/*
* #%L
* gitools-core
* %%
* Copyright (C) 2013 Universitat Pompeu Fabra - Biomedical Genomics group
* %%
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as
* published by the Free Software Foundation, either version 3 of the
* License, or (at your option) any later version.
*
* This program 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public
* License along with this program. If not, see
* <http://www.gnu.org/licenses/gpl-3.0.html>.
* #L%
*/
package org.gitools.analysis.groupcomparison;
import org.gitools.analysis.AnalysisProcessor;
import org.gitools.analysis.groupcomparison.dimensiongroups.DimensionGroup;
import org.gitools.analysis.groupcomparison.format.math33Preview.CombinatoricsUtils;
import org.gitools.analysis.stats.mtc.MTCFactory;
import org.gitools.analysis.stats.test.EnrichmentTest;
import org.gitools.analysis.stats.test.MannWhitneyWilcoxonTest;
import org.gitools.analysis.stats.test.OneWayAnovaTest;
import org.gitools.analysis.stats.test.results.GroupComparisonResult;
import org.gitools.analysis.stats.test.results.OneWayAnovaResult;
import org.gitools.api.analysis.IProgressMonitor;
import org.gitools.api.matrix.*;
import org.gitools.api.resource.ResourceReference;
import org.gitools.heatmap.Heatmap;
import org.gitools.heatmap.HeatmapLayer;
import org.gitools.heatmap.decorator.impl.PValueDecorator;
import org.gitools.heatmap.decorator.impl.PValueLogSumDecorator;
import org.gitools.heatmap.header.ColoredLabel;
import org.gitools.heatmap.header.HeatmapColoredLabelsHeader;
import org.gitools.heatmap.header.HeatmapHeader;
import org.gitools.matrix.model.hashmatrix.HashMatrix;
import org.gitools.matrix.model.hashmatrix.HashMatrixDimension;
import org.gitools.matrix.model.matrix.AnnotationMatrix;
import org.gitools.matrix.model.matrix.element.LayerAdapter;
import org.gitools.matrix.model.matrix.element.MapLayerAdapter;
import org.gitools.utils.CloneUtils;
import org.gitools.utils.aggregation.SumAbsAggregator;
import org.gitools.utils.color.ColorGenerator;
import java.util.ArrayList;
import java.util.Date;
import java.util.Iterator;
import static org.gitools.api.matrix.MatrixDimensionKey.COLUMNS;
import static org.gitools.api.matrix.MatrixDimensionKey.ROWS;
public class GroupComparisonProcessor implements AnalysisProcessor {
private final GroupComparisonAnalysis analysis;
public GroupComparisonProcessor(GroupComparisonAnalysis analysis) {
this.analysis = analysis;
}
@Override
public void run(IProgressMonitor monitor) {
Date startTime = new Date();
// Prepare input data matrix
IMatrix dataMatrix = analysis.getData().get();
IMatrixLayer<Double> layer = dataMatrix.getLayers().get(analysis.getLayerName());
// Prepare dimensions to compare
IMatrixDimension sourceRows = (analysis.isTransposeData() ? dataMatrix.getDimension(COLUMNS) : dataMatrix.getDimension(ROWS));
IMatrixDimension sourceColumns = (analysis.isTransposeData() ? dataMatrix.getDimension(ROWS) : dataMatrix.getDimension(COLUMNS));
EnrichmentTest test = analysis.getTest();
Heatmap resultHeatmap;
// Prepare group predicates
NullConversion nullConversionFunction = new NullConversion(analysis.getNullConversion());
if (test.getClass() == MannWhitneyWilcoxonTest.class) {
resultHeatmap = mannWhitneyWilcoxon(monitor, dataMatrix, layer, sourceRows, sourceColumns, (MannWhitneyWilcoxonTest) test, nullConversionFunction);
} else {
resultHeatmap = oneWayAnova(monitor, dataMatrix, layer, sourceRows, sourceColumns, (OneWayAnovaTest) test, nullConversionFunction);
}
// Finish
analysis.setStartTime(startTime);
analysis.setElapsedTime(System.currentTimeMillis() - startTime.getTime());
analysis.setResults(new ResourceReference<>("results", resultHeatmap));
}
private Heatmap oneWayAnova(IProgressMonitor monitor, IMatrix dataMatrix, IMatrixLayer<Double> layer, IMatrixDimension sourceRows, IMatrixDimension sourceColumns, OneWayAnovaTest test, NullConversion nullConversionFunction) {
LayerAdapter<OneWayAnovaResult> adapter = new LayerAdapter<>(OneWayAnovaResult.class);
DimensionGroup[] groups = analysis.getGroups().toArray(new DimensionGroup[analysis.getGroups().size()]);
ArrayList<String> resultColnames = new ArrayList<>();
resultColnames.add("oneWayAnova");
HashMatrixDimension resultColumns = new HashMatrixDimension(COLUMNS, resultColnames);
HashMatrixDimension resultsRows = new HashMatrixDimension(ROWS, sourceRows);
Heatmap resultHeatmap = new Heatmap(
new HashMatrix(
adapter.getMatrixLayers(),
resultsRows,
resultColumns
)
);
// Run comparison
dataMatrix.newPosition()
.iterate(sourceRows)
.monitor(monitor, "Running group comparison analysis")
.transform(
new OneWayAnovaFunction(
test,
sourceColumns,
layer,
nullConversionFunction,
groups)
)
.store(resultHeatmap, new MapLayerAdapter<>(resultColumns, adapter));
// Run multiple test correction
IMatrixPosition position = resultHeatmap.newPosition();
IMatrixFunction<Double, Double> mtcFunction = MTCFactory.createFunction(analysis.getMtc());
for (String condition : position.iterate(resultColumns)) {
// Two-tail p-Value
position.iterate(adapter.getLayer(Double.class, "two-tail-p-value"), resultsRows)
.transform(mtcFunction)
.store(resultHeatmap, adapter.getLayer(Double.class, "corrected-two-tail-p-value"));
}
// Results formatting
for (HeatmapLayer resultLayer : resultHeatmap.getLayers()) {
if (resultLayer.getId().contains("p-value")) {
resultLayer.setDecorator(new PValueDecorator());
}
}
resultHeatmap.setTitle(analysis.getTitle() + " (results)");
if (analysis.isCopyAnnotation() && dataMatrix instanceof Heatmap &&
((Heatmap) dataMatrix).getRows().getAnnotations() != null) {
resultHeatmap.getRows().addAnnotations(
((Heatmap) dataMatrix).getRows().getAnnotations()
);
for (HeatmapHeader header : ((Heatmap) dataMatrix).getRows().getHeaders()) {
HeatmapHeader headerClone = null;
try {
headerClone = CloneUtils.clone(header);
} catch (Exception e) {
e.printStackTrace();
}
resultHeatmap.getRows().addHeader(headerClone);
headerClone.init(resultHeatmap.getRows());
}
}
return resultHeatmap;
}
private Heatmap mannWhitneyWilcoxon(IProgressMonitor monitor, IMatrix dataMatrix, IMatrixLayer<Double> layer, IMatrixDimension sourceRows, IMatrixDimension sourceColumns, MannWhitneyWilcoxonTest test, NullConversion nullConversionFunction) {
// Prepare results data matrix
LayerAdapter<GroupComparisonResult> adapter = new LayerAdapter<>(GroupComparisonResult.class);
DimensionGroup[] groups = analysis.getGroups().toArray(new DimensionGroup[analysis.getGroups().size()]);
ArrayList<String> resultColnames = new ArrayList<>();
AnnotationMatrix resultColumnAnnotations = new AnnotationMatrix();
Iterator<int[]> combIterator = CombinatoricsUtils.combinationsIterator(groups.length, 2);
while (combIterator.hasNext()) {
int[] groupIndices = combIterator.next();
DimensionGroup dimensionGroup1 = groups[groupIndices[0]];
DimensionGroup dimensionGroup2 = groups[groupIndices[1]];
String combName = dimensionGroup1.getName() + " VS " + dimensionGroup2.getName();
resultColnames.add(combName);
resultColumnAnnotations.setAnnotation(combName, "Group 1", dimensionGroup1.getName());
resultColumnAnnotations.setAnnotation(combName, "Group 2", dimensionGroup2.getName());
}
HashMatrixDimension resultColumns = new HashMatrixDimension(COLUMNS, resultColnames);
HashMatrixDimension resultsRows = new HashMatrixDimension(ROWS, sourceRows);
Heatmap resultHeatmap = new Heatmap(
new HashMatrix(
adapter.getMatrixLayers(),
resultsRows,
resultColumns
)
);
resultHeatmap.getColumns().addAnnotations(resultColumnAnnotations);
// Run comparison
dataMatrix.newPosition()
.iterate(sourceRows)
.monitor(monitor, "Running group comparison analysis")
.transform(
new GroupComparisonFunction(
test,
sourceColumns,
layer,
nullConversionFunction,
groups)
)
.store(resultHeatmap, new MapLayerAdapter<>(resultColumns, adapter));
// Run multiple test correction
IMatrixPosition position = resultHeatmap.newPosition();
IMatrixFunction<Double, Double> mtcFunction = MTCFactory.createFunction(analysis.getMtc());
for (String condition : position.iterate(resultColumns)) {
// Left p-Value
position.iterate(adapter.getLayer(Double.class, "left-p-value"), resultsRows)
.transform(mtcFunction)
.store(resultHeatmap, adapter.getLayer(Double.class, "corrected-left-p-value"));
// Right p-Value
position.iterate(adapter.getLayer(Double.class, "right-p-value"), resultsRows)
.transform(mtcFunction)
.store(resultHeatmap, adapter.getLayer(Double.class, "corrected-right-p-value"));
// Two-tail p-Value
position.iterate(adapter.getLayer(Double.class, "two-tail-p-value"), resultsRows)
.transform(mtcFunction)
.store(resultHeatmap, adapter.getLayer(Double.class, "corrected-two-tail-p-value"));
}
// Results formatting
for (HeatmapLayer resultLayer : resultHeatmap.getLayers()) {
if (resultLayer.getId().contains("p-value")) {
resultLayer.setDecorator(new PValueDecorator());
}
resultHeatmap.getLayers().get("p-value-log-sum").setDecorator(
new PValueLogSumDecorator()
);
}
resultHeatmap.getLayers().setTopLayerById("p-value-log-sum");
resultHeatmap.setTitle(analysis.getTitle() + " (results)");
if (analysis.isCopyAnnotation() && dataMatrix instanceof Heatmap &&
((Heatmap) dataMatrix).getRows().getAnnotations() != null) {
resultHeatmap.getRows().addAnnotations(
((Heatmap) dataMatrix).getRows().getAnnotations()
);
for (HeatmapHeader header : ((Heatmap) dataMatrix).getRows().getHeaders()) {
HeatmapHeader headerClone = null;
try {
headerClone = CloneUtils.clone(header);
} catch (Exception e) {
e.printStackTrace();
}
resultHeatmap.getRows().addHeader(headerClone);
headerClone.init(resultHeatmap.getRows());
}
}
// generate colors for annation
ColorGenerator cg1 = new ColorGenerator();
ColorGenerator cg2 = new ColorGenerator();
ArrayList<ColoredLabel> group1List = new ArrayList<ColoredLabel>();
ArrayList<ColoredLabel> group2List = new ArrayList<ColoredLabel>();
for (DimensionGroup g : analysis.getGroups()) {
group1List.add(new ColoredLabel(g.getName(), cg1.next(g.getName())));
group2List.add(new ColoredLabel(g.getName(), cg2.next(g.getName())));
}
resultHeatmap.getColumns().setCellSize(40);
resultHeatmap.getLayers().get("p-value-log-sum").setSortDirection(SortDirection.DESCENDING);
resultHeatmap.getLayers().get("p-value-log-sum").setAggregator(SumAbsAggregator.INSTANCE);
HeatmapColoredLabelsHeader group1Header = new HeatmapColoredLabelsHeader(resultHeatmap.getColumns());
group1Header.setClusters(group1List);
group1Header.setAnnotationPattern("${Group 1}");
group1Header.setTitle("Group 1");
HeatmapColoredLabelsHeader group2Header = new HeatmapColoredLabelsHeader(resultHeatmap.getColumns());
group2Header.setClusters(group2List);
group2Header.setAnnotationPattern("${Group 2}");
group2Header.setTitle("Group 2");
resultHeatmap.getColumns().addHeader(group1Header);
resultHeatmap.getColumns().addHeader(group2Header);
return resultHeatmap;
}
}