Estimation of alternative splicing variability in human populations

Area of Science:

• Genomics
• Transcriptomics
• Bioinformatics

Background:

• DNA arrays and RNA-sequencing (RNA-seq) are used for transcriptome-wide gene expression analysis.
• RNA-seq enables quantification of alternative transcript isoforms and estimation of alternative splice form ratios.
• Changes in splicing ratios, independent of gene expression levels, can significantly impact phenotypes.

Purpose of the Study:

• Develop statistical methodology to measure and compare splicing ratio variability within and between conditions.
• Identify genes with condition-specific splicing ratios.
• Differentiate the contributions of gene expression variability and splicing ratio variability to overall transcript abundance variability.

Main Methods:

• Developed statistical methods for analyzing splicing ratio variability from RNA-seq data.
• Applied methodology to RNA-seq data from lymphoblastoid cells of Caucasian and Yoruban individuals.
• Quantified the relative contributions of gene expression and splicing to transcript abundance variability.

Main Results:

• Protein-coding genes show low splicing variability within populations; many have constant splicing ratios.
• Up to 10% of protein-coding genes exhibit population-specific splicing ratios between Caucasian and Yoruban individuals.
• ~60% of total transcript isoform abundance variability is attributed to transcription variability; splicing variability accounts for a significant portion of the remainder.

Conclusions:

• Splicing variability is generally low within populations but can be condition-specific.
• Variability in transcription is the primary driver of transcript isoform abundance variability.
• Splicing variability often co-occurs with transcription variability.