MetaPCA: Meta-analysis in the Dimension Reduction of Genomic data

Share:

Description

Preprocessing for microarray meta-analysis. It is about gene filtering and missing value imputation.

Usage

1
PreprocessMetaAnalysis(DList, cutRatioByMean=.4, cutRatioByVar=.4, doImpute=FALSE, na.rm.pct=.1, na.rm.pct.each=.5, verbose=FALSE)

Arguments

DList

A list of all data matrices; Each data name should be set as the name of each list element. Each data should be a numeric matrix that has genes in the rows and samples in the columns. Row names should be official gene symbols and column names be sample labels.

cutRatioByMean

Proportion of genes filtered by study-wise mean. Default is 40%.

cutRatioByVar

Proportion of genes filtered by study-wise variance. Default is 40%.

doImpute

Whether to impute missing genes. Default is TRUE, and default imputation method is knn.

na.rm.pct

Proportion of genes filtered by study-wise missing proportion. Default is 10%.

na.rm.pct.each

Proportion of genes filtered by each study's missing proportion. Default is 50%.

verbose

Whether to print logs. Default is FALSE.

Value

list object of all data matrices after filtering and imputation.

Author(s)

Don Kang (donkang75@gmail.com) and George Tseng (ctseng@pitt.edu)

References

Dongwan D. Kang and George C. Tseng. (2011) Meta-PCA: Meta-analysis in the Dimension Reduction of Genomic data.

Examples

1
2
3
4
5
6
## Not run: 
DList <- PreprocessMetaAnalysis(list(Yu=Yu, Lapointe=Lapointe, Tomlins=Tomlins, Varambally=Varambally), 
		cutRatioByMean=.1, cutRatioByVar=.1, doImpute=T, na.rm.pct=.2)
str(DList)

## End(Not run)