ProbAnnoWeb and ProbAnnoPy: probabilistic annotation and gap-filling of metabolic reconstructions

Area of Science:

• Metabolic Engineering
• Computational Biology
• Systems Biology

Background:

• Genome-scale metabolic reconstructions require gap-filling for accurate flux-balance simulations.
• Existing gap-filling tools often employ organism-agnostic approaches, limiting their specificity.
• The accuracy of metabolic models is crucial for predicting cellular behavior and engineering applications.

Purpose of the Study:

• To develop and present novel computational tools for improved gap-filling in metabolic reconstructions.
• To introduce an organism-specific, likelihood-based approach for selecting candidate reactions.
• To provide accessible implementations of probabilistic annotation and likelihood-based gap-filling.

Main Methods:

• Development of ProbAnnoWeb, a web service for probabilistic annotation and gap-filling.
• Implementation of ProbAnnoPy, a standalone Python package for local use.
• Utilizing likelihood scores derived from the target organism's genome for reaction selection.

Main Results:

• ProbAnnoWeb and ProbAnnoPy offer enhanced gap-filling capabilities.
• The likelihood-based approach improves the specificity of reaction selection.
• These tools facilitate the creation of higher-quality metabolic models.

Conclusions:

• ProbAnnoWeb and ProbAnnoPy represent significant advancements in metabolic reconstruction gap-filling.
• The organism-specific approach enhances the reliability of flux-balance simulations.
• These tools are readily available for the research community, promoting wider adoption and application.