ArchaeaFun

ArchaeaFun predicts and classifies enzymes from archaeal genome sequences using ab initio protein feature–based analyses to identify enzymatic functions without relying on sequence similarity.

Key Features:

• Ab initio prediction: Uses predicted protein features including cotranslational and posttranslational modifications, secondary structure predictions, and simple physical/chemical properties to identify enzyme candidates.
• Non-reliance on sequence similarity: Eschews sequence similarity as a criterion, enabling detection of enzymes that lack recognizable homologs in existing databases.
• Comprehensive feature integration: Integrates multiple predicted protein features into a unified predictive model to improve identification accuracy.
• Enzyme class classification: Assigns predicted enzyme candidates to enzyme classes based on their computed feature profiles.

Scientific Applications:

• Discovery of novel enzymes: Enables identification of candidate biocatalysts from archaeal genomes that may be undetectable by homology-based methods.
• Functional annotation of genomes: Supports annotation of uncharacterized archaeal genes by assigning probable enzymatic functions.
• Biotechnological applications: Provides candidate enzymes for evaluation in biofuel production, pharmaceutical development, and environmental remediation.

Methodology:

Computationally predicts protein features (cotranslational and posttranslational modifications, secondary structure, and simple physical/chemical properties), constructs a feature profile for each gene product, integrates multiple features into a predictive model to assess enzymatic likelihood and assigns enzyme class, explicitly without using sequence similarity.