Heterogeneity Mediated System Complexity: The Ultimate Challenge for Studying Common and Complex Diseases

Abstract

Influenced by the gene-centric conceptual framework, current genetic studies of disease mainly focus on the characterization of gene mutations or other molecular mechanisms. It is generally accepted that gene mutations are the drivers of diseases. However, it has been very hard to identify these common drivers for most common and complex diseases, despite extensive efforts spanning decades. Furthermore, various—omics based studies have illustrated that the common feature of most common diseases is genetic and non-genetic heterogeneity at multiple levels of organization.

Heterogeneity is traditionally considered to be insignificant “noise”, and carefully designed research strategies are often needed to reduce or eliminate this in order to determine the signal pattern. However, if heterogeneity represents a key feature of disease, such approaches are fundamentally limited. It is challenging to translate this “noise free” knowledge back into “noise rich” systems. This paradoxical situation is reflective of the current status of disease research. On the one hand, we know much about the genes linked to these common diseases, and yet, it is extremely difficult to apply this gene-based knowledge to effectively treat these diseases.

The key to solving this dilemma is to directly study system heterogeneity, the key contributor to biological complexity. For example, without heterogeneity, cancer would not form in the first place. It is only when the degree of heterogeneity is used as an index of system instability or evolutionary potential that cancer evolution can be realistically studied. In fact, monitoring system behaviours such as stability often results in better prediction power than focusing on lower level components, as most disease systems are highly dynamic by nature.

Here we use cancer, Gulf War Illness and chronic fatigue syndrome as examples of complex diseases/illnesses to briefly illustrate these concepts. We will discuss how system inheritance and fuzzy inheritance or inherited heterogeneity contribute to diseases more strongly than individual genes, and why the time factor must be incorporated within the multiple level landscape model in order to explain the stochastic behaviour of disease progression and response to treatment. Stress-induced heterogeneity serves not only as a tool of cellular adaptation, but as a trade-off that also contributes to diseases.