Needle in the haystack: structure-based toxin discovery

Area of Science:

• Bioinformatics and computational biology
• Structural biology
• Protein science

Background:

• Modern research generates vast amounts of data, necessitating advanced bioinformatics tools for knowledge discovery.
• Expanding knowledge of protein families with low sequence identity but high structural conservation is challenging.
• Mono-ADP-ribosylating toxins exemplify protein families where sequence-based methods are limited.

Purpose of the Study:

• To develop and evaluate a strategy for expanding protein-family knowledge, focusing on cases with low sequence identity.
• To advance protein-family expansion methodology towards a structure-based approach.

Main Methods:

• Evolution of strategies from PSI-BLAST searches to incorporating secondary-structure prediction and pattern-based searches.
• Implementation of a novel tactic dominated by fold recognition to reduce reliance on sequence similarity.

Main Results:

• The fold recognition-dominated tactic significantly reduces dependence on sequence similarity for identifying protein family members.
• This approach facilitates a more accurate, structure-based protein-family expansion.

Conclusions:

• A structure-based methodology, prioritizing fold recognition, is superior for expanding protein families with conserved structures but divergent sequences.
• This strategy enhances the discovery of new protein family members, particularly for toxins like mono-ADP-ribosylating toxins.