Changes in exon-intron structure during vertebrate evolution affect the splicing pattern of exons.

Area of Science:

• Genomics
• Evolutionary Biology
• Molecular Biology

Background:

• Exon-intron architecture guides the splicing machinery to exons within introns.
• The evolutionary history of exon-intron structure and its effect on splicing remain largely unexplored.

Purpose of the Study:

• To investigate the evolutionary dynamics of exon-intron architecture in vertebrates.
• To understand the impact of splice site strength on intron length and exon recognition.
• To identify regulatory sequences involved in alternative splicing and develop predictive models.

Main Methods:

• Comparative genomic analysis of 17 vertebrate genomes.
• Reconstruction of ancestral splice site motifs and exon/intron lengths.
• Analysis of evolutionary rate shifts and identification of cis-acting regulatory sequences.
• Machine-learning approach to predict alternative splicing.

Main Results:

• Vertebrate introns have increased in length throughout evolution.
• Weak splice sites restrict intron length, while strong splice sites facilitate recognition of exons in long introns.
• Novel cis-acting regulatory sequences were identified and experimentally validated.
• A machine-learning algorithm achieved high accuracy (AUC=0.91) in predicting alternative exons, leading to the discovery of new alternatively spliced exons.

Conclusions:

• Splice site strength and exon features play crucial roles in shaping exon-intron architecture during vertebrate evolution.
• Evolutionary constraints on exons and their recognition by the splicing machinery have been elucidated.
• The developed predictive model offers a powerful tool for identifying novel alternatively spliced exons.