Recapitulating phylogenies using k-mers: from trees to networks

Area of Science:

• Genomics
• Evolutionary Biology
• Bioinformatics

Background:

• Phylogenetic inference traditionally relies on molecular sequences, facing challenges with microbial genome evolution due to recombination and lateral gene transfer.
• Ernst Haeckel's ontogenic recapitulation theory, once foundational for evolutionary trees, is largely discredited.
• Standard multiple sequence alignment is computationally intensive and may oversimplify complex microbial evolutionary histories.

Purpose of the Study:

• To develop a scalable and robust method for inferring microbial phylogenies that accounts for non-tree-like evolutionary events.
• To construct a phylogenetic network based on shared k-mers from whole-genome sequences of bacteria and archaea.
• To explore the utility of k-mer networks in capturing complex microbial genome evolution.

Main Methods:

• Genome sequences from 143 bacterial and archaeal species were analyzed.
• A phylogenetic network was constructed by quantifying shared k-mers (fixed-length subsequences) across whole genomes.
• The k-mer network approach was compared to traditional tree-based inferences.

Main Results:

• The k-mer network successfully captured key aspects of microbial genome evolution, consistent with tree-based inferences.
• The network revealed non-treelike evolutionary features, such as those arising from recombination and lateral gene transfer.
• The method demonstrated high scalability for analyzing large numbers of genomes.

Conclusions:

• Inferring genome phylogenies using k-mers from whole-genome sequences offers a powerful alternative to traditional methods.
• Phylogenetic networks provide a more comprehensive representation of microbial evolution than simple tree structures.
• This dynamic network approach facilitates rapid and intuitive investigation of microbial genome evolution.