SPAED

SPAED leverages the predicted aligned error (PAE) matrix from AlphaFold to segment protein domains, specifically identifying domain-linker boundaries and detecting terminal disordered regions in phage endolysins.

Key Features:

• Hierarchical clustering of PAE matrices: SPAED applies hierarchical clustering to AlphaFold's PAE matrix to segment phage endolysin domains.
• Domain-linker boundary identification: SPAED identifies boundaries between protein domains and linkers using information contained in the PAE matrix.
• Terminal disordered region detection: SPAED detects terminal disordered regions in proteins.
• Benchmarked performance: On a dataset of 376 bacteriophage endolysins, SPAED achieves a mean intersect-over-union score of 96% and a domain-boundary-distance score of 89%, compared to 94% and 70% for Chainsaw.

Scientific Applications:

• Phage endolysin domain segmentation: SPAED accurately segments domains in bacteriophage endolysins, which are proteins that degrade the bacterial cell wall.
• Protein domain boundary prediction: SPAED improves identification of domain-linker boundaries compared to state-of-the-art tools for proteins with AlphaFold predictions.

Methodology:

SPAED applies hierarchical clustering to AlphaFold's predicted aligned error (PAE) matrix.