EvANI benchmarking workflow for evolutionary distance estimation

Area of Science:

• Genomics and Evolutionary Biology
• Bioinformatics and Computational Biology

Background:

• Advances in long-read sequencing yield high-quality genome assemblies, enabling comparative genomics across the Tree of Life.
• Average nucleotide identity (ANI) quantifies genetic similarity between genomes, aiding species delineation and phylogenetic analysis.
• Traditional ANI calculation via genome alignment is computationally intensive, prompting the development of faster, sketch-based methods.

Purpose of the Study:

• To introduce EvANI, a novel evaluation framework for assessing the accuracy and efficiency of different average nucleotide identity (ANI) estimation algorithms.
• To analyze the impact of assumptions and heuristics in sketch-based ANI methods on distance estimations.
• To guide the selection of appropriate ANI estimation strategies for diverse genomic datasets and evolutionary studies.

Main Methods:

• Development of a benchmark dataset comprising simulated and real genomic data for rigorous evaluation.
• Implementation of a rank-correlation-based metric to quantify the accuracy of ANI estimates against evolutionary distances.
• Comparative analysis of various ANI estimation algorithms, including alignment-based (ANIb) and k-mer-based approaches.

Main Results:

• ANIb provides the most accurate tree distance estimation but is computationally inefficient.
• K-mer-based methods demonstrate high efficiency and consistent accuracy across various datasets.
• Optimal k-mer lengths can vary by clade, suggesting the utility of multiple k-mer values (e.g., k=10 and k=19 for Chlamydiales).
• Maximal exact match approaches offer a balance between computational efficiency and accuracy.

Conclusions:

• EvANI serves as a robust framework for evaluating genome comparison tools.
• K-mer-based strategies are highly effective for rapid and accurate ANI estimation in large-scale genomic comparisons.
• The choice of k-mer length and the consideration of alternative methods like maximal exact matches are important for optimizing phylogenetic analyses.