Similarity-driven multiview linear reconstruction model (simlr) for two modalities

This function simplifies calculating image-wide multivariate beta maps from that is similar to CCA.

simlr2(
  voxmats,
  basisK,
  smoothingMatrixX,
  smoothingMatrixY,
  iterations = 10,
  gamma = 0.000001,
  sparsenessQuantileX = 0.5,
  sparsenessQuantileY = 0.5,
  positivityX = c("positive", "negative", "either"),
  positivityY = c("positive", "negative", "either"),
  initialUMatrix,
  orthogonalize = TRUE,
  repeatedMeasures = NA,
  verbose = FALSE
)

Arguments

voxmats	A list that contains the named x and y matrices.
basisK	an integer determining the size of the basis.
smoothingMatrixX	allows parameter smoothing, should be square and same size as input matrix on left side of equation
smoothingMatrixY	allows parameter smoothing, should be square and same size as input matrix on right side of equation
iterations	number of gradient descent iterations
gamma	step size for gradient descent
sparsenessQuantileX	quantile to control sparseness - higher is sparser
sparsenessQuantileY	quantile to control sparseness - higher is sparser
positivityX	restrict to positive or negative solution (beta) weights. choices are positive, negative or either as expressed as a string.
positivityY	restrict to positive or negative solution (beta) weights. choices are positive, negative or either as expressed as a string.
initialUMatrix	initialization matrix size n by k. If this is missing, a random matrix will be used.
orthogonalize	boolean to control whether we orthogonalize the solutions explicitly
repeatedMeasures	list of repeated measurement identifiers. this will allow estimates of per identifier intercept.
verbose	boolean to control verbosity of output

Value

A list of u, x and y-related matrices.

See also

milr mild

Examples

set.seed(1500)
nsub = 12
npix = 100
outcome = rnorm( nsub )
covar = rnorm( nsub )
mat = replicate( npix, rnorm( nsub ) )
mat2 = replicate( npix + 10, rnorm( nsub ) )
mat3 = replicate( npix + 10, rnorm( nsub ) )
nk = 3
result = simlr2( list( vox = mat, vox2 = mat2, vox3 = mat3 ), basisK = 3 )