TrAnsFuSE refines the search for protein function: oxidoreductases.

Area of Science:

• Biochemistry and Molecular Biology
• Evolutionary Biology
• Bioinformatics

Background:

• Non-equilibrium catalysis by oxidoreductases, utilizing transition metals, is crucial for biological macromolecule element flux.
• These ancient enzyme domains, originating in microbes, are difficult to identify due to sequence divergence.
• Current homology-based methods for annotating uncharacterized protein sequences have limitations in accuracy and reliability.

Purpose of the Study:

• To develop a robust method for identifying redox enzyme domains in large, unannotated sequence datasets.
• To improve the accuracy of protein function annotation transfer for oxidoreductases.
• To enhance the study of evolutionary relationships and functional diversity of these critical enzymes.

Main Methods:

• Developed TrAnsFuSE, a computational approach combining profile-based searches with catalytic site annotations.
• Validated annotations of 104 InterPro domains involved in transition metal-mediated redox reactions.
• Utilized experimentally identified catalytic residues to refine sequence alignment-based function annotations.

Main Results:

• TrAnsFuSE demonstrated 11% and 14% higher accuracy than PSI-BLAST and InterPro, respectively.
• The method proved robust, achieving higher accuracy with comparable coverage when including metal-binding sites.
• Successfully identified redox domains across diverse oxidoreductases, primarily those using iron, copper, and molybdenum.

Conclusions:

• TrAnsFuSE significantly enhances the detection accuracy of redox domains in unannotated sequences.
• This method facilitates the study of vast amounts of genomic and metagenomic data.
• Improves understanding of oxidoreductase evolution and function, particularly ancient metalloenzymes.