fssemR (version 0.1.8)

opt.multiNFSSEMiPALM2: opt.multiNFSSEMiPALM2

Description

optimize multiNFSSEMiPALM's parameters by minimize BIC, when feature size is large (> 300), BIC methods will be much faster than Cross-validation

Usage

opt.multiNFSSEMiPALM2(
  Xs,
  Ys,
  Bs,
  Fs,
  Sk,
  sigma2,
  nlambda = 20,
  nrho = 20,
  p,
  q,
  wt = TRUE
)

Arguments

Xs

eQTL matrices

Ys

Gene expression matrices

Bs

initialized GRN-matrices

Fs

initialized eQTL effect matrices

Sk

eQTL index of genes

sigma2

initialized noise variance

nlambda

number of hyper-parameter of lasso term in CV

nrho

number of hyper-parameter of fused-lasso term in CV

p

number of genes

q

number of eQTLs

wt

use adaptive lasso or not. Default TRUE.

Value

list of model selection result