betAS: intuitive analysis and visualization of differential alternative splicing using beta distributions.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Next-generation RNA sequencing provides high-resolution alternative splicing (AS) quantification.
• Percent spliced-in (PSI) values, commonly used for AS quantification, lack precision information related to read coverage.
• Beta distributions can model AS inclusion levels and their precision effectively.

Purpose of the Study:

• To develop a computational pipeline for quantitative and visual comparison of alternative splicing between sample groups.
• To introduce a differential splicing significance metric that accounts for intergroup differences, estimation uncertainty, and intragroup variability.
• To provide an accessible tool for differential splicing analysis for both computational and non-computational biologists.

Main Methods:

• Utilizing beta distributions to model PSI values and their associated precision based on read counts for inclusion and exclusion.
• Developing a computational pipeline for quantitative and visual comparison of AS events across multiple sample groups.
• Implementing a novel differential splicing significance metric incorporating magnitude of difference, estimation uncertainty, and variability.

Main Results:

• A computational pipeline and methodology based on beta distributions for accurate PSI value modeling and precision interpretation.
• A differential splicing significance metric suitable for multiple-group comparisons, considering various statistical factors.
• The development of betAS, an R package and web app for user-friendly, visual differential splicing analysis.

Conclusions:

• Beta distribution modeling offers a robust approach to quantify and interpret AS precision from RNA sequencing data.
• The betAS tool provides an accessible and intuitive platform for differential splicing analysis, enhancing biological insights.
• This methodology improves the quantitative and visual comparison of alternative splicing events across diverse biological samples.