ARD2

ARD2 identifies alpha-rod repeats (alpha-solenoids), including HEAT and Armadillo repeats, in protein sequences using a neural network to detect structural repeat domains.

Key Features:

• Neural Network-Based Detection: ARD2 uses a neural network trained on an extensive dataset of protein structures to identify alpha-solenoids even when sequence homology is minimal.
• Improved Algorithm: The detection algorithm was refined using recently solved alpha-solenoid structures to improve accuracy and reliability.
• Discovery of Hidden Alpha-Solenoids: ARD2 has identified previously overlooked alpha-solenoids, including occurrences within the core of the catalytic subunit of PI3KC.

Scientific Applications:

• Expansion of Known Alpha-Solenoids: Application across diverse proteins has expanded the catalogue of known alpha-solenoids.
• Protein-Protein Interaction Insights: Analyses reveal alpha-solenoids are enriched in proteins involved in protein–protein interactions, implicating them in cellular interaction networks.
• Taxonomic Distribution Analysis: ARD2 facilitates study of taxonomic distribution and supports that alpha-solenoids may have arisen independently across taxa via convergent evolution.
• Genomic Enrichment Patterns: The tool highlights higher prevalence of alpha-solenoids in eukaryotic genomes and in prokaryotic families such as Cyanobacteria and Planctomycetes.

Methodology:

ARD2 applies a neural network trained on an extensive dataset of protein structures to analyze protein sequences, with its detection algorithm refined using recently solved alpha-solenoid structures.