Istep_Z

Impute missing data in Z by maximizing the likelihood given fixed
parameters of LUCID

An implementation of the LUCID model (Peng (2019) <doi:10.1093/bioinformatics/btz667>).
LUCID conducts integrated clustering using exposures, omics data (and outcome
as an option). An EM algorithm is implemented to estimate MLE of the LUCID model.
LUCIDus features integrated variable selection, incorporation of missing omics
data, bootstrap inference, prediction and visualization of the model.

Yinqi Zhao

LUCIDus

Latent Unknown Clustering Integrating Multi-View Data

David Conti

Jesse Goodrich

Cheng Peng

Istep_Z function

<dl><dt>Z</dt>
<dd>an N by P matrix representing the omics data</dd>
<dt>p</dt>
<dd>an N by K matrix representing posterior inclusion probability for each 
latent cluster</dd>
<dt>mu</dt>
<dd>an M by K matrix representing cluster-specific means</dd>
<dt>sigma</dt>
<dd>an M by M by K array representing cluster-specific covariance</dd>
<dt>index</dt>
<dd>an N by M matrix representing missing values in Z</dd></dl>

Arguments

I-step of LUCID — Istep_Z

<dl>

<dt>Z</dt>
<dd>an N by P matrix representing the omics data</dd>


<dt>p</dt>
<dd>an N by K matrix representing posterior inclusion probability for each 
latent cluster</dd>


<dt>mu</dt>
<dd>an M by K matrix representing cluster-specific means</dd>


<dt>sigma</dt>
<dd>an M by M by K array representing cluster-specific covariance</dd>


<dt>index</dt>
<dd>an N by M matrix representing missing values in Z</dd>

</dl>

Istep_Z: I-step of LUCID

Description

Usage

Value

Arguments