A phylogenetic network is a graph that represents evolutionary relationships among species, incorporating events like horizontal gene transfer.A phylogenetic network is a model of evolutionary relationships that, in addition to branching (speciation), accounts for reticulation events: horizontal gene transfer, hybridization, recombination. Unlike a phylogenetic tree, the network contains nodes with multiple parent ancestors, making it more suitable for describing the evolution of prokaryotes, viruses, and some plants.
The concept of phylogenetic networks arose in response to discrepancies between gene trees and species trees. In the 1980s, Walter Fitch realized the need to account for horizontal transfer, and in the 1990s, with the development of sequencing methods and computer power, the first algorithms for network construction appeared (e.g., the split decomposition algorithm of Bandelt and Dress, 1992). Pioneers of computational phylogenetics included David Swofford, Joseph Felsenstein, and later David Bryant, Daniel Huson, and others who developed modern methods.
Construction of a phylogenetic network begins with the choice of an evolutionary model and data (usually DNA sequences). Algorithms estimate distances or likelihoods and search for a topology that minimizes the number of reticulations or maximizes likelihood. Networks can be explicit — where reticulation nodes denote specific transfer events — or implicit, showing alternative evolutionary scenarios as intertwined structures (e.g., split networks). Validation and network selection are often performed using information criteria (AIC, BIC) and bootstrap.
Phylogenetic networks help track the spread of viruses (e.g., COVID-19) and identify hybrid plant species for breeding.
Up to 10% of the human genome may be of viral origin, and network models help trace these unusual genetic connections.















