CFITIntegrate: Integration of multiple data source.
In pengminshi/cFIT: cFIT (common Factor Space Integration & Transfer)

Solve the model parameters through Iterative Nonnegative Matrix Factorization (iNMF), by minimizing the objective function

1/N ∑_j||X_j -(H_jW^TΛ_j + 1_{n_j} b_j^T)||_F^2

with penalties.

CFITIntegrate(
  X.list,
  r = 15,
  max.niter = 100,
  tol = 1e-05,
  nrep = 1,
  init = NULL,
  future.plan = c("sequential", "transparent", "multicore", "multisession", "cluster"),
  workers = parallel::detectCores() - 1,
  verbose = T,
  seed = 0
)

`X.list`	a list of m ncells-by-ngenes, gene expression matrices from m data sets
`r`	scalar, dimension of common factor matrix, which can be chosen as the rough number of identifiable cells types in the joint population (default 15).
`max.niter`	integer, max number of iterations (default 100).
`tol`	numeric scalar, tolerance used in stopping criteria (default 1e-5).
`nrep`	integer, number of repeated runs (to reduce effect of local optimum, default 1)
`init`	a list of parameters for parameter initialization. The list either contains all parameter sets: W,lambda.list, b.list, H.list, or only W will be used if provided (default NULL).
`future.plan`	plan for future parallel computation, can be chosen from 'sequential','transparent','multicore','multisession' and 'cluster'. Default is 'sequential'. Note that Rstudio does not support 'multicore'.
`workers`	additional parameter for `future::plan()`, in cases of 'multicore','multisession' and 'cluster'.
`verbose`	boolean scalar, whether to show extensive program logs (default TRUE)
`seed`	random seed used (default 0)

a list containing

W: ngenes-by-r numeric matrix, estimated common factor matrix
H.list: A list of m factor loading matrix of size ncells-by-r, estimated factor loading matrices
b.list: A list of estimated shift vector of size p (ngenes).
lambda.list: A list of estimated scaling vector of size p (ngenes).
convergence: boolean, whether the algorithm converge
obj: numeric scalar, value of the objective function at convergence or when maximum iteration achieved
niter: integer, the iteration at convergence (or maximum iteration if not converge)
delta: numeric scalar, the relative difference in last objective update
params: list of parameters used for the algorithm: max.iter, tol, nrep