spatialFastmap-methods: Spatially-aware FastMap projection

In Cardinal: A mass spectrometry imaging toolbox for statistical analysis

Description

Performs spatially-aware FastMap projection.

Usage

## S4 method for signature 'SparseImagingExperiment'
spatialFastmap(x, r = 1, ncomp = 3,
    method = c("gaussian", "adaptive"),
    metric = c("average", "correlation", "neighborhood"),
    dist = "chebyshev", tol.dist = 1e-9,
    iter.max = 1, BPPARAM = getCardinalBPPARAM(), ...)

## S4 method for signature 'SpatialFastmap2'
summary(object, ...)

## S4 method for signature 'SImageSet'
spatialFastmap(x, r = 1, ncomp = 3,
    method = c("gaussian", "adaptive"),
    metric = c("average", "correlation", "neighborhood"),
    iter.max = 1, ...)

Arguments

x	The imaging dataset for which to calculate the FastMap components.
r	The neighborhood spatial smoothing radius.
ncomp	The number of FastMap components to calculate.
method	The method to use to calculate the spatial smoothing kernels for the embedding. The 'gaussian' method refers to spatially-aware (SA) weights, and 'adaptive' refers to spatially-aware structurally-adaptive (SASA) weights.
metric	The dissimilarity metric to use when comparing spectra, where ‘average’ (the default) means to use the differences of spatially-smoothed spectra, ‘correlation’ means to use the correlations of spatially-smoothed spectra, and ‘neighborhood’ means to use pairwise differences of each spectrum in the neighborhoods. Previous versions used ‘neighborhood’, which is the algorithm of Alexandrov & Kobarg; ‘average’ is the current default, because it handles non-gridded pixels better than ‘neighborhood’.
dist	The type of distance metric to use when calculating neighboring pixels based on r. The options are ‘radial’, ‘manhattan’, ‘minkowski’, and ‘chebyshev’ (the default).
tol.dist	The distance tolerance used for matching pixels when calculating pairwise distances between neighborhoods. This parameter should only matter when the data is not gridded. (Only considers ‘radial’ distance.)
iter.max	The number of iterations to perform when choosing the pivot vectors for each dimension.
...	Ignored.
object	A fitted model object to summarize.
BPPARAM	An optional instance of BiocParallelParam. See documentation for bplapply.

Value

An object of class SpatialFastmap2, which is a ImagingResult, or an object of class SpatialFastmap, which is a ResultSet. Each element of the resultData slot contains at least the following components:

scores:

A matrix with the FastMap component scores.

correlation:

A matrix with the feature correlations with each FastMap component.

sdev:

The standard deviations of the FastMap scores.

Author(s)

Kylie A. Bemis

References

Alexandrov, T., & Kobarg, J. H. (2011). Efficient spatial segmentation of large imaging mass spectrometry datasets with spatially aware clustering. Bioinformatics, 27(13), i230-i238. doi:10.1093/bioinformatics/btr246

Faloutsos, C., & Lin, D. (1995). FastMap: A Fast Algorithm for Indexing, Data-Mining and Visualization of Traditional and Multimedia Datasets. Presented at the Proceedings of the 1995 ACM SIGMOD international conference on Management of data.

See Also

PCA, spatialKMeans, spatialShrunkenCentroids

Examples

setCardinalBPPARAM(SerialParam())

set.seed(1)
data <- simulateImage(preset=2, npeaks=20, dim=c(6,6),
    representation="centroid")

# project to FastMap components
fm <- spatialFastmap(data, r=1, ncomp=2, method="adaptive")

# visualize first 2 components
image(fm, superpose=FALSE)

Cardinal documentation built on Nov. 8, 2020, 11:10 p.m.