Written and maintained by CASRAI Editorial Board
Last updated
Short answer: A Turing pattern is a stable spatial pattern, such as spots or stripes, that arises from a uniform starting state when chemicals react with each other and diffuse at different rates. Alan Turing proposed the idea in 1952. In the classic version a self-promoting activator spreads more slowly than the inhibitor it triggers, which sets the spacing of the pattern.
Definition and scope
A Turing pattern is the outcome of a reaction-diffusion mechanism: substances (morphogens) are produced and consumed by local reactions while also spreading through tissue. Turing’s paper, “The Chemical Basis of Morphogenesis” (Philosophical Transactions of the Royal Society of London B, 237(641):37-72, 1952, doi:10.1098/rstb.1952.0012), proposed that reacting and diffusing morphogens could be sufficient to account for the main phenomena of morphogenesis. He analysed a ring of cells to show how a uniform state becomes unstable, and suggested candidates such as Hydra tentacles, whorled leaves and phyllotaxis. The identity of the morphogens was unknown at the time.
The term covers a theoretical mechanism and the patterns it predicts. Whether a given biological pattern actually arises this way is a separate empirical question, discussed below. For the wider field this belongs to, see the CASRAI guide to mathematical biology.
How the mechanism works
In the activator-inhibitor description summarised in a PubMed Central review (PMC5697439, read 2026-11-08), an activator promotes its own production and also triggers a second chemical. That second chemical diffuses faster and acts as a lateral inhibitor, suppressing activator production in the surrounding region. The balance of reaction and diffusion rates fixes the spacing of the pattern, sometimes called its chemical wavelength.
- Local activation: the activator amplifies itself where its concentration is already higher.
- Long-range inhibition: the inhibitor spreads further than the activator, so peaks cannot form close together.
- Spacing from rates: the distance between spots or stripes follows from the reaction and diffusion rates, not from a pre-existing template.
The same review notes that other network types can also produce Turing patterns, including substrate-depletion models and models in which an activator precursor breaks down locally. A paper by Smith and Dalchau, “Beyond activator-inhibitor networks: the generalised Turing mechanism” (arXiv:1803.07886), argues that a wide variety of networks can generate stable Turing patterns, for example systems built from two self-activating components, so the activator-inhibitor pair is not the only route.
Variants and related concepts
| Concept | What it means | Source |
|---|---|---|
| Classic (morphostatic) Turing model | The chemical pattern forms first and then acts on cells; cell movement is not part of creating the pattern. | PMC5697439 |
| Morphodynamic model | Cell movement is part of how the pattern develops, so a purely chemical account is incomplete. | PMC5697439 |
| Substrate-depletion and local-breakdown networks | Alternative reaction networks that can also give Turing patterns. | PMC5697439 |
| Generalised Turing mechanism | Many network designs, including two self-activators, can yield stable patterns. | Smith and Dalchau, arXiv:1803.07886 |
Where it is supported and where it is contested
Models are easy to write; showing that one operates in a real tissue is harder. Two sourced examples show the range:
- Digit patterning: A review by Cooper (2015) describes a Turing mechanism proposed for digit patterning. It cites work by Sheth and colleagues (2012) in which Hox genes regulate digit patterning by controlling the wavelength. The review notes that the core activator and inhibitor have not been identified.
- Galectin-based models: The PMC5697439 review states that the generalizability of the two-galectin models across vertebrate clades is unknown, and that Gal1 null mice lacking a limb phenotype is a challenge to the model.
Checklist: evidence behind a claim of a Turing mechanism
This checklist is an editorial synthesis of the limits reported in the sources above, not a formal standard. Use it to read or write a methods section.
- Components identified: are the proposed activator and inhibitor named molecules, or only placeholders? (The digit-patterning case lacks identified core components.)
- Rate asymmetry shown: does the work show or justify that the inhibitor effectively spreads further than the activator, rather than assuming it?
- Perturbation test: does removing or altering a proposed component change the pattern as the model predicts? (The Gal1 null result is an example of a perturbation that did not match.)
- Cell behaviour considered: does the model allow for cell movement, as in morphodynamic accounts, where relevant?
- Generality stated: is the claim limited to the species or tissue tested, given that generalizability across clades can be unknown?
Where it matters for modellers and research groups
Turing models are usually reported as equations with parameters, so a reader needs the reaction terms, diffusion coefficients, boundary conditions and simulation settings to reproduce a figure; see the dictionary entry on reproducibility. When a pattern is instead produced by simulated individual cells, the CASRAI guide to the agent-based model covers structured reporting. Theoretical work of this kind is published in venues such as the Journal of Theoretical Biology.
Common misunderstandings
- “Turing patterns are proven to cause animal markings.” The mechanism is well established as a theoretical possibility; for specific tissues, support varies and is contested, as in the digit and galectin examples above.
- “It always needs an activator and an inhibitor.” Other network types can also produce the patterns, per the generalised Turing mechanism.
- “A pattern that looks like Turing spots is a Turing pattern.” Similar appearance does not show mechanism; that requires the evidence listed in the checklist.
FAQ
What are Turing patterns in simple terms?
Spatial patterns, such as spots and stripes, that form from a uniform state when interacting chemicals spread at different rates.
Who proposed them, and when?
Alan Turing, in “The Chemical Basis of Morphogenesis,” published in 1952.
What sets the size of the spots or stripes?
In the classic description, the balance of reaction and diffusion rates sets the spacing, called the chemical wavelength.
Are Turing patterns seen in real organisms?
They have been proposed for several systems, including digit patterning, but the evidence is incomplete or contested for the examples above.
Sources
- Turing AM. The Chemical Basis of Morphogenesis. Phil Trans R Soc Lond B 1952;237(641):37-72. doi:10.1098/rstb.1952.0012
- PubMed Central review PMC5697439, pmc.ncbi.nlm.nih.gov/articles/PMC5697439 (read 2026-11-08)
- Cooper KL. 2015 review on digit patterning, UC San Diego Cooper lab (read 2026-11-08)
- Smith SA, Dalchau N. Beyond activator-inhibitor networks: the generalised Turing mechanism. arxiv.org/abs/1803.07886 (read 2026-11-08)
