Are animal models as good as we think?

Models have been a tool of science at least since the 18th century and serve a variety of purposes from focusing abstract thoughts to representing scaled down version of things for study. Generally, animal models are needed when it is impractical or unethical to study the target animal. Biologists have taken modeling by analogy beyond most other disciplines, deriving the relationship between model and target through evolution. The "unity in diversity" concept suggests that homology between model and target foretells functional similarities. Animal model studies have been invaluable for elucidating general strategies, pathways, processes and guiding the development of hypotheses to test in target animals. The vast majority of animals used as models are used in biomedical preclinical trials. The predictive value of those animal studies is carefully monitored, thus providing an ideal dataset for evaluating the efficacy of animal models. On average, the extrapolated results from studies using tens of millions of animals fail to accurately predict human responses. Inadequacies in experimental designs may account for some of the failure. However, recent discoveries of unexpected variation in genome organization and regulation may reveal a heretofore unknown lack of homology between model animals and target animals that could account for a significant proportion of the weakness in predictive ability. A better understanding of the mechanisms of gene regulation may provide needed insight to improve the predictability of animal models.