ExperimentSubset: an R package to manage subsets of Bioconductor Experiment objects.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Genomic data analysis often requires subsetting large matrices for downstream methods.
• Existing R objects like SummarizedExperiment can be inefficient for subsetting, creating duplicate data and losing provenance.
• Subsetting is crucial for tasks like excluding low-quality samples or selecting variable features.

Purpose of the Study:

• To develop an R package for efficient storage and retrieval of subsetted genomic assays.
• To address challenges of data duplication and provenance tracking in genomic data analysis.
• To provide a flexible container for managing subsets of high-throughput genomic data.

Main Methods:

• Developed the ExperimentSubset R package with specialized classes for subsetted assays.
• Implemented inherent data provenance by maintaining relationships between subsetted and parent assays.
• Demonstrated utility using a single-cell RNA-sequencing dataset.

Main Results:

• ExperimentSubset enables efficient storage and streamlined retrieval of row/column subsetted assays.
• The package maintains data provenance, linking subsets to their original data.
• Utilizing ExperimentSubset on a single-cell RNA-seq dataset resulted in a lower memory footprint.

Conclusions:

• ExperimentSubset offers a flexible and memory-efficient solution for managing subsets of genomic data.
• The package enhances data provenance, crucial for reproducible research in genomics.
• This tool streamlines common computational tasks in high-throughput data analysis workflows.