sim1

A simulated dataset for integrated clustering with normal outcome. The data is simulated under cluster number K = 2.

datasets

An implementation for the 'LUCID' model (Peng (2019) <doi:10.1093/bioinformatics/btz667>) to jointly estimate latent unknown clusters/subgroups with integrated data.
An EM algorithm is used to obtain the latent cluster assignment and model parameter estimates.
Feature selection is achieved by applying the L1 regularization method.

Yinqi Zhao

LUCIDus

Latent Unknown Clustering with Integrated Data

sim1 function

A matrix of 22 columns, which are<dl class="dl-horizontal">
 <dt>G1 - G10</dt><dd>Genetic features, G1 to G5 are causal genes contributed to clustering, with OR = 2; G6 to G10 are null genes that is not related to clustering</dd>
 <dt>Z1 - Z10</dt><dd>Biomarkers, Z1 to Z5 are causal biomarkers with delta Z = 4 between 2 clusters, Z6 to Z10 are noises with delta Z = 0. All biomarkers are assumed to be independent with each other</dd>
 <dt>Y</dt><dd>Outcome of interest, which follows 2 normal distribution with N(-1, 1) and N(1, 1)</dd>
 <dt>X</dt><dd>Latent cluster assignment for each observation</dd>
</dl>

Format

simulated dataset 1 — sim1

A matrix of 22 columns, which are<dl class='dl-horizontal'>
 <dt>G1 - G10</dt><dd>Genetic features, G1 to G5 are causal genes contributed to clustering, with OR = 2; G6 to G10 are null genes that is not related to clustering</dd>
 <dt>Z1 - Z10</dt><dd>Biomarkers, Z1 to Z5 are causal biomarkers with delta Z = 4 between 2 clusters, Z6 to Z10 are noises with delta Z = 0. All biomarkers are assumed to be independent with each other</dd>
 <dt>Y</dt><dd>Outcome of interest, which follows 2 normal distribution with N(-1, 1) and N(1, 1)</dd>
 <dt>X</dt><dd>Latent cluster assignment for each observation</dd>
</dl>

sim1: simulated dataset 1

Description

Usage

Arguments

Format