jD2Stat

jD2Stat computes k-mer-based D2 statistics to generate pairwise distances from nucleotide and amino acid sequences for alignment-free phylogenetic inference and downstream tree construction such as neighbour-joining.

Key Features:

• Alignment-free methodology: Extracts k-mers and computes distances without performing multiple sequence alignment.
• Subsequence property extraction: Uses properties of sub-sequences (e.g., identity or match length) to inform distance calculations.
• Pairwise distance computation: Calculates D2-statistic-based pairwise distances for every sequence pair to produce a distance matrix.
• Robustness and scalability: Demonstrates robustness to among-site rate heterogeneity, compositional biases, genetic rearrangements, and insertions/deletions, supporting large-scale phylogenomic analyses.
• Efficiency in low divergence scenarios: Provides faster computation than alignment-based methods for datasets with low sequence divergence.
• Sensitivity considerations: Shows sensitivity to recent sequence divergence and sequence truncation, which can affect distance estimates.

Scientific Applications:

• Large-scale phylogenomics: Infers phylogenies from extensive datasets where alignment is computationally prohibitive or unreliable due to complex evolutionary histories.
• Phylogenetic analysis of nucleotide and amino acid sequences: Applies to inference from both nucleotide and amino acid sequences across diverse evolutionary scenarios.

Methodology:

jD2Stat extracts k-mers from input sequences, computes pairwise distances using D2 statistics, and produces distance matrices that can be used for phylogenetic tree construction such as neighbour-joining in an alignment-free framework.