This repository contains R code to reproduce the results in the paper "Spike-and-Slab Lasso Biclustering" (Moran, Rockova and George, 2020).

The SSLB R package is available at the Github repository: https://github.com/gemoran/SSLB

Please follow the following instructions to run the code.

1. Install the SSLB R package using the package devtools by running the following code in your R GUI:

install.packages("devtools")
library(devtools)
install_github("gemoran/SSLB")

2. Install the following R packages used in simulations and data analysis:

install.packages(c("mvtnorm", "isa2", "biclust", "pals", "reshape2", "tidyverse", "gridExtra", "clue"))

if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install(c("breastCancerNKI", "preprocessCore", "impute", "fabia", "clusterProfiler", "org.Hs.eg.db", "enrichplot", "org.Mm.eg.db"))

3. We compare SSLB to the method BicMix (Gao et al. 2016). We downloaded the BicMix software from here. Follow the instructions in the BicMix software to create the BicMix executable file. Change bicmix_directory.txt to reflect where the executable file is relative to the folder SSLB_examples. The R code will then call the BicMix exectuable file from within the R session.

4. We also compare SSLB to the method SSBiEM (Denitto et al 2017). We downloaded the SSBiEM software from here. To run SSBiEM, place the files kronSpeye.m and SSBiEM.m in each of the folders: sim_study_1, sim_study_1, sim_study_1, sim_study_1, breastCancerNKI and zeisel. Then, to run SSBiEM, run the Matlab code in [folder-name]_ssbi.m.

• SSLB_functions.R: helper functions required for all of the R scripts

• K.R: this code creates the tables of number of biclusters estimated by each method for all of the simulation studies.

• sim_study_1: folder contains code to reproduce the results in Simulation 1.

  • sim_study_1_ssbi.m: this code runs the method SSBiEM
  • sim_study_1.R: this code runs all other biclustering methods.
  • plots.R: this code produces the consensus, relevance, recovery and variance explained plots
  • plot_matrices.R: this code plots the factor and loadings matrices found by each method for one dataset from the simulation study.

• sim_study_2: folder contains code to reproduce the results in Simulation 2. Code structure the same as sim_study_1.

• sim_study_3: folder contains code to reproduce the results in Simulation 3. Code structure the same as sim_study_1.

• sim_study_4: folder contains code to reproduce the results in Simulation 4. Code structure the same as sim_study_1.

• breastCancerNKI: folder contains code to (i) process data from the breastCancerNKI R package for biclustering and (ii) run code to reproduce the results in Section 4 of Moran et al (2020).

• zeisel: folder contains code to (i) process data from Zeisel et al (2015) and (ii) run code to reproduce the results in Section 5 of Moran et al (2020).

NOTE: for both breastCancerNKI and zeisel, the script 0_process_data.R does not need to be run as processed data is stored in ../data. The script is included here for completeness.

• Moran G, Rockova V, George E (2020) "Spike-and-Slab Lasso Biclustering" Annals of Applied Statistics (Accepted)

• Denitto, M., Bicego, M., Farinelli, A. and Figueiredo, M. A. (2017). "Spike and slab biclustering." Pattern Recognition

• Gao, C., McDowell, I. C., Zhao, S., Brown, C. D. and Engelhardt, B. E. (2016). "Context Specific and Differential Gene Co-expression Networks via Bayesian Biclustering." PLoS Comput Biol.

• Schroeder M, Haibe-Kains B, Culhane A, Sotiriou C, Bontempi G, Quackenbush J (2019). breastCancerNKI: Genexpression dataset published by van't Veer et al. [2002] and van de Vijver et al. [2002] (NKI). R package version 1.22.0, http://compbio.dfci.harvard.edu/.

• Amit Zeisel, Ana B. Muñoz Manchado, Peter Lönnerberg, Gioele La Manno, Simone Codeluppi, Anna Juréus, Sueli Marques, Hermany Munguba, Liqun He, Christer Betsholtz, Charlotte Rolny, Gonçalo Castelo-Branco, Jens Hjerling-Leffler and Sten Linnarsson (2015) "Cell types in the mouse cortex and hippocampus revealed by single-cell RNA-seq" Science