Class ComBat<R,C>
- java.lang.Object
-
- ubic.gemma.core.analysis.preprocess.batcheffects.ComBat<R,C>
-
public class ComBat<R,C> extends Object
An implementation of the ComBat algorithm described by Johson et al as described in:Johnson, WE, Rabinovic, A, and Li, C (2007). Adjusting batch effects in microarray expression data using Empirical Bayes methods. Biostatistics 8(1):118-127.
-
-
Constructor Summary
Constructors Constructor Description ComBat(DoubleMatrix<R,C> data, ObjectMatrix<C,String,?> sampleInfo)
-
Method Summary
All Methods Instance Methods Concrete Methods Modifier and Type Method Description DoubleMatrix2D
getDesignMatrix()
void
plot(String filePrefix)
Make diagnostic plots.DoubleMatrix2D
run()
DoubleMatrix2D
run(boolean parametric)
-
-
-
Constructor Detail
-
ComBat
public ComBat(DoubleMatrix<R,C> data, ObjectMatrix<C,String,?> sampleInfo)
-
-
Method Detail
-
getDesignMatrix
public DoubleMatrix2D getDesignMatrix()
-
plot
public void plot(String filePrefix)
Make diagnostic plots. FIXME: As in the original ComBat, this only graphs the first batch's statistics. In principle we can (and perhaps should) examine these plots for all the batches.- Parameters:
filePrefix
- file prefix
-
run
public DoubleMatrix2D run() throws ubic.gemma.core.analysis.preprocess.batcheffects.ComBatException
- Returns:
- data corrected using parametric prior estimator
- Throws:
ubic.gemma.core.analysis.preprocess.batcheffects.ComBatException
- combat problems
-
run
public DoubleMatrix2D run(boolean parametric) throws ubic.gemma.core.analysis.preprocess.batcheffects.ComBatException
- Parameters:
parametric
- if false, use the non-parametric (slower) method for estimating the priors.- Returns:
- corrected data
- Throws:
ubic.gemma.core.analysis.preprocess.batcheffects.ComBatException
- combat problems
-
-