Homology-aware average nucleotide identity estimation at scale with TurboANI

TurboANI enables homology-aware average nucleotide identity estimation at scale, delivering up to ~40,000-fold speedup over alignment-based ANI with near-perfect agreement.

• Why it matters: Fast, accurate ANI is needed to compare many microbial genomes, but alignment-based methods are too slow and existing fast tools trade speed for accuracy.
• What they did: TurboANI uses a seed-chain-quasi-alignment scheme with SIMD seeding, compact indexing, lightweight chaining, and cache-friendly bitset sliding minwise hashing to estimate ANI.
• The result: It is >40-fold faster than FastANI and ten-fold faster than skani, and in sensitive mode it recovers true-ANI at a ten-fold higher rate than skani.

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