SHARK.capture

SHARK.capture identifies functional sequence motifs within intrinsically disordered protein regions (IDRs) to characterize sequence–function relationships and detect motifs that mediate biomolecular interactions.

Key Features:

• Alignment-Free Approach: Performs motif detection without sequence alignment, addressing the rapid evolution and alignment challenges of IDRs.
• Word-Based Motif Detection: Scans sequences for recurring patterns or "words" to detect candidate motifs.
• Incorporation of Amino Acid Physicochemistry: Uses amino acid physicochemical properties to assess motif similarity without relying on rigid equivalency groups.
• Systematic Benchmarking and Performance: Exhibits consistently strong performance in systematic benchmarks, particularly at residue-level detection accuracy.
• Experimental Validation: Recovers known functional motifs across orthologs of the microtubule-associated zinc finger protein BuGZ and identifies a short motif in the IDR of S. cerevisiae RNA helicase Ded1p that was experimentally verified to enhance ATPase activity.
• Extensive Motif Database: Facilitated the systematic calculation of 10,889 motifs across 2,695 yeast IDRs.

Scientific Applications:

• Sequence–Function Mapping in IDRs: Identify conserved and functionally significant motifs to elucidate roles of IDRs in protein–protein interactions, signaling pathways, and other cellular functions.
• Comparative Motif Detection: Detect motifs across orthologs, exemplified by analyses of BuGZ orthologs.
• Functional Characterization: Prioritize motifs for experimental validation, such as the Ded1p IDR motif that enhances ATPase activity.
• Study of Rapidly Evolving Regions: Detect motifs in rapidly evolving IDRs across diverse species without relying on alignments.

Methodology:

Employs an alignment-free, word-based scanning method that incorporates amino acid physicochemical properties to assess motif similarity, and was used for systematic calculation of motifs across yeast IDRs and for benchmarking residue-level detection accuracy.