BLOG

Please post a comment

Michael Parkinson discuss experimental design and the factors involved in the efficient an ethical use of animals in scientific research.

Variability and scope.

Experimental material is inherently variable and one of the ways to control this variability is to make the experimental material more homogeneous. This cuts down on variability and makes the experiment more powerful. However one of the inherent dangers of this approach is that it may reduce scope and this can be very important. This is highlighted in a recent (26th june)  paper in Nature Communications which showed that most traits in mice showed a strong gender effect. Using only males to reduce variability can substantially reduce scope and you may, for example in a drug discovery programme,  end up with a drug that has dramatically reduced effects on females.  This is probably not what you want.

One way to get around this is to use more heterogeneous material, for example both males and females or outbred lines but this addition of extra variability will then require much larger sample sizes to show significant effects. With appropriate experimental design it is possible to factor in sources of variability to have the best of both worlds; small variability, big scope. This needs to be designed into the experiment.

Festing, Michael. (2010). Improving Toxicity Screening and Drug Development by Using Genetically Defined Strains. Methods in molecular biology (Clifton, N.J.). 602. 1-21. 10.1007/978-1-60761-058-8_1.

Mike