Phylogenetic Analysis — Evolutionary Tree Reconstruction from Molecular Data
Phylogenetic Analysis of Molecular Sequence Data · Also known as: molecular phylogenetics, phylogenetic inference, evolutionary tree reconstruction, phylogenomics
Phylogenetic analysis reconstructs the evolutionary history of organisms, genes, or proteins by comparing molecular sequence data and estimating the branching tree that best explains observed similarities and differences. Rooted in the work of Felsenstein and colleagues from the 1960s onward, it is a cornerstone technique in evolutionary biology, microbiology, epidemiology, and comparative genomics, supporting tasks from tracing viral outbreak origins to classifying novel species.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
+3 more
When to use it
Use phylogenetic analysis when the central question concerns evolutionary relationships, common ancestry, divergence times, or the spread of sequences such as viral lineages over time or geography. It is appropriate whenever homologous sequences can be obtained for all taxa and aligned reliably. It is particularly powerful for tracing infectious disease outbreaks, classifying novel organisms, dating divergence events with calibrated molecular clocks, and studying gene family evolution. Do not use it when sequences are too short or too diverged to align reliably, when the question concerns within-population genetic variation (use GWAS or admixture analysis instead), or when horizontal gene transfer or hybridisation is so pervasive that a tree model is inappropriate and phylogenetic network methods should be used instead.
Strengths & limitations
- Provides an explicit, testable evolutionary hypothesis with quantified branch support rather than a mere cluster diagram.
- Maximum likelihood and Bayesian methods are statistically principled and handle substitution saturation and rate heterogeneity through explicit substitution models.
- Scales from a handful of sequences to thousands of taxa with modern tools such as IQ-TREE and RAxML-NG.
- Bayesian inference via BEAST integrates temporal information to estimate divergence dates under a molecular clock model.
- Applicable across the full spectrum of molecular data including mitochondrial markers, whole genomes, transcriptomes, viral genomes, and protein families.
- The inferred tree is only as good as the sequence alignment; errors in alignment propagate directly into topology errors.
- Model misspecification using an overly simple substitution model can produce statistically confident but biologically incorrect trees.
- A single gene tree may not reflect the species tree due to incomplete lineage sorting, gene duplication, or horizontal gene transfer.
- Bayesian MCMC analyses can be computationally intensive and require careful convergence diagnostics including effective sample size and trace plots.
- Long-branch attraction causes fast-evolving lineages to cluster artifactually with the outgroup under parsimony or poorly specified maximum likelihood models.
Frequently asked
Which tree inference method should I use — neighbor-joining, maximum likelihood, or Bayesian?
For most modern analyses, maximum likelihood with IQ-TREE or RAxML is the default choice: it is statistically principled, fast on large datasets, and provides bootstrap support. Neighbor-joining is acceptable for rapid exploratory work or very large sequence sets where runtime is prohibitive. Bayesian inference via MrBayes or BEAST is preferred when you need full posterior distributions on topologies and branch lengths or when integrating temporal information for molecular clock dating. Parsimony is now rarely recommended for molecular data due to inconsistency under long-branch attraction.
How do I know if my bootstrap support values are trustworthy?
Bootstrap values above 70-75 are conventionally considered moderate support and above 95 is strong support. However, high bootstrap values can still accompany a wrong tree if the data are systematically misleading due to model misspecification or long-branch attraction. Always interpret support values alongside biological plausibility and, where possible, corroborate with independent loci or morphological evidence.
What is incomplete lineage sorting and why does it matter?
Incomplete lineage sorting occurs when ancestral polymorphism persists across multiple speciation events, causing individual gene trees to differ from the true species tree. It is especially common in rapid radiations and recently diverged taxa. If you are working with multiple loci across closely related species, concatenation alone may recover the wrong species topology with high confidence. Use a species-tree method such as ASTRAL (coalescent-based) or BEAST multispecies coalescent to account for this discordance.
Can I use whole-genome data for phylogenetics?
Yes. Phylogenomics using hundreds or thousands of loci is now standard and greatly increases resolution. It requires careful ortholog identification, alignment trimming, and handling of missing data. Concatenation with partition models in IQ-TREE and coalescent species-tree methods such as ASTRAL are the dominant approaches. For prokaryotes where horizontal gene transfer is common, a core-genome or recombination-filtered approach is advisable before inferring a species tree.
My sequences are from very divergent taxa — is alignment still reliable?
High divergence above roughly 40-50% nucleotide identity or 25% protein identity makes alignment increasingly unreliable and phylogenetic signal may be obscured by substitution saturation. In such cases, use protein-coding data aligned at the amino acid level, trim ambiguous columns aggressively, apply saturation tests, and consider whether a network or supertree approach is more appropriate than a bifurcating tree.
Sources
- Felsenstein, J. (2004). Inferring Phylogenies. Sinauer Associates. ISBN: 978-0878931774
- Felsenstein, J. (1981). Evolutionary trees from DNA sequences: A maximum likelihood approach. Journal of Molecular Evolution, 17(6), 368-376. link ↗
How to cite this page
ScholarGate. (2026, June 3). Phylogenetic Analysis of Molecular Sequence Data. ScholarGate. https://scholargate.app/en/bioinformatics/phylogenetic-analysis
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- eQTL AnalysisBioinformatics↔ compare
- Genome-wide association studyBioinformatics↔ compare
- RNA-seq Differential ExpressionBioinformatics↔ compare
- Sequence AlignmentBioinformatics↔ compare
- Variant CallingBioinformatics↔ compare