LUCIDus (version 2.0.0)

est.lucid: Estimate latent unknown clusters with multi-omics data

Description

This function estimates the latent clusters by integrating genetic features/environmental exposures, biomarkers with/without the outcome of interest. Variable selection is available for analyzing the high-dimensional data.

Usage

est.lucid(
  G,
  Z,
  Y,
  CoG = NULL,
  CoY = NULL,
  K = 2,
  family = "normal",
  useY = TRUE,
  control = def.control(),
  tune = def.tune(),
  Z.var.str = NULL
)

Arguments

G

Genetic features/environmental exposures, a matrix.

Z

Biomarkers/other omics data, a matrix.

Y

Disease outcome, it is suggested to transform it into a n by 1 matrix.

CoG

Optional, matrix. Covariates to be adjusted for estimating the latent cluster.

CoY

Optional, matrix. Covariates to be adjusted for estimating the outcome.

K

Number of latent clusters.

family

Type of outcome Y. It should be choose from "normal", "binary".

useY

Whether or not to include the information of Y to estimate the latent clusters. Default is TRUE.

control

A list of tolerance parameters used by EM algorithm. See def.control.

tune

A list of tuning parameters used by variable selection procedure. See def.tune

Z.var.str

The variance-covariance structure for the biomarkers. See mclustModelNames for details.

Value

A list which contains the several features of LUCID, including:

pars

Estimates of parameters of LUCID, including beta (estimates of genetic feature/environmental exposure), mu (estimates of cluster-specific biomarker means), sigma (estimates of the cluster-specific biomarker variance-covariance matrix) and gamma(estimates of cluster-specific effect and covariates effect related to the outcome)

K

Number of latent cluster

Z.var.str

The model used to estimate the cluster-specific variance-covariance matrix, for further details, see mclust

likelihood

The log likelihood of the LUCID model

post.p

Predicted probability of belonging to each latent cluster

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.

Examples

# NOT RUN {
set.seed(10)
fit1 <- est.lucid(G = G1,Z = Z1,Y = Y1, CoY = CovY, K = 2, family = "binary")
fit2 <- est.lucid(G = G1,Z = Z1,Y = Y1, CoY = CovY, K = 2, family = "binary", 
  tune = def.tune(Select_Z = TRUE, Rho_Z_InvCov = 0.1, Rho_Z_CovMu = 90, 
  Select_G = TRUE, Rho_G = 0.02))
# }