R/map_diagnostics_bootstrap_blobs.R
bootstrapBlobs.RdThis function takes a map for which the function bootstrapMap() has already
been applied and draws contour blobs for each point illustrating how point
position varies in each bootstrap repeat. The blobs are calculated using
kernal density estimates according to these point distribution and drawn
so as to encompass a given proportion of this variation according to the
parameter conf.level. A conf.level set at 0.95 for example will draw
blobs that are calculated to encompass 95% of the positional variation seen
in the bootstrap repeats. Note however that the accuracy of these estimates
will depend on the number of bootstrap repeats performed, for example whether
100 or 1000 repeats were performed in the initial calculations using
bootstrapMap().
bootstrapBlobs(
map,
conf.level = 0.68,
smoothing = 6,
gridspacing = 0.25,
antigens = TRUE,
sera = TRUE,
method = "ks"
)The acmap data object
The proportion of positional variation captured by each blob
The amount of smoothing to perform when performing the kernel density estimate, larger equates to more smoothing
grid spacing to use when calculating blobs, smaller values will produce more accurate blobs with smoother edges but will take longer to calculate.
Should blobs be calculated for antigens
Should blobs be calculated for sera
One of "MASS", the default, or "ks", specifying the algorithm to use when calculating blobs in 2D. 3D will always use ks::kde.
Returns an acmap object that will then show the corresponding bootstrap blobs when viewed or plotted.
Other map diagnostic functions:
agCohesion(),
bootstrapMap(),
checkHemisphering(),
dimensionTestMap(),
logtiterTable(),
map-table-distances,
mapBootstrapCoords,
mapDistances(),
mapRelaxed(),
mapResiduals(),
pointStress,
ptBootstrapBlob,
ptBootstrapCoords(),
ptLeverage,
ptTriangulationBlob,
recalculateStress(),
stressTable(),
tableColbases(),
tableDistances(),
triangulationBlobs(),
unstableMaps