Mutation Biology and Consequences

Short version. A mutation is a heritable change in DNA sequence. Mutations vary in scale (a single nucleotide to entire chromosomes), in mechanism (replication error, chemical damage, repair failure, retrotransposition, slippage at repeats), and in consequence (silent, regulatory, protein-altering, structural, dominant, recessive, lethal). Modern variant interpretation rests on a published framework (the ACMG/AMP 2015 guidelines, refined by ClinGen) that combines population-frequency, computational, segregation, and functional evidence into a five-tier classification. This pillar introduces three subtopics: types of mutation; functional consequences of mutation; and mutation detection and interpretation.

A short history

The modern study of mutation is a twentieth-century discipline. Hermann Muller's 1927 demonstration that X-rays could induce heritable variation in Drosophila established that mutation rates are not fixed by nature but are sensitive to the environment. Two decades later, Charlotte Auerbach and J. M. Robson, working in Edinburgh during the Second World War, showed in 1947 that nitrogen mustard (an alkylating agent used as a chemical weapon) was strongly mutagenic — the first chemical demonstrated to act like ionising radiation in inducing genetic change. Their work, classified during the war and published only after, founded chemical mutagenesis and grounded the field in mechanism rather than phenomenology.

The discovery that DNA is the genetic material (Avery, MacLeod, McCarty 1944; Hershey-Chase 1952) and the elucidation of the double helix (Watson, Crick, Franklin, Wilkins 1953) gave mutation a chemical address. By the 1960s, Benzer's fine-structure mapping of the rII locus in T4 phage had reduced mutation to base-pair resolution, and the demonstration that the genetic code was universal allowed any nucleotide change to be read as a protein-level event. The molecular era of mutation biology had begun.

How often does mutation happen?

The mutation rate is not a single number. It varies by organism, by genomic region, by sequence context, and by the kind of mutation under consideration. The most-cited synthesis for the spontaneous rate across organisms is John Drake's 1991 paper in PNAS (88:7160), which proposed a roughly constant rate of about 0.003 mutations per genome per generation across DNA-based microbes — a regularity that has come to be called Drake's rule.

For humans, the analogous figure was established by direct measurement only with the arrival of trio whole-genome sequencing. Kong et al. 2012 (Nature 488:471), sequencing 78 Icelandic trios, estimated the de novo single-nucleotide variant (SNV) rate at approximately 1.2 × 10⁻⁸ per site per generation, equivalent to roughly 60–80 new SNVs in each newborn relative to their parents. They also showed that paternal age increases the rate substantially: roughly two additional mutations per year of paternal age, because spermatogonial stem cells divide repeatedly across the male reproductive lifespan. Maternal age has a smaller effect on de novo SNVs but a larger effect on aneuploidy.

Mutation rates differ by class. Single-nucleotide changes are common; small indels are roughly an order of magnitude rarer; structural variants rarer still per genome but each contributes far more sequence change. Repeat expansions, mitochondrial mutations, and somatic mutations follow their own dynamics, addressed in the types of mutation page.

Subtopic 1: Types of mutation

The first concern of mutation biology is descriptive: what kinds of change exist, and at what scale? The conventional taxonomy starts with point mutations — single base substitutions, classified as transitions (purine→purine, pyrimidine→pyrimidine) or transversions (purine→pyrimidine and back). Substitutions are then classified by coding consequence: synonymous (no amino-acid change), missense (one amino-acid change), nonsense (premature stop codon), or splice-site disrupting.

Insertions and deletions (indels) are the next scale up. Small indels can preserve the reading frame (in-frame, multiples of three) or disrupt it (frameshifts). Microsatellite slippage at short tandem repeats produces a characteristic class of indels. Trinucleotide repeat expansion, a peculiar class of dynamic mutation, gives rise to disorders such as Huntington disease, fragile X syndrome, Friedreich ataxia, myotonic dystrophy, and the spinocerebellar ataxias; the expansion size grows across generations, producing the phenomenon of anticipation.

Above the indel scale lie structural variants — large deletions, duplications, inversions, translocations — and copy-number variants. Mosaicism, the presence of multiple cell-genotype lineages within an individual, is a crosscutting category: somatic mosaicism arises during development and is a major substrate of cancer; germline mosaicism (covered on our germline mosaicism calculator page) explains apparent de novo recurrences within a sibship. Mutational signatures — the patterns of trinucleotide context that different mutagenic processes produce — were characterised systematically by Alexandrov et al. 2013 (Nature 500:415) and form the basis of the COSMIC mutational signatures catalogue. The full taxonomy is on the types of mutation subtopic page.

Subtopic 2: Functional consequences of mutation

Description is not enough. The geneticist also asks what a mutation does. Loss of function (LoF) variants — nonsense, frameshift, large deletion, canonical splice-site disrupting — abolish or severely reduce the protein product of the affected allele. Whether this is consequential depends on dosage sensitivity: if the gene is haploinsufficient, a single LoF allele produces a phenotype (autosomal dominant); if not, two LoF alleles are needed (autosomal recessive). The systematic survey of LoF in healthy human exomes by MacArthur et al. 2014 (Nature 508:469) showed that even healthy individuals carry roughly 100 LoF variants, of which about 20 affect both copies of the gene — a foundational observation for population-level interpretation.

Gain-of-function variants change protein behaviour rather than removing it. Constitutive activation, novel binding partners, increased expression, and stabilisation of a dosage-sensitive product can all produce a gain-of-function phenotype. The classic example is FGFR3 G380R in achondroplasia, where a single recurrent missense substitution produces ligand-independent activation of the receptor.

Dominant-negative variants are a third category: a mutant subunit poisons the function of a multimeric complex. Type I and type IV collagen variants in osteogenesis imperfecta and Alport syndrome are textbook examples; mutant subunits are incorporated into the triple helix and destabilise it. Splice-disrupting and regulatory variants form a fourth category whose interpretation has been transformed by deep-learning predictors such as SpliceAI (Jaganathan et al. 2019, Cell 176:535). The full treatment is on the functional consequences of mutation subtopic page.

Subtopic 3: Detection and interpretation

The third pillar of mutation biology is methodological. How are variants detected, and how are they interpreted? Sanger sequencing, which dominated the 1980s and 1990s, has largely been replaced for screening purposes by short-read next-generation sequencing on Illumina platforms (described in Bentley et al. 2008, Nature 456:53) and is increasingly supplemented by long-read platforms (PacBio HiFi, Oxford Nanopore) that resolve repeat expansions and structural variants short reads miss.

Variant interpretation has been standardised. The ACMG/AMP 2015 guidelines (Richards et al., Genetics in Medicine 17:405) define a five-tier classification — pathogenic, likely pathogenic, uncertain significance, likely benign, benign — built from population-frequency evidence (gnomAD), computational predictors (REVEL, CADD, AlphaMissense), segregation evidence within families, and functional evidence at the bench. ClinGen variant-curation expert panels apply and refine these criteria gene-by-gene, and ClinVar publishes the resulting interpretations. The full treatment, including the interpretive frameworks, the major databases, and how the field has converged on a shared vocabulary, is on the mutation detection and interpretation page.

Where Evagene fits

Evagene is an academic, research, and educational pedigree modelling platform. It is not a sequencing service; it is not a variant-calling pipeline; it does not return classifications under the ACMG/AMP framework. The platform draws pedigrees, manages family-history information, and runs implementations of published family-history-based risk-model algorithms (Claus 1994, Couch 1997, Frank 2002, Evans 2004, Vasen 1999, Umar 2004, Gail 1989, Tyrer / Duffy / Cuzick 2004 as an IBIS-style approximation, BayesMendel BRCAPRO / MMRpro / PancPRO, and family-history scoring) for teaching, research, and exploratory use. Where computation against the canonical BOADICEA implementation is wanted, Evagene exports a ##CanRisk 2.0 pedigree file the user uploads at canrisk.org; BOADICEA is licensed by the University of Cambridge and is not bundled in Evagene.

The connection between this educational pillar and the platform is conceptual. A pedigree records the segregation of phenotypes across generations — the very pattern that mutation produces. Reading a pedigree well requires knowing what kinds of mutation exist, what their consequences are, and how a candidate variant moves from "found by sequencing" to "interpreted with confidence". The three subtopic pages elaborate.

Three subtopic pages

Types of mutation — substitutions, indels, repeat expansions, structural variants, copy-number variants, mosaicism, de novo mutation, the paternal age effect, and mutational signatures.
Functional consequences of mutation — loss of function, gain of function, dominant negative, haploinsufficiency, splice and regulatory variants, allele-frequency interpretation.
Mutation detection and interpretation — sequencing platforms, variant calling, annotation, the ACMG/AMP five-tier framework, gnomAD, ClinVar, in silico predictors.

Caveats and educational positioning

This page and the three subtopic pages are educational. They present published frameworks and cite the original literature so a student or researcher can follow the chain of evidence and read the original papers. The pages do not return a clinical interpretation, do not advise on testing decisions, and do not replace the work of a clinical geneticist or genetic counsellor. The ACMG/AMP framework is the published interpretive standard the field uses; it is not a service Evagene offers. Where a specific variant is in question, the appropriate route is a genetic-testing laboratory operating under the framework, not a marketing page.

Foundational sources cited

Auerbach C, Robson JM. Production of mutations by allyl isothiocyanate. Nature 1947 — nature.com/articles/157302a0 (mustard-gas mutagenesis foundational paper).
Drake JW. A constant rate of spontaneous mutation in DNA-based microbes. PNAS 1991;88:7160 — PMID 1831267.
Kong A, et al. Rate of de novo mutations and the importance of father's age. Nature 2012;488:471 — PMID 22914163.
Alexandrov LB, et al. Signatures of mutational processes in human cancer. Nature 2013;500:415 — PMID 23945592.
MacArthur DG, et al. A systematic survey of loss-of-function variants in human protein-coding genes. Nature 2014;508:469 — PMID 24482476.
Richards S, et al. Standards and guidelines for the interpretation of sequence variants. Genetics in Medicine 2015;17:405 — PMID 25741868 (the ACMG/AMP guidelines).
Jaganathan K, et al. Predicting splicing from primary sequence with deep learning. Cell 2019;176:535 — PMID 30661751 (SpliceAI).

Mutation biology and consequences