est_lucid: Estimating latent clusters with multi-omics data

Description Usage Arguments Value Author(s) References Examples

View source: R/est_lucid.R

Description

est_lucid estimates an integrated cluster assignment of genetic effects using complete biomarker data with/without disease outcomes. Options to produce sparse solutions for cluster-specific parameter estimates under a circumstance of analyzing high-dimensional data are also provided. An IntClust object will be produced.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
est_lucid(
  G = NULL,
  CoG = NULL,
  Z = NULL,
  Y,
  CoY = NULL,
  useY = TRUE,
  family = "binary",
  K = 2,
  Pred = TRUE,
  initial = def_initial(),
  itr_tol = def_tol(),
  tunepar = def_tune()
)

Arguments

G

Genetic features, a matrix

CoG

Covariates to be included in the G->X path

Z

Biomarker data, a matrix, can be incomplete and have missing values

Y

Disease outcome, a vector

CoY

Covariates to be included in the X->Y path

useY

Using Y or not, default is TRUE

family

"binary" or "normal" for Y

K

Pre-specified # of latent clusters, default is 2

Pred

Flag to compute posterior probability of latent cluster with fitted model, default is TRUE

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

tunepar

A list of tuning parameters and settings will be returned for integrative clustering

Value

est_lucid returns an object of list containing parameters estimates, predicted probability of latent clusters, and other features:

beta

Estimates of genetic effects, matrix

mu

Estimates of cluster-specific biomarker means, matrix

sigma

Estimates of cluster-specific biomarker covariance matrix, list

gamma

Estimates of cluster-specific disease risk, vector

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.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
# Integrative clustering without feature selection
set.seed(10)
IntClusFit <- est_lucid(G=G1,Z=Z1,Y=Y1,K=2,family="binary",Pred=TRUE)

## Not run: 
# Re-run the model with covariates in the G->X path
IntClusCoFit1 <- est_lucid(G=G1,CoG=CoG,Z=Z1,Y=Y1,K=2,family="binary",Pred=TRUE)

# Re-run the model with covariates in the X->Y path
IntClusCoFit2 <- est_lucid(G=G1,Z=Z1,Y=Y1,CoY=CoY,K=2,family="binary",Pred=TRUE)

# Re-run the model with covariates in both G->X and X->Y paths
IntClusCoFit3 <- est_lucid(G=G1,CoG=CoG,Z=Z1,Y=Y1,CoY=CoY,K=2,family="binary",Pred=TRUE)

# Model fit with incomplete biomarker data and covariates in both G->X & X->Y paths
IntClusCoFit3_Incomp <- est_lucid(G=G1,CoG=CoG,Z=Z1_Incomp,Y=Y1,CoY=CoY,K=2,family="binary")

## End(Not run)

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