splitMADOutlier {discordant} | R Documentation |
Identify features with outliers using left and right median absolute deviation (MAD).
splitMADOutlier(mat, filter0 = TRUE, threshold = 2)
mat |
mxn matrix of -omics data, where rows are features and columns samples. |
filter0 |
Option to filter out features if they have at least one 0 value. Default is TRUE. |
threshold |
Threshold of how many MADs outside the left or right median is used to determine features with outliers. |
The purpose of this function is to determine outliers in non-symmetric distributions. The distribution is split by the median. Outliers are identifed by being however many median absolute deviations (MAD) from either split distribution.
mat.filtered |
Input matrix where features with outliers filtered out. |
index |
Index of features that have no outliers. |
Leys C, Klein O, Bernard P and Licata L. "Detecting Outliers: Do Not Use Standard Deviation Around the Mean, Use Absolute Deivation Around the Median." Journal of Experimental Social Psychology, 2013. 49(4), 764-766. Magwene, PM, Willis JH, Kelly JK and Siepel A. "The Statistics of Bulk Segregant Analysis Using Next Generation Sequencing." PLoS Computational Biology, 2011. 7(11), e1002255.
## Simulate matrix of continuous -omics data. data(TCGA_Breast_miRNASeq) ## Filter matrix based on outliers. mat.filtered <- splitMADOutlier(TCGA_Breast_miRNASeq)$mat.filtered