demo/README.md

In xiangli2pro/hbcm: HBCM model for cluster analysis

Code for manuscript Community Detection with Heterogeneous Block Covariance Model

The folder provides comprehensive code scripts to replicate the numerical results and graphs presented in manuscript Community Detection with Heterogeneous Block Covariance Model. The readme doc gives instructions to navigate the files.

Data

Real Gene Expression Data

The yeast gene expression data sets used in the manuscript Figure 3 and Figure 4 are embedded in the hbcm package, and can be directly called by their names (data_gene_sample, data_gene_group) in R (NO need to use the load() function).

The data sets can also be downloaded from the package github webpage. Those data sets are processed following the script here. Raw data pre-process is available in package github webpage.

Simulation Data

Simulation data can be generated by running the R scripts provided in this directory. Note that the manuscript used 100 simulation data sets for each scenario, which requires intensive computing resources and was executed on cloud computing platform. If the purpose is to test the code and have a simple demo, we recommend to use fewer (<5) simulation data for each scenario.

Scripts

Each script has a markdown (.qmd) version and a pdf version (.pdf). Users can either download the markdown version and execute it as instructed, OR, check the code and output in the pdf version.

To replicate manuscript Table 1 simulation - Run table1_simulation.qmd or use table1_simulation.pdf

To replicate manuscript Figure 1 simulation - Run figure1_left_simulation.qmd or use figure1_left_simulation.pdf; - Run figure1_middle_simulation.qmd or use figure1_middle_simulation.pdf; - Run figure1_right_tdist_simulation.qmd or use figure1_right_tdist_simulation.pdf; - Run figure1_right_tdist_corrected_simulation.qmd or use figure1_right_tdist_corrected_simulation.pdf.

To replicate manuscript Figure 2 simulation - Run figure2_simulation.qmd or use figure2_simulation.pdf.

To replicate manuscript Figure 3 simulation - Run figure3_plot.qmd or use figure3_plot.pdf.

To replicate manuscript Figure 4 result - Run figure4_plot.qmd or use figure4_plot.pdf.

xiangli2pro/hbcm documentation built on Nov. 15, 2024, 9:15 a.m.