Skip to content

MultiAssayExperiment

MultiAssayExperiment#

MultiAssayExperiment (MAE) is an R package used to harmonize data management of multiple experimental assays performed on an overlapping set of specimens. It is designed to manage and integrate multiple types of omics or experimental data (e.g., RNA-seq, mutation data, methylation, proteomics) into a single, structured object. The object stores different data modalities (assays) together, keep track of which samples have which data types, with a goal to facilitate joint analysis, visualization, and subsetting across multiple experiments.

You can create a MultiAssayExperiment object if you have a multiomics dataset (RNA-seq, mutation, proteomics data) for the same or overlapping sets of samples, and if you want to apply multi-omics integration methods, ML models, or visualizations that require synchronized datasets.

"It provides a familiar Bioconductor user experience by extending concepts from SummarizedExperiment, supporting an open-ended mix of standard data classes for individual assays, and allowing subsetting by genomic ranges or rownames. Facilities are provided for reshaping data into wide and long formats for adaptability to graphing and downstream analysis."

Citation

Ramos M, Schiffer L, Re A, Azhar R, Basunia A, Rodriguez Cabrera C, Chan T, Chapman P, Davis S, Gomez-Cabrero D, Culhane A, Haibe-Kains B, Hansen K, Kodali H, Louis M, Mer A, Reister M, Morgan M, Carey V, Waldron L (2017). “Software For The Integration Of Multi-Omics Experiments In Bioconductor.” Cancer Research, 77(21), e39-42. doi:10.1158/0008-5472.CAN-17-0344, https://cancerres.aacrjournals.org/content/77/21/e39.

Structure of a MAE Object#

  • ExperimentList: A list of assays (i.e. SummarizedExperiment, matrix).
  • colData: Metadata about the primary samples (patients/cell lines), such as age, treatment, outcome.
  • sampleMap: A table mapping how each assay's sample IDs relate to the primary colData sample IDs (handles cases where IDs differ across experiments).
    MultiAssayExperiment
|-- ExperimentList
        |-- RNA-seq matrix
        |-- Mutation calls (binary matrix)
        |-- Proteomics data
|-- colData (patient-level metadata)
|-- sampleMap (which patient → which sample in each assay)

Installation requirement#

R (version "4.5") 

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

    BiocManager::install("MultiAssayExperiment")

Build a MAE Object#

    library(MultiAssayExperiment)

    # Suppose you have rna, mutation data and patient information
    rna <- SummarizedExperiment(assays = list(counts = matrix(rnorm(1000), ncol = 10)))
    snv <- matrix(sample(0:1, 100, replace = TRUE), ncol = 10)
    patient_info <- as.data.frame(
    PatientID = paste0("Patient", 1:10),
    Age = sample(40:80, 10)
    )

    # Create the MultiAssayExperiment object with function
    mae <- MultiAssayExperiment(
    experiments = ExperimentList(rna = rna, mutations = mutations),
    colData = patient_info
    )

Work with a MAE#

We will give an example Using Immune Checkpoint Blockade Dataset. Suppose you have downloaded a Immune Checkpoint Blockade Dataset from ORCESTRA

    library(MultiAssayExperiment)

    # Load dataset
    icb_mae <- readRDS(icb_file)

    # Extract clinical data (patient information, treatment, disease etc.)
    clinical_data <- colData(icb_mae) %>% as.data.frame()

    # view RNA
    rna_se <- experiments(icb_mae)[["expr"]]
    rna_sample <- colnames(rna_se) # Sample names inside RNA assay
    rna_expr <- assay(rna_se) # save the rna expression with sample to a matrix

    # view mutation data
    snv_se <- experiments(icb_mae)[["snv"]]
    snv_matrix <- assay(snv_se)

    # Mapping between assays and primary samples
    sample_map <- sampleMap(icb_mae)

    # View metadata if available, this include any extra information about the experiment (i.e. batch info, project details)
    meta_data <- metadata(icb_mae)

Additional References#