sem_lucid: SEM for latent cluster estimation

In USCbiostats/LUCid: Latent Unknown Clustering with Integrated Data

Description

sem_lucid provides standard errors (SE) of parameter estimates when performing latent cluster analysis with multi-omics data. SEs are obtained through supplemented EM-algorithm (SEM).

Usage

sem_lucid(
  G = NULL,
  Z = NULL,
  Y,
  family = "binary",
  useY = TRUE,
  K = 2,
  initial = def_initial(),
  itr_tol = def_tol(),
  Pred = TRUE,
  Get_SE = TRUE,
  Ad_Hoc_SE = FALSE
)

Arguments

G	Genetic features, a matrix
Z	Biomarker data, a matrix
Y	Disease outcome, a vector
family	"binary" or "normal" for Y
useY	Using Y or not, default is TRUE
K	Pre-specified # of latent clusters, default is 2
initial	A list of initial model parameters will be returned for integrative clustering
itr_tol	A list of tolerance settings will be returned for integrative clustering
Pred	Flag to compute predicted disease probability with fitted model, boolean, default is TRUE
Get_SE	Flag to perform SEM to get SEs of parameter estimates, default is TRUE
Ad_Hoc_SE	Flag to fit ad hoc regression models to get SEs of parameter estimates, default is FALSE

Value

sem_lucid returns an object of list containing parameters estimates, their corresponding standard errors, and other features:

beta	Estimates of genetic effects, matrix
se_beta	SEM standard errors of Beta
se_ah_beta	Ad hoc standard errors of Beta
mu	Estimates of cluster-specific biomarker means, matrix
se_mu	SEM standard errors of Mu
se_ah_mu	Ad hoc standard errors of Mu
sigma	Estimates of cluster-specific biomarker covariance matrix, list
gamma	Estimates of cluster-specific disease risk, vector
se_gamma	SEM standard errors of Gamma
se_ah_gamma	Ad hoc standard errors of Gamma
pcluster	Probability of cluster, when G is null
pred	Predicted probability of belonging to each latent cluster

Author(s)

Cheng Peng, Zhao Yang, David V. Conti

References

Cheng Peng, Jun Wang, Isaac Asante, Stan Louie, Ran Jin, Lida Chatzi, Graham Casey, Duncan C Thomas, David V Conti, A Latent Unknown Clustering Integrating Multi-Omics Data (LUCID) with Phenotypic Traits, Bioinformatics, , btz667, https://doi.org/10.1093/bioinformatics/btz667.

Meng, X., & Rubin, D. B. (1991). Using EM to Obtain Asymptotic Matrices : The SEM Algorithm. Journal of the American Statistical Association, 86(416), 899-909. http://doi.org/10.2307/2290503

Examples

## Not run: 
sem_lucid(G=G2,Z=Z2,Y=Y2,useY=TRUE,K=2,Pred=TRUE,family="normal",Get_SE=TRUE,
            itr_tol = def_tol(MAX_ITR=1000,MAX_TOT_ITR=3000))

## End(Not run)

USCbiostats/LUCid documentation built on Feb. 22, 2020, 8:57 p.m.