Perfect model

Biologists use a variety of animal models for experiments. But a mouse or a fruitfly cannot truly represent a human being. Jessica Bolker, associate professor at the department of biological sciences of University of New Hampshire in the US, discusses with Vibha Varshney the need for diverse animal models to study the impact of the environment on developmental biology

By Vibha Varshney
Published: Saturday 15 December 2012

Jessica BolkerWhy do we need a wide variety of experimental models?

Biology is a diverse subject and we cannot appreciate or understand that diversity unless we look at a wide range of models. The few animal models that dominate experimental research in biological and biomedical fields are not sufficient to answer all the questions. These models are poorly suited to answer questions asked by scientists in emerging fields like ecological developmental biology which focuses on external influences on the development of an organism.

How does having a limited number of models adversely affect research output?

There are several examples where results of model-based research have been, at best, irrelevant and, at worst, misleading; particularly in translational research that aims to apply basic science to clinical practices.

A subtler but very important effect of using limited number of models is that some problems may not get addressed at all while others are addressed using models not necessarily well-suited for the solution. For example, in case of Parkinson’s disease potential treatments are often assessed by measuring muscle movement in rat model.

But this model does not clearly represent other significant symptoms of Parkinson’s in human patients like cognitive decline. Disparities between animal models and humans may explain why millions of dollars spent on research have yielded frustratingly few clinical advances. Few lab models account for the environment the animals are kept in, despite increasing recognition that this may affect the outcome of the experiment.

What should be the criteria to choose a research model?

The key thing is to identify, as part of experimental design, exactly what the model needs to do or represent. For instance, is it important that the model has particular physiological characteristics or metabolic pathways, or be of a similar size or share a certain disease pathway with humans as a basis for testing possible therapies? Which trait is important for a given study depends on what the study is focused on, how it will be done, and the nature of the answers we hope to get out of it.

Are some animal models more suited for certain experiments than others?

Pigs are well-suited for tests on drug metabolism or surgical techniques because their body size and organ functions are similar to humans. But they are impractical for genetic studies. In contrast, the techniques available for studying genes in mice make them a good model to study mammalian genes, but the difference in metabolic rates may make mice less reliable as a model for other studies.


Dog breeds with naturally occurring gene mutations are good models for studying human genetic diseases and their possible treatments. We need to broaden our range of models to include species such as Antarctic icefish, comb jellies, cichlids, dune mice and finches that naturally have features relevant for studies on human diseases.

Can we compare results of experiments on existing models with those on more suitable models?

I don’t think anything can ensure that results are comparable; rather, in each case we need to assess what aspects of the model are relevant to the question at hand. If those aspects are aligned from one model to the other, chances are the results too will match. And if different models yield different results, then we learn something very important about the models’ ability (or the lack of ability thereof) to serve as a basis for generalisation.

There is advocacy against using animals for research. What is your opinion on the alternatives?

Though I firmly believe that use of animals in research must be carried out in a careful and humane manner, I think animal models are, and will remain, a necessary part of biological and biomedical research. The general principle that the necessary characteristics of a model are driven by the nature of the question applies here, too. If we want to examine cellular metabolism, then cells in labs may well be the best way to do it. But for studying organism-scale mechanisms and influences, a cell- or tissue-culture model may not yield a full answer as it does not adequately represent all the important aspects of an organ system.

Subscribe to Daily Newsletter :

Comments are moderated and will be published only after the site moderator’s approval. Please use a genuine email ID and provide your name. Selected comments may also be used in the ‘Letters’ section of the Down To Earth print edition.