XRate: a fast prototyping, training and annotation tool for phylo-grammars

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Phylo-grammars, combining Markov chains and stochastic grammars, are emerging for genome annotation.
• Previous phylo-grammar implementations were complex, hindering adoption by computational biologists.

Purpose of the Study:

• To develop an accessible, open-source software tool for genome annotation using phylo-grammars.
• To enable efficient estimation of phylogenetic trees and model parameters.

Main Methods:

• Developed 'xrate', an open-source tool supporting various substitution models and stochastic context-free grammars.
• Implemented a novel 'phylo-EM' algorithm for efficient maximum-likelihood parameter and phylogenetic tree estimation.
• Grammar specification in external files allows user customization without code recompilation.

Main Results:

• xrate efficiently estimates maximum-likelihood parameters and phylogenetic trees.
• The tool was used to measure codon substitution rates and predict protein/RNA secondary structures.
• Biologically meaningful rates were estimated, with prediction accuracy comparable to specialized tools.

Conclusions:

• xrate successfully implements phylo-grammar-based genome annotation.
• The software provides an accessible and efficient solution for computational biologists.
• xrate's predictions demonstrate accuracy comparable to existing specialized tools.