Detect the outlier sample (or gene)

Description

Detect the outlier sample (or gene) based on distance to the cluster center

Usage

1
detectOutlier(x, metric = "euclidean", standardize = TRUE, Th = 2, ifPlot = FALSE)

Arguments

x

a LumiBatch object, ExpressionSet object or a matrix with each column corresponding to a sample or other profile

metric

the distance matric

standardize

standardize the profile or not

Th

the threshold of outlier,

ifPlot

to plot the result (as a hierarchical tree) or not

Details

The current outlier detection is based on the distance from the sample to the center (average of all samples after removing 10 percent samples farthest away from the center). The assumption of the outlier detection is that there is only one single cluster and the distance from the sample to the center is Gaussian distributed.

The outlier is detected when its distance to the center is larger than a certain threshold. The threshold is calculated as Th * median distances to the center.

The profile relations can be visualized as a hierarchical tree.

Value

Plot the results or return the outlier (a logic vector) with the distance matrix and threshold as attributes.

Author(s)

Pan Du

See Also

lumiQ

Examples

1
2
3
4
5
## load example data
data(example.lumi)

## detect the outlier (Further improvement needed.)
temp <- detectOutlier(example.lumi, ifPlot=TRUE)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.