assignR (version 2.2.3)

qtlRaster: Sample assignment using thresholds

Description

Selects the grid cells of probability density rasters with the highest probability and returns rasters with these cell values set to 1. Cells are selected based on the user-specified quantile threshold so that the most-probable cells representing a given fraction of the assignment area or posterior probability are returned.

Usage

qtlRaster(pdR, threshold, thresholdType = "area", genplot = TRUE, outDir = NULL)

Value

RasterStack or RasterLayer including a binary assignment surface for each individual in pdR. If outDir is not NULL, writes individual rasters in GeoTIFF format and a single PDF file with images for each raster to the designated directory.

Arguments

pdR

RasterLayer, RasterStack, or RasterBrick. Probability density maps for individual samples, e.g., as output by pdRaster.

threshold

numeric from 0 to 1. Quantile to be selected.

thresholdType

character. Either “area” (default) or “prob”. If “area”, the most probable cells constituting threshold percent of the assignment area are selected. If “prob”, the most probable cells constituting threshold percent of the posterior probability are selected.

genplot

logical.Plot results in R.

outDir

character string. Directory path to which output will be saved. If NULL no files are written.

Examples

# load North America boundary and global isoscape
data("naMap")
data("d2h_lrNA")

# load hydrogen isotope data for human hair in North America
d = subOrigData(group = "Modern human", mask = naMap, niter = 100)

# rescale from environmental isoscape to tissue isoscape
r = calRaster(known = d, isoscape = d2h_lrNA, mask = naMap)

# four unknown-origin examples
id = c("A", "B", "C", "D")
d2H = c(-110, -90, -105, -102)
un = data.frame(id, d2H)

# posterior probability surface
asn = pdRaster(r, unknown=un, mask=naMap)

# assign to most probable 10 percent of area
if (FALSE) qtlRaster(asn, threshold = 0.1)

# assign to most probable 10 percent of proabability distribution
qtlRaster(asn, threshold = 0.1, thresholdType = "prob")