recount3: summaries and queries for large-scale RNA-seq expression and splicing

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Publicly available RNA sequencing (RNA-seq) data is vast but often difficult to access and analyze uniformly.
• Standardized processing of large-scale transcriptomic datasets is crucial for reproducible biological research.

Purpose of the Study:

• To introduce recount3, a comprehensive resource of over 750,000 human and mouse RNA sequencing samples.
• To provide accessible tools (R/Bioconductor packages and web resources) for querying and analyzing RNA-seq data.
• To enable comparison of local or private RNA-seq data with the public recount3 dataset using the Monorail pipeline.

Main Methods:

• Uniform processing of over 750,000 human and mouse RNA sequencing samples using the Monorail analysis pipeline.
• Development of recount3 and snapcount R/Bioconductor packages for data access and manipulation.
• Provision of web resources for data exploration and querying at various levels (study, exon-exon junctions, genes, samples).

Main Results:

• Creation of recount3, a large-scale, uniformly processed RNA sequencing dataset.
• Availability of user-friendly R/Bioconductor packages and web tools for data retrieval and analysis.
• The Monorail pipeline allows for the integration and comparison of public and private RNA-seq data.

Conclusions:

• recount3 and its associated tools significantly enhance the utility of public RNA sequencing data.
• Biologists can leverage these resources to gain deeper insights into their own data and advance biological understanding.
• The Monorail pipeline facilitates reproducible analysis and comparison across diverse RNA-seq studies.