LUCIDus: Integreted clustering with multi-view data Version 3.0.1

The LUCIDus package implements the statistical method LUCID proposed in the research paper A Latent Unknown Clustering Integrating Multi-Omics Data (LUCID) with Phenotypic Traits (Bioinformatics, 2020). LUCID conducts integrated clustering by using multi-view data, including exposures, and omics data with/without outcome. LUCIDus features variable selection, incorporating missingness in omics data, visualization of the LUCID model via Sankey diagram, bootstrap inference, and functions for tuning model parameters.

LUCID version 3.0.1, a major update and enhancement from the original release, implements different integration strategies for multi-omics data with multiple layers including LUCID early integration, LUCID in parallel, and LUCID in serial. It also incorporates methods to deal with missingness in multi-omics data. The following DAG illustrates the three different LUCID models for three integration strategies.

If you are interested in the integration of omic data to estimate mediator or latent structures, please check out Conti Lab to learn more.

Installation

You can install the development version of LUCIDus 3.0.1 from GitHub with:

# install.packages("devtools")
devtools::install_github("ContiLab-usc/LUCIDus-3.0",ref="main",auth_token = "xxx")

Note that this repo is now private, so only authorized users can download this package. Please go to tokens to obtain your personal authorized token and input it into auth_token = "xxx" to download this package.

Workflow

The following figure illustrate the workflow of LUCIDus 3.0.1.

Usage

Please refer to the tutorial.

Citation

#> 
#> To cite LUCID methods, please use:
#> 
#>   Cheng Peng, Jun Wang, Isaac Asante, Stan Louie, Ran Jin, Lida Chatzi,
#>   Graham Casey, Duncan C Thomas, David V Conti (2019). A latent unknown
#>   clustering integrating multi-omics data (LUCID) with phenotypic
#>   traits. Bioinformatics, btz667. URL
#>   https://doi.org/10.1093/bioinformatics/btz667
#> 
#> To cite LUCIDus R package, please use:
#> 
#>   Yinqi Zhao (2022). LUCIDus: an R package to implement the LUCID
#>   model. CRAN. R package version 2.2 URL
#>   https://yinqi93.github.io/LUCIDus/index.html
#> 
#> To see these entries in BibTeX format, use 'print(<citation>,
#> bibtex=TRUE)', 'toBibtex(.)', or set
#> 'options(citation.bibtex.max=999)'.