Phylogenetics

Phylogenetics is the branch of biology that investigates the evolutionary history of life by inferring relationships among organisms from observable traits—such as DNA sequences, protein structures, and physical form—a process called phylogenetic inference. The outcome is a phylogenetic tree, a diagram showing hypothetical evolutionary connections. The tips of the tree represent observed entities, which may be living species or fossils. Trees can be rooted, indicating a hypothetical common ancestor for all included taxa, or unrooted, which makes no assumption about evolutionary direction or origin. Beyond species relationships, phylogenetic analyses are also used to map connections among genes or individual organisms, and have become essential for understanding biodiversity, evolution, ecology, and genomes. Phylogenetics is a part of systematics, using similarities and differences in species’ characteristics to interpret their evolutionary origins and relationships.

In cancer research, phylogenetics helps study the clonal evolution of tumors and molecular chronology, using whole genome sequencing to predict and show how cell populations change during disease progression and treatment. Because cancer cells reproduce mitotically, the evolutionary processes differ from those in sexually reproducing species, leading to distinct types of aberrations, higher mutation rates, greater heterogeneity among tumor subclones, and an absence of genetic recombination. Phylogenetics also aids drug design by organizing species to identify which are likely to have inherited medically useful traits, such as producing biologically active compounds. For example, venom-producing animals yield important drugs like ACE inhibitors and Prialt (Ziconotide). To discover new venoms, scientists use phylogenetics to screen for closely related species that may share the same trait; the phylogenetic tree highlights venomous fish, snakes, and lizards, as well as related species that might also be venomous. In forensic science, phylogenetic tools assess DNA evidence for court cases, having been used to exonerate or implicate individuals. HIV forensics specifically uses phylogenetic analysis to compare HIV genes and determine relatedness between two samples, though it cannot serve as sole proof of transmission and does not indicate the direction of transmission.

Taxonomy involves identifying, naming, and classifying organisms. The Linnaean system, developed in the 1700s by Carolus Linnaeus, is the foundation of modern classification, originally relying on physical characteristics. With biochemistry, classifications now often use DNA sequence data or a combination of DNA and morphology. Many systematists argue that only monophyletic groups should be recognized as named taxa. How much classification depends on inferred evolutionary history varies by school: phenetics ignores phylogeny and focuses on overall similarity; cladistics (phylogenetic systematics) bases classification solely on shared, derived characters (synapomorphies) to reflect phylogeny; evolutionary taxonomy balances branching patterns with “degree of difference” to compromise between common ancestry and evolutionary distinctness.

Phylogenetic inference typically uses computational methods with optimality criteria, including parsimony, maximum likelihood, and MCMC-based Bayesian inference. All rely on an explicit or implicit mathematical model of character state transformation probabilities. Phenetics, popular in the mid-20th century but now largely obsolete, used distance matrix-based methods to build trees from overall similarity in morphology or observable traits, often assumed to approximate phylogeny. Neighbor joining, a phenetic method, is still used for similarity trees in DNA barcoding. Before 1950, phylogenetic inferences were mostly narrative scenarios, often ambiguous and lacking explicit criteria for evaluating alternatives.

In phylogenetic analysis, taxon sampling selects a small group of exemplar taxa to infer a clade’s evolutionary history, also called stratified or clade-based sampling. Judicious sampling is crucial given limited resources and computational constraints; poor sampling can lead to inaccurate results.