A virologist studying how viruses infect host cells, a developmental biologist tracking birth defects in an embryo, and an oncologist seeking new chemotherapeutic targets all use one common approach to discover new mechanisms: describing how a mutation exerts its effects to change molecular or morphological features.

Gain-of-function and loss-of-function are two widely used terms to describe how a specific genetic mutation changes protein structure, how it affects interactions with other molecular players and pathways, how it shapes cellular morphology, or tissue organization, or whole organism physiology. But on what basis do researchers label mutations as either gain-of-function or loss-of-function? Is there broad consensus among researchers on the meaning of these terms?

In the latest perspective article in GENETICS, Tobias Warnecke provides a historical account of how researchers settled on the term “gain-of-function”, points out substantial heterogeneity in how the term is used in practice—making it confusing—and asks whether there could be labelling system for gain-of-function mutations that is contextual and consistent. Warnecke demonstrates contradictions in how the term gain-of-function has been used by different scientists, emphasizing the pitfalls of using one simple descriptor for diverse, complex, and messy biological phenomena.

The etymology

First, Warnecke points out that the early geneticist Hermann Muller and other fly pushers classified mutations as hypomorphic, hypermorphic, amorphic, and neomorphic based on an increase or decrease of gene dosage and its impact on the appearance of mutant flies. Thirty years later, the developmental Drosophilist Ed Lewis introduced the terms loss- and gain-of-function, simplifying the myriad effects of mutations into two outcomes. These broader terms are prevalent in the literature, and mutations of all kinds as well as their physical manifestations are described by these two terms, often providing a misleading impression of the molecular mechanisms involved.

Loss versus gain: From molecular changes to systemic realities

To illustrate how the gain-of-function mutation label can mislead one’s intuition, Warnecke uses several examples of mutations and summarizes their mechanisms of action at the molecular, cellular, and organismal levels. While the functional impact of a mutation at a molecular and/or protein level might suggest a gain-of-function effect, the mutation can have a loss-of-function effect further down the molecular cascade or at a cellular or organismal level, and vice versa. According to Warnecke, the gain-of-function label is often derived from the level of biological complexity at which researchers chose to examine the effect of the mutation. From ion channels to prion proteins, to immune activators, a deeper look into known mutations depicts a common pattern: one researcher’s gain-of-function mutation is a loss-of-function mutation for others.

Incidentally, Warnecke argues that such inconsistent labeling of mutations can complicate efforts to train machine learning algorithms, especially in clinical applications. Warnecke discusses whether systemic re-labeling of mutations—for example by providing context on whether the mutation acts at the protein level or higher—could be useful to reliably employ machine learning approaches.

Overall, Warnecke’s thoughtful perspective is a reminder to geneticists, clinicians, and computational biologists that categorization of mutations into a limited set of categories often misses important complexity and is done very differently by different practitioners of science, which should be kept in mind when comparing mutations in bulk based on their labels.  

References

  • Warnecke T. Actually, what is a gain-of-function mutation?
    GENETICS. May 2026. 233(1).
    https://doi.org/10.1093/genetics/iyag059

Sejal Davla is a freelance science writer and data scientist with expertise in neuroscience and genetics. She is a motivated storyteller and works on projects at the intersection of science, data, and policy.

View all posts by Sejal Davla »

Read more in