCount();
Logger.getLogger("dr.app.beagle.evomodel").info(" Complex Substitution Model (stateCount=" + states + ")");
ratesParameter = (Parameter) cxo.getChild(Parameter.class);
int rateCount = (dataType.getStateCount() - 1) * dataType.getStateCount();
if (ratesParameter == null) {
if (rateCount == 1) {
// simplest model for binary traits...
} else {
throw new XMLParseException("No rates parameter found in " + getParserName());
}
} else if (ratesParameter.getDimension() != rateCount) {
throw new XMLParseException("Rates parameter in " + getParserName() + " element should have " + rateCount + " dimensions.");
}
if (!xo.hasChildNamed(INDICATOR)) {
return new ComplexSubstitutionModel(COMPLEX_SUBSTITUTION_MODEL,dataType, freqModel, ratesParameter);
}
cxo = xo.getChild(INDICATOR);
Parameter indicatorParameter = (Parameter) cxo.getChild(Parameter.class);
if (indicatorParameter == null || ratesParameter == null || indicatorParameter.getDimension() != ratesParameter.getDimension())
throw new XMLParseException("Rates and indicator parameters in " + getParserName() + " element must be the same dimension.");
if (xo.hasAttribute(BSSVS_TOLERANCE)) {
double tolerance = xo.getAttribute(BSSVS_TOLERANCE,
BayesianStochasticSearchVariableSelection.Utils.getTolerance());
if (tolerance > BayesianStochasticSearchVariableSelection.Utils.getTolerance()) {
// Only increase smallest allowed tolerance
BayesianStochasticSearchVariableSelection.Utils.setTolerance(tolerance);
Logger.getLogger("dr.app.beagle.evomodel").info("\tIncreasing BSSVS tolerance to " + tolerance);
}
}
if (xo.hasAttribute(BSSVS_SCALAR)) {
double scalar = xo.getAttribute(BSSVS_SCALAR,
BayesianStochasticSearchVariableSelection.Utils.getScalar());
if (scalar < BayesianStochasticSearchVariableSelection.Utils.getScalar()) {
BayesianStochasticSearchVariableSelection.Utils.setScalar(scalar);
Logger.getLogger("dr.app.beagle.evomodel").info("\tDecreasing BSSVS scalar to " + scalar);
}
}
SVSComplexSubstitutionModel model = new SVSComplexSubstitutionModel(SVS_COMPLEX_SUBSTITUTION_MODEL,dataType, freqModel, ratesParameter, indicatorParameter);
boolean randomize = xo.getAttribute(RANDOMIZE, false);
if (randomize) {
// Randomization may need multiple tries
int tries = 0;
boolean valid = false;
while (!valid && tries < maxRandomizationTries) {
BayesianStochasticSearchVariableSelection.Utils.randomize(indicatorParameter,
dataType.getStateCount(),false);
valid = !Double.isInfinite(model.getLogLikelihood());
tries++;
}
Logger.getLogger("dr.app.beagle.evomodel").info("\tRandomization attempts: " + tries);
}
Logger.getLogger("dr.app.beagle.evomodel").info("\t\tPlease cite: Edwards, Suchard et al. (2011)\n");
return model;
}
//************************************************************************
// AbstractXMLObjectParser implementation
//************************************************************************
public String getParserDescription() {
return "A general irreversible model of sequence substitution for any data type.";
}
public Class getReturnType() {
return SubstitutionModel.class;
}
public XMLSyntaxRule[] getSyntaxRules() {
return rules;
}
private final XMLSyntaxRule[] rules = {
new XORRule(
new StringAttributeRule(DataType.DATA_TYPE, "The type of sequence data",
DataType.getRegisteredDataTypeNames(), false),
new ElementRule(DataType.class),
true // Optional
),
AttributeRule.newBooleanRule(RANDOMIZE,true),
new XORRule(
new ElementRule(FREQUENCIES,FrequencyModel.class),
new ElementRule(ROOT_FREQUENCIES,FrequencyModel.class)),
new ElementRule(RATES,
new XMLSyntaxRule[]{
new ElementRule(Parameter.class, true)}
),
new ElementRule(INDICATOR,
new XMLSyntaxRule[]{
new ElementRule(Parameter.class)
},true),
AttributeRule.newDoubleRule(BSSVS_TOLERANCE, true),
AttributeRule.newDoubleRule(BSSVS_SCALAR, true),
};
}