RabbitCD-HIT: Distributed High-Sensitivity Greedy Incremental Clustering for Billion-Scale Protein Sequences in Hours

RabbitCD-HIT clustered the March 10, 2026 NCBI nr snapshot (1.02 billion sequences; 487 GB FASTA) at 90\% identity in 5 hours on 64 nodes.

• Why it matters: Billion-scale protein clustering is bottlenecked by the performance limits of traditional greedy incremental methods, slowing high-sensitivity analyses.
• What they did: The tool uses distributed greedy incremental clustering that fully exploits cluster resources, and it was benchmarked against CD-HIT on a 256-core server under matched settings.
• The result: RabbitCD-HIT delivered a 2.1$\times$–6.6$\times$ speedup over CD-HIT while producing identical results, and it releases representatives and assignments as a public database resource.

AI summary of the abstract · may contain errors · not medical advice · read the paper.