Intron evolution: a statistical comparison of two models

Area of Science:

• Molecular Biology
• Evolutionary Biology
• Bioinformatics

Background:

• The origin and distribution of introns (non-coding DNA sequences within genes) remain a significant question in molecular evolution.
• Two primary hypotheses exist: introns are ancient relics ('introns-early') or were acquired later in evolution ('introns-late').

Purpose of the Study:

• To mathematically model and statistically evaluate the 'introns-early' and 'introns-late' hypotheses.
• To assess which model best explains the observed intron distribution in specific genes across different species.

Main Methods:

• Construction of two distinct mathematical models representing intron gain and intron loss scenarios.
• Incorporation of evolutionary tree structures and branch-specific rates of intron dynamics.
• Estimation of rate constants from intron counts and simulation of intron evolution to calculate probability distributions.

Main Results:

• Analysis of actin and alpha-tubulin genes, with the most extensive data, favored the 'introns-early' model.
• The mathematical models calculated probabilities for intron arrangements under both hypotheses.
• Results suggest that the observed intron patterns are more likely explained by introns being ancient.

Conclusions:

• The study provides evidence supporting the hypothesis that introns are largely ancient features of the genome.
• This finding contrasts with some previous analyses, highlighting the importance of robust modeling in evolutionary studies.
• The 'introns-early' model appears to be a better fit for the intron distribution in the studied genes.