Skip to main content

The Importance of Initial Conditions

The reason nonhuman animals cannot be predictive models for human response to drugs and diseases involves the fact that all animals are examples of evolved complex systems. I have addressed what complexity means in science (see Animal models and conserved processes, for example), including the fact that complex biological systems are dependent on initial conditions. Relevant to complexity, the fact that animals and humans evolved has implications for, among other things, the initial conditions of various species. As I have pointed out, variation in initial conditions also explains why individual humans respond differently to drugs and disease. This was recently reinforced by Fu et al. (2012).

Fu et al. examined “15,336 genes in 6,515 individuals of European American and African American ancestry and inferred the age of 1,146,401 autosomal single nucleotide variants (SNVs).” They discovered that “approximately 73% of all protein-coding SNVs and approximately 86% of SNVs predicted to be deleterious arose in the past 5,000–10,000 years,” and that “disease genes contained a significantly higher proportion of recently arisen deleterious SNVs than other genes.” (Fu et al. 2012) A great amount of variation has happened since humans last shared a common ancestor with another species. Therefore, even a close evolutionary relationship, such as exists between humans and chimps for example, will be inadequate in terms of expecting predictive value from that species.

Along the same lines, a study published in PLoS Biology on November 20, 2012 reveals: “hundreds of small regions of the genome that appear to be uniquely regulated in human neurons,” according to a Public Library of Science press release. These differences may explain our intellect compared to other primates as well as why certain diseases appear to be human-specific. Humans living in a certain area of the Argentinian Andes have a gene variant that allows them to consume higher amounts of arsenic. The variant results in less toxic metabolism. This means that the process of evolution is happening to humans in a time frame that we can appreciate. Scientists have discovered that even cells from the same individual differ in genetic make-up. O'Huallachain et al. discovered copy number variations (CNVs) in healthy cells from the same individual. CNVs had previously been discovered in stem cells. (O'Huallachain et al. 2012) Research from Stanford and Yale also suggests that there is intra-individual variability in DNA. (Abyzov et al. 2012) This has implications for drug and disease research. As I have stated many times, a complex system has a hierarchy of levels of organization and how a perturbation affects one level may vary from how it affects a different level. Apparently, how a perturbation affects one cell may differ from how it affects another.

The above reinforces statements from scientists I have previously quoted in this blog. For example, Belmaker et al:

Individual differences in response to pharmacologic treatment limits the usefulness of mean data obtained from randomized controlled trials. These individual differences exist even in genetically uniform inbred mouse strains. While stratification can be of value in large studies, the individual patient history is the most effective currently available guide for personalized medicine in psychopharmacology. (Belmaker, Bersudsky, and Agam 2012)

Moreover, the same genes can be used differently by different species so even when initial conditions appear the same, important differences may exist. An October 12, 2012 press release from Michigan Technological University announced that Werner and Raja have discovered that the same three genes that code for the spots on the body of fruit flies also cause cancer or other diseases in humans. Flies shared a common ancestor with humans ~600 MYA. Werner states: “All the genes needed to build a body were already present in that ancestor, and today we still share virtually all of our body-building genes with fruit flies. This is why we are able to study human diseases like cancer in fruit flies.” No, actually it doesn’t mean that at all. It means that different species use the same genes for different functions and this again means more differences in initial conditions.

The common flaw in suggesting that we can use flies to study human cancer as well as using other species in similar fashion is an irrational dependence on reductionism. Reductionism has an important place in biomedical research but complex systems cannot be fully understood by reductionism alone. Yet, this notion persists in biomedical research as typified by the following statement from Harvard:

It's one of the basic tenets of biological research – by studying simple "model" systems, researchers hope to gain insight into the workings of more complex organisms. Caenorhabditis elegans –tiny, translucent worms with just 302 neurons – have long been studied to understand how a whole nervous system is capable of translating sensory input into motion and behavior.

(As it turns out, even C. elegans had some surprises for scientists.)

Complex systems have many layers of organization. When one layer is somewhat understood, you may not be any closer to understanding the whole. Yong, writing in The Scientist refers to this when he discusses using pigs for organ transplants:

The organs from the genetically modified pigs were no longer instantly rejected when tested in nonhuman primates, but subtler problems appeared. “It’s like an onion,” says Sachs. “You take away a layer and you get another.”(Yong 2012)

Koch also alludes to this in his discussion of the complex systems like the human brain:

Such systems are characterized by large numbers of highly heterogeneous components, be they genes, proteins, or cells. These components interact causally in myriad ways across a very large spectrum of space-time, from nanometers to meters and from microseconds to years. A complete understanding of these systems demands that a large fraction of these interactions be experimentally or computationally probed. This is very difficult. . . . fields as diverse as neuroscience and cancer biology have proven resistant to facile predictions about imminent practical applications. Improved technologies for observing and probing biological systems has only led to discoveries of further levels of complexity that need to be dealt with. This process has not yet run its course. We are far away from understanding cell biology, genomes, or brains, and turning this understanding into practical knowledge. (Koch 2012)

The notion of evolved complex systems is not intellectually challenging hence one must wonder why, in light of the volumes of data being published on the differences in initial conditions among species, the animal model community appears to be ignoring the implications. There are many possible explanations for this, including one outlined by Massimo Pigliucci in his blog: scientists simply do not keep up with developments outside their field of expertise and will use creative reasoning to get their research funded regardless of applicability. Scientists are human and humans are prone to dishonesty.

A recent example of such dishonesty, that was eventually corrected, was a study out of Harvard that suggested aspartame causes cancer. The university retracted the press release it had sent out because it was so misleading. The entire issue of PR departments and scientists exaggerating conclusion of research in press releases has been the subject of comment by many. (See here and here.)

Bolker puts the above in perspective:

For most experimental biologists, life revolves around a handful of species: the mouse (Mus musculus), the nematode worm (Caenorhabditis elegans), the fruitfly (Drosophila melanogaster) and the thale cress (Arabidopsis thaliana). We assume that model organisms offer universal insights, and funding agencies largely support work on a shortlist of favoured species ( . . . Disparities between mice and humans may help to explain why the millions of dollars spent on basic research have yielded frustratingly few clinical advances (Davis 2008; von Herrath and Nepom 2005; Geerts 2009; Schnabel 2008). (Bolker 2012)

An Institute of Medicine workshop, “Accelerating the Development of New Drugs and Diagnostics,” similarly reported:

[Joshua] Boger [founder of Vertex] remarked that Vertex has put three drugs into the market and none has had an animal model. He termed animal models “overrated,” . . . On the other hand, [James Bradner of Harvard] noted, there is no obvious animal model for sickle cell disease, but that has not been a barrier to moving the science forward.(IOM 2012)

Animal models are indeed overrated in addition to being misleading and a waste of scarce research funds. In a more rational universe, governments would acknowledge the obvious and change policies. In our universe however, governments listen to lobbyists.


Abyzov et al. 2012. Somatic copy number mosaicism in human skin revealed by induced pluripotent stem cells. Nature advance online publication.

Belmaker, R., Y. Bersudsky, and G. Agam. 2012. Individual differences and evidence-based psychopharmacology. BMC Medicine 10 (1):110.

Bolker, Jessica. 2012. Model organisms: There's more to life than rats and flies. Nature 491 (7422):31-33.

Davis, M. M. 2008. A prescription for human immunology. Immunity 29 (6):835-8.

Fu et al. 2012. Analysis of 6,515 exomes reveals the recent origin of most human protein-coding variants. Nature advance online publication.

Geerts, H. 2009. Of mice and men: bridging the translational disconnect in CNS drug discovery. CNS drugs 23 (11):915-26.

IOM. 2012. Accelerating the Development of New Drugs and Diagnostics: Maximizing the Impact of the Cures Acceleration Network: Workshop Summary. Washington DC: National Academies Press.

Koch, C. 2012. Systems biology. Modular biological complexity. Science 337 (6094):531-2.

O'Huallachain, M., K. J. Karczewski, S. M. Weissman, A. E. Urban, and M. P. Snyder. 2012. Extensive genetic variation in somatic human tissues. Proceedings of the National Academy of Sciences of the United States of America.

Schnabel, J. 2008. Neuroscience: Standard model. Nature 454 (7205):682-5.

von Herrath, M. G., and G. T. Nepom. 2005. Lost in translation: barriers to implementing clinical immunotherapeutics for autoimmunity. The Journal of Experimental Medicine 202 (9):1159-62.

Yong, Ed. 2012. Replacement Parts. The Scientist (August).


Popular Video