AcrDB

AcrDB catalogs computationally predicted anti-CRISPR (Acr) proteins and associated Aca operons to enable analysis of Acr occurrence, genomic context, and potential inhibition of prokaryotic CRISPR-Cas systems.

Key Features:

• Genome-scale catalog: Contains 39,799 predicted Acr-Aca operons derived from a broad survey of genomes.
• Biological scope: Focuses on short anti-CRISPR (Acr) proteins predominantly found in phages, prophages, and other mobile genetic elements that inhibit CRISPR-Cas systems of prokaryotic hosts.
• Genomic context emphasis: Emphasizes the genomic neighborhoods of Acr and Aca candidates rather than isolated protein families.
• Integrated prediction: Integrates results from three independent programs—AcrFinder, AcRanker, and PaCRISPR—each using distinct algorithms for data mining and prediction.
• Machine-learning scoring: Proteins identified in predicted operons are assigned numerical scores or ranks by the machine learning-based programs AcRanker and PaCRISPR.
• Genome coverage: Based on scanning approximately 19,000 genomes from prokaryotes and viruses.

Scientific Applications:

• CRISPR-Cas research: Supports identification and comparative analysis of Acr proteins and Aca operons to study inhibition of CRISPR-Cas systems.
• Genome editing control: Provides candidate Acrs relevant to mitigating excessive or prolonged Cas nuclease activity during genome editing applications.
• Evolutionary and ecological studies: Facilitates investigation of evolutionary dynamics between phages/viruses, mobile genetic elements, and their prokaryotic hosts and defense mechanisms.

Methodology:

AcrDB was built by scanning ~19,000 prokaryote and virus genomes with AcrFinder to predict Acr-Aca operons, followed by scoring/ranking of proteins in those operons using AcRanker and PaCRISPR, and integrating outputs from AcrFinder, AcRanker, and PaCRISPR.