Mercator (version 1.1.6)

downsample: Downsampling a Distance Visualization to Facilitate iGraph

Description

The downsample function implements a structured reduction of data points with a parent Mercator distance visualization to improve visualization and computational time, especially for the implementation of the iGraph visualization.

Usage

downsample(target, distanceMat, cutoff)

Value

downsample returns an object of class Mercator containing a structured subset of items from the parent Mercator object.

Arguments

target

An integer number of points to which the user wishes to reduce the parent Mercator object.

distanceMat

An object of class matrix containing the distance matrix component of the parent Mercator object.

cutoff

An inclusion cutoff for selected points based on the local density within the parent data.

Author

Kevin R. Coombes <[e-mail]>, Caitlin E. Coombes

Details

Mercator can be used to visualize complex networks using iGraph. To improve clarity of the visualization and computational time, we implement the downsample function to reduce the number of data points to be linked and visualized. The conceptual grounding for downsample draws on Peng Qiu's implementation of the SPADE clustering algorithm for mass cytometry data. The downsample function under-samples the densest regions of the data space to make it more likely that rarer clusters will still be adequately sampled.

References

Qiu, P., et. al. (2011). Extracting a cellular hierarchy from high-dimensional cytometry data with SPADE. Nature biotechnology, 29(10), 886.

Examples

#Form a BinaryMatrix
data("iris")
my.data <- as.matrix(iris[,c(1:4)])
my.rows <- as.data.frame(c(1:length(my.data[,1])))
my.binmat <- BinaryMatrix(my.data, , my.rows)
my.binmat <- t(my.binmat)
summary(my.binmat)

# Form and plot Mercator object
# Set K to the known number of species in the dataset
my.vis <- Mercator(my.binmat, "euclid", "tsne", K=3)
summary(my.vis)
plot(my.vis, view = "tsne", main="t-SNE plot of all data points")

#Downsample the Mercator object
M <- as.matrix(my.vis@distance)
set.seed(21340)
DS <- downsample(50, M, 0.1)
red.vis <- my.vis[DS]

#Visualize the down sampled t-SNE plot
plot(red.vis, view = "tsne", main="Down sampled t-SNE Plot")