BumpyMatrix

BumpyMatrix implements the BumpyMatrix class and several subclasses for holding non-scalar objects in each entry of a matrix, enabling natural representation of multivariate data in formats compatible with two-dimensional Bioconductor containers.

Key Features:

• BumpyMatrix class: Provides a matrix structure where each entry holds non-scalar objects, with raggedness in the third dimension akin to a bumpy surface.
• BumpyDataFrameMatrix subclass: Stores Bioconductor data frames as individual matrix entries for multivariate data representation.
• Bioconductor container compatibility: Integrates with two-dimensional containers including SummarizedExperiment and MultiAssayExperiment objects.

Scientific Applications:

• Multivariate data representation: Enables storage of complex multivariate data structures within matrix-based frameworks in Bioconductor workflows.
• Multi-omics data integration: Supports compatibility with SummarizedExperiment and MultiAssayExperiment for combined analysis of heterogeneous biological datasets.

Methodology:

The BumpyMatrix class implements a ragged array structure with raggedness in the third dimension, allowing non-scalar objects to occupy individual matrix entries.