/*
* GeoSpatialDistribution.java
*
* Copyright (c) 2002-2012 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.geo;
import dr.inference.distribution.MultivariateDistributionLikelihood;
import dr.inference.model.Likelihood;
import dr.inference.model.Parameter;
import dr.math.distributions.MultivariateDistribution;
import dr.xml.*;
import java.awt.geom.Point2D;
import java.util.ArrayList;
import java.util.List;
import java.util.logging.Logger;
/**
* @author Marc A. Suchard
* @author Philippe Lemey
* @author Alexei J. Drummond
*/
public class GeoSpatialDistribution implements MultivariateDistribution {
public static final String FLAT_SPATIAL_DISTRIBUTION = "flatGeoSpatialPrior";
public static final String DATA = "data";
public static final String TYPE = "geoSpatial";
public static final String NODE_LABEL = "taxon";
public static final String KML_FILE = "kmlFileName";
public static final String INSIDE = "inside";
public static final String UNION = "union";
private static final String DEFAULT_LABEL = "";
public static final int dimPoint = 2; // Assumes 2D points only
public GeoSpatialDistribution(String label) {
this.label = label;
}
public GeoSpatialDistribution(String label, Polygon2D region, boolean inside) {
this.label = label;
this.region = region;
this.outside = !inside;
}
public double logPdf(double[] x) {
final boolean contains = region.containsPoint2D(new Point2D.Double(x[0], x[1]));
if (outside ^ contains)
return 0;
return Double.NEGATIVE_INFINITY;
}
public double[][] getScaleMatrix() {
return null;
}
public double[] getMean() {
return null;
}
public String getType() {
return TYPE;
}
public String getLabel() {
return label;
}
public boolean getOutside() {
return outside;
}
public Polygon2D getRegion() {
return region;
}
protected Polygon2D region;
protected String label = null;
private boolean outside = false;
public static XMLObjectParser FLAT_GEOSPATIAL_PRIOR_PARSER = new AbstractXMLObjectParser() {
public String getParserName() {
return FLAT_SPATIAL_DISTRIBUTION;
}
public Object parseXMLObject(XMLObject xo) throws XMLParseException {
String label = xo.getAttribute(NODE_LABEL, DEFAULT_LABEL);
boolean inside = xo.getAttribute(INSIDE, true);
boolean union = xo.getAttribute(UNION, false);
boolean readFromFile = false;
List<GeoSpatialDistribution> geoSpatialDistributions = new ArrayList<GeoSpatialDistribution>();
if (xo.hasAttribute(KML_FILE)) {
// read file
String kmlFileName = xo.getStringAttribute(KML_FILE);
List<Polygon2D> polygons = Polygon2D.readKMLFile(kmlFileName);
for (Polygon2D region : polygons)
geoSpatialDistributions.add(new GeoSpatialDistribution(label, region, inside));
readFromFile = true;
} else {
for (int i = 0; i < xo.getChildCount(); i++) {
if (xo.getChild(i) instanceof Polygon2D) {
Polygon2D region = (Polygon2D) xo.getChild(i);
geoSpatialDistributions.add(
new GeoSpatialDistribution(label, region, inside)
);
}
}
}
List<Parameter> parameters = new ArrayList<Parameter>();
XMLObject cxo = xo.getChild(DATA);
for (int j = 0; j < cxo.getChildCount(); j++) {
Parameter spatialParameter = (Parameter) cxo.getChild(j);
parameters.add(spatialParameter);
}
if (geoSpatialDistributions.size() == 1) {
MultivariateDistributionLikelihood likelihood = new MultivariateDistributionLikelihood(geoSpatialDistributions.get(0));
for (Parameter spatialParameter : parameters) {
if (spatialParameter.getDimension() != dimPoint)
throw new XMLParseException("Spatial priors currently only work in " + dimPoint + "D");
likelihood.addData(spatialParameter);
}
return likelihood;
}
if (geoSpatialDistributions.size() == 0) {
throw new XMLParseException("Error constructing geo spatial distributions in " + xo.getId());
}
if (parameters.size() == 1) {
Parameter parameter = parameters.get(0);
if (parameter.getDimension() % dimPoint != 0)
throw new XMLParseException("Spatial priors currently only work in " + dimPoint + "D");
if (!label.equals(DEFAULT_LABEL)) { // For a tip-taxon
Logger.getLogger("dr.geo").info(
"\nConstructing a multiple-region spatial prior:\n" +
"\tTaxon: " + label + "\n" +
"\tNumber of regions: " + geoSpatialDistributions.size() + "\n\n");
MultivariateDistributionLikelihood likelihood = new MultivariateDistributionLikelihood(
new MultiRegionGeoSpatialDistribution(label, geoSpatialDistributions, union));
likelihood.addData(parameter);
return likelihood;
} else {
Logger.getLogger("dr.geo").info(
"\nConstructing a GeoSpatialCollectionModel:\n" +
"\tParameter: " + parameter.getId() + "\n" +
"\tNumber of regions: " + geoSpatialDistributions.size() + "\n\n");
return new GeoSpatialCollectionModel(xo.getId(), parameter, geoSpatialDistributions, !union);
}
}
throw new XMLParseException("Multiple separate parameters and multiple regions not yet implemented");
}
public XMLSyntaxRule[] getSyntaxRules() {
return rules;
}
private XMLSyntaxRule[] rules = new XMLSyntaxRule[]{
AttributeRule.newStringRule(NODE_LABEL, true),
AttributeRule.newBooleanRule(INSIDE, true),
AttributeRule.newBooleanRule(UNION, true),
new XORRule(
AttributeRule.newStringRule(KML_FILE),
new ElementRule(Polygon2D.class, 1, Integer.MAX_VALUE)
),
new ElementRule(DATA,
new XMLSyntaxRule[]{new ElementRule(Parameter.class, 1, Integer.MAX_VALUE)}
)
};
public String getParserDescription() {
return "Calculates the likelihood of some data under a 2D geospatial distribution.";
}
public Class getReturnType() {
return Likelihood.class;
}
};
}