Affine transformations

Apply an affine transformation matrix to a spatial object. Currently only works for 2D transforms.

Usage

# S4 method for class 'giotto,matrix'
affine(
  x,
  y,
  inv = FALSE,
  spat_unit = ":all:",
  feat_type = ":all:",
  images = ":all:",
  ...
)

# S4 method for class 'missing,missing'
affine(x)

# S4 method for class 'ANY,missing'
affine(x)

# S4 method for class 'ANY,affine2d'
affine(x, y, ...)

# S4 method for class 'SpatVector,matrix'
affine(x, y, inv = FALSE, pre_multiply = FALSE, ...)

# S4 method for class 'giottoPoints,matrix'
affine(x, y, inv = FALSE, pre_multiply = FALSE, ...)

# S4 method for class 'giottoPolygon,matrix'
affine(x, y, inv = FALSE, pre_multiply = FALSE, ...)

# S4 method for class 'spatLocsObj,matrix'
affine(x, y, inv = FALSE, pre_multiply = FALSE, ...)

# S4 method for class 'giottoLargeImage,matrix'
affine(x, y, inv = FALSE, pre_multiply = FALSE, ...)

# S4 method for class 'giottoAffineImage,matrix'
affine(x, y, inv = FALSE, pre_multiply = FALSE, ...)

# S4 method for class 'affine2d,matrix'
affine(x, y, inv = FALSE, pre_multiply = FALSE, ...)

Arguments

x

object to affine transform or a matrix

y

matrix or coercible to matrix (such as affine2d). Should be a matrix with either 2 or 3 columns (linear or affine).

inv

logical. Whether the inverse of the affine transform should be applied.

spat_unit

character vector. spatial units to affect. The :all: token to affect all can be used.

feat_type

character vector. feature types to affect. The :all: token to affect all can be used.

images

character vector. Images to affect. The :all: token to affect all can be used.

...

additional args to pass (none implemented)

pre_multiply

logical. Either pre: t(A %*% t(xy)) or post: xy %*% A convention to use when applying the affine matrix. The general convention is to set TRUE, but Giotto internally uses FALSE (this may be updated in the future).

Value

affine transformed object or an affine2d if a matrix was passed to x

Examples

m <- diag(rep(1, 3))
trans_m <- matrix(c(1, 0, 0, 0, 1, 0, 200, 300, 1), nrow = 3)
scale_m <- matrix(c(2, 0, 0, 0, 3, 0, 0, 0, 1), nrow = 3)
aff_m <- matrix(c(2, 3, 0, 0.2, 3, 0, 100, 29, 1), nrow = 3)

gpoints <- GiottoData::loadSubObjectMini("giottoPoints")
gpoly <- GiottoData::loadSubObjectMini("giottoPolygon")
sl <- GiottoData::loadSubObjectMini("spatLocsObj")

# creation of affine2d
aff <- affine(m)
aff <- spin(flip(shear(aff, fx = 0.2)), 45)
plot(aff) # blue is start, red is end


# giottoPoints ##############################################
plot(gpoints)

plot(affine(gpoints, trans_m))


# giottoPolygon #############################################
plot(gpoly)

plot(affine(gpoly, scale_m))

plot(affine(gpoly, aff)) # affine() with `affine2d`


# spatLocsObj ###############################################
plot(affine(sl, m))

plot(affine(sl, trans_m))

plot(affine(sl, scale_m))

# this transformation can be inverted
aff_sl <- affine(sl, aff_m)
plot(aff_sl)

plot(affine(aff_sl, aff_m, inv = TRUE))