Noisy splicing drives mRNA isoform diversity in human cells

Area of Science:

• Molecular Biology
• Genomics
• Evolutionary Biology

Background:

• Alternative splicing is common in human genes, but the functional relevance of most splice forms remains unknown.
• Understanding the extent and accuracy of alternative splicing is crucial for interpreting gene expression and evolution.

Purpose of the Study:

• To investigate the prevalence and functional significance of alternative splicing in human cells.
• To identify and characterize novel splice junctions and assess their evolutionary conservation.

Main Methods:

• Deep RNA sequencing of human cells.
• De novo identification of splice junctions.
• Analysis of splice site conservation and associated sequence motifs.

Main Results:

• Discovery of approximately 150,000 unannotated splice junctions, primarily low-abundance isoforms.
• Newly identified splice sites lack evolutionary conservation, suggesting they result from splicing errors.
• Splicing error rate is estimated at 0.7% per intron, with higher accuracy in highly expressed genes.

Conclusions:

• A significant fraction of observed alternative splicing events may represent noise rather than functional variation.
• Noisy splicing, particularly in less conserved splice sites, is a notable feature of genome evolution.
• Gene expression levels influence splicing accuracy, potentially linked to intron length.