Skip to contents

DIVAS Vignette: GNP Dataset Example

Yinuo Sun

2025-07-11

Source: vignettes/DIVAS_GNP_Dataset_Example.Rmd
DIVAS_GNP_Dataset_Example.Rmd

Example 2: Using the GNP Dataset

In addition to the toy dataset, we can also demonstrate how to use DIVAS with a different type of data format. The following example uses a dataset stored in qs format.

About the GNP Dataset

The gnp_imputed.qs dataset contains multi-omics data organized in a list structure. When loaded, it provides:

  1. gnp$datablock: A list of three named data matrices:
    • RNA: Gene expression data (8432 genes × 18 samples)
    • PRO: Protein expression data (3156 proteins × 18 samples)
    • MIC: MicroRNA expression data (283 miRNAs × 18 samples)
  2. gnp$metaData: A data frame with 18 rows (samples) and 8 columns of metadata, including information about sample stage, donor, sex, and age.

All three data matrices share the same samples (18 columns with identical sample IDs), making this a perfect example for multi-modal data integration with DIVAS. Each data type (RNA, protein, miRNA) represents a different “view” of the same biological samples.

To run this example, you’ll need the qs package:

# install.packages("qs")
# library(qs)

Loading and Analyzing the GNP Dataset 

# library(qs)
# gnp_data_path <- system.file("extdata", "gnp_imputed.qs", package = "DIVAS") 
# gnp <- qs::qread(gnp_data_path)
# 
# datablock_gnp <- gnp$datablock
# divasRes_gnp <- DIVAS::DIVASmain(datablock_gnp, nsim = 400, colCent = TRUE)
# 
# # Create names for the data blocks
# dataname_gnp <- paste0("DataBlock_", names(datablock_gnp))
# plots_gnp <- DIVAS::DJIVEAngleDiagnosticJP(datablock_gnp, dataname_gnp, divasRes_gnp, 566, "GNP Demo")
# print(plots_gnp)

References

  1. Prothero, J., …, Marron J. S. (2024). Data integration via analysis of subspaces (DIVAS).

  2. DIVAS R package. https://github.com/ByronSyun/DIVAS_Develop.

  3. Klein, C., Hesse, S., Mao, J., Hadziahmetovic, A., et al. (2025). A molecular atlas of human granulopoiesis. https://www.researchsquare.com/article/rs-6184761/v1

  4. GNP dataset provided by the Comprehensive Childhood Research Center
    at the Dr. von Hauner Children’s Hospital (Klein Lab). https://granulopoiesis.com/

For more detailed technical information about the DIVAS method, please refer to the primary publication (1). For details on the R package implementation, please visit the GitHub repository (2). The GNP dataset used in Example 2 is from the human granulopoiesis atlas project (3,4).