Molecular Evolution and Phylogenetics

Introduction to Molecular Evolution

Molecular evolution studies how genetic sequences change over time through mutation, selection, genetic drift, and recombination. Since the 1960s discovery that protein sequences carry evolutionary information (molecular clock hypothesis—some proteins evolve at constant rates enabling time estimation of divergences), molecular data have transformed evolutionary biology. DNA and protein sequences contain a richer evolutionary record than morphology—reflecting the actual molecular mechanisms of inheritance, mutation, and selection. Modern genomic analyses examine millions of positions across hundreds of species, reconstructing the tree of life with unprecedented resolution and revealing the molecular mechanisms underlying evolutionary adaptations.

Phylogenetics—reconstruction of evolutionary relationships—uses sequence alignments and mathematical models of sequence evolution to infer branching trees (phylogenies) representing the evolutionary history of genes, genomes, and species. Applications range from tracing infectious disease outbreaks (tracking SARS-CoV-2 variant spread in real time), reconstructing ancient human population history (identifying admixture with Neanderthals and Denisovans), detecting positive selection on adaptive genes, inferring horizontal gene transfer, and classifying new species. Phylogenomics—using genome-scale data—resolves long-standing evolutionary puzzles impossible with single-gene approaches.

Mechanisms of Molecular Evolution

Neutral Theory and Molecular Clock

Motoo Kimura's neutral theory (1968) proposed that most molecular variation is selectively neutral—neither beneficial nor harmful. Neutral mutations accumulate at a rate equal to the mutation rate (approximately 1.1×10-8 per site per generation for humans). This molecular clock enables calibration of divergence times using fossil-calibrated phylogenies. Synonymous substitutions (not changing amino acid) evolve faster than non-synonymous; intergenic regions evolve faster than coding sequences; highly conserved genes (histones, ribosomal proteins) evolve slowly due to functional constraints. The ratio of non-synonymous to synonymous substitutions (dN/dS or Ka/Ks) measures selection: dN/dS >1 indicates positive selection; <1 indicates purifying selection removing deleterious mutations.

Positive Selection and Adaptation

Positive (Darwinian) selection drives advantageous mutations to fixation faster than expected by drift. The McDonald-Kreitman test distinguishes positive selection from relaxed constraint by comparing ratios of fixed and polymorphic synonymous/non-synonymous substitutions. Selective sweeps—rapid fixation of advantageous alleles carrying neighbouring variants hitch-hiking to high frequency—create characteristic patterns of reduced diversity and long haplotype blocks detectable in population genomic data. Lactase persistence (allowing adults to digest milk lactose) due to LCT gene upstream variants evolved independently 2,000-7,000 years ago in African and European cattle-herding populations—a clear example of recent human positive selection driven by cultural-genetic co-evolution.

Phylogenomic Methods

Tree Inference

Phylogenetic tree inference methods include parsimony (minimising total character state changes), maximum likelihood (finding the tree maximising the probability of observed data under a substitution model), and Bayesian inference (integrating over model parameter uncertainty using MCMC sampling to obtain posterior probability distributions of trees). Model selection—choosing appropriate substitution models (GTR+Gamma for DNA; WAG, LG for proteins) accounting for rate variation across sites—is critical for accurate inference. Bootstrap resampling or posterior probabilities quantify node support. Concatenation of many genes increases information; coalescent-based species tree methods (ASTRAL) account for gene tree discordance from incomplete lineage sorting in rapid radiations.

Ancient DNA and Palaeogenomics

Ancient DNA preserved in permafrost, dry caves, and temperate environments enables direct sampling of past genetic diversity. Next-generation shotgun sequencing of DNA from archaeological samples—using enzymatic damage repair and library preparation adapted for short, damaged fragments—recovers full genomes from fossils. The Denisovan genome recovered from a finger bone in a Siberian cave changed human evolutionary biology—revealing a distinct archaic human lineage that introgressed into ancestral non-African populations, with Melanesians and Aboriginal Australians retaining ~4-6% Denisovan DNA. Ancient pathogen genomes trace Yersinia pestis (plague) and other disease histories through archaeological time, revealing origins and spread of historical pandemics.

Comparative Genomics

Comparative genomics identifies functional elements through evolutionary conservation—sequences conserved across diverse species likely have functional importance (constraint). ENCODE-funded comparative analyses of vertebrate genomes identified ~10% of the human genome under evolutionary constraint including non-coding regulatory regions. Synteny—conserved gene order across chromosomes over long evolutionary distances—reveals ancient chromosomal relationships; human-mouse conserved synteny blocks reconstruct the ancestral mammalian karyotype. Comparative genomics of extreme phenotypes (echolocation in bats and dolphins—convergent molecular evolution in the same genes) identifies the genetic basis of adaptive traits, and comparisons across longevous species identify potential longevity genes.

Examples and Applications

Example 1: SARS-CoV-2 Phylogenetic Surveillance

GISAID and Nextstrain platforms enabled real-time phylogenetic tracking of SARS-CoV-2 evolution throughout the COVID-19 pandemic. Genome sequencing of positive samples with phylogenetic analysis identified variant emergence (Alpha, Delta, Omicron), traced geographic spread, detected outbreak clusters, inferred timing of origin, and revealed recombination events. UK COG-Consortium sequenced >50% of all UK COVID-19 cases at peak pandemic, enabling identification of transmission chains in healthcare settings, vaccination escape mutations, and international importation events. Phylogenomic surveillance has become standard public health infrastructure for pandemic preparedness across multiple pathogens.

Example 2: Human Population Genetics and Admixture

Genome-wide SNP data from thousands of geographically diverse individuals reconstruct human population history through admixture analysis, principal component analysis, and explicit demographic modelling. STRUCTURE and ADMIXTURE software reveal distinct genetic clusters broadly corresponding to continental geography. The SGDP (Simons Genome Diversity Project) of 300 diverse individuals documented the Out-of-Africa bottleneck, multiple independent ancestry components in Africa, ancient admixture between ancestral Eurasians and Neanderthals (~2%), and complex migration patterns including the peopling of the Americas via at least one large migration (~15,000 years ago) with subsequent subsidiary migrations producing the Athabascan and Paleo-Eskimo ancestral components.

Example 3: Convergent Molecular Evolution

Convergent evolution—independent evolution of similar traits in unrelated lineages—sometimes involves mutations at the same molecular sites. MYH7 mutations affecting the same residues affect hearing in multiple echolocating mammal lineages (bats, dolphins) using sonar. Alpha-keratin mutations affecting webbing occur independently in diving birds and in separately evolved diving (e.g., swimming) mammal lineages. SLC45A2 pigmentation gene shows convergent light pigmentation variants in independently depigmented cave fish populations. These repeated molecular substitutions under similar selective pressures reveal the limited molecular solutions available to achieve specific phenotypic ends—the genetic basis of evolutionary constraint.

Example 4: Horizontal Gene Transfer in Bacteria

Horizontal gene transfer (HGT)—movement of genes between organisms other than parent-to-offspring vertical inheritance—is rampant in bacteria, driving antibiotic resistance spread and metabolic innovation. Pathogenicity islands—large genomic insertions from HGT—confer virulence factors to previously avirulent strains: E. coli O157:H7 acquired Shiga toxin genes on bacteriophage. Antibiotic resistance genes on mobile elements (plasmids, transposons, integrons) spread between diverse bacterial genera by HGT—explaining why resistance to new antibiotics spreads rapidly across genera despite no direct selective pressure. Detecting HGT uses phylogenetic incongruence between gene trees and species trees, or GC content/codon usage anomalies relative to genome background.

Example 5: Neanderthal Introgression and Human Traits

Sequencing of the Neanderthal reference genome (2010) enabled identification of genome segments inherited by modern humans from interbreeding ~50,000-65,000 years ago in the Middle East or Central Asia. Neanderthal-derived alleles associated with specific traits in modern humans include increased COVID-19 severity (LZTFL1 3p21 risk haplotype strongly enriched for Neanderthal origin), reduced COVID-19 severity (OAS antiviral genes at 12q24), Tibetan adaptation to high altitude (EPAS1 variant from Denisovans), and immune response variants. Certain Neanderthal alleles were advantageous in novel Eurasian environments and swept to high frequency; others affecting skin pigmentation, hair texture, and immunity reflect adaptive introgression.

Example 6: Phylogenomics of Life's Tree

The three-domain tree of life (Bacteria, Archaea, Eukaryotes) proposed by Woese based on rRNA is now refined by phylogenomics. Asgard archaea discovery (from deep-sea sediment metagenomics) places eukaryotes within a lineage of complex Archaea, reframing eukaryotic origin as a two-domain tree—with eukaryotes evolved from archaea through serial endosymbiosis or syntrophic merger with alphaproteobacteria (mitochondrial ancestor). Plant phylogenomics using hundreds of genes from thousands of species resolved the Angiosperm Phylogeny Group classification, resolving previously intractable rapid radiation events including the rapid diversification of early angiosperms (Darwin's 'abominable mystery').

Example 7: Mutation Rate Estimation

Human germline mutation rate (~1.1×10-8 per bp per generation, ~35 mutations per generation) is estimated from complete genome sequencing of parent-offspring trios identifying de novo mutations. Paternal age effect: sperm undergo 23 cell divisions per year while oocytes undergo fixed division number, so de novo mutation rate increases by ~1-2 mutations per additional year of paternal age. De novo mutations are enriched in autism spectrum disorder, schizophrenia, and intellectual disability, explaining why these conditions persist despite reproductive fitness reduction (new mutations from older fathers replenish the pool). Somatic mutation rates are 10-100 times higher than germline, reflecting cell-type-specific replication and repair differences estimated from large-scale cancer genome and normal tissue sequencing studies.

Example 8: Molecular Systematics of Microbes

16S rRNA gene sequencing enabled culture-independent identification of the vast majority of microbial diversity—originally estimated that 99% of environmental bacteria cannot be cultured. SILVA and Greengenes databases contain millions of classified 16S sequences enabling taxonomic assignment from sequencing alone. Environmental genomics (metagenomics) recovered entirely new phyla undetectable by culture—Candidate Phyla Radiation (CPR/Patescibacteria), DPANN archaea—dramatically expanding the known tree of life. Tara Oceans expedition metagenomics characterised global ocean microbial diversity, discovering millions of novel viral species and hundreds of novel microbial protein families with unknown function, revealing the biological dark matter of marine ecosystems.

Try it live

Everything above runs in your browser — open Molecular Clock & Phylogenetic Tree Simulator and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.