Robustness Increases Variability: A Fundamental Law of Biology (Mathematical Sciences/biomathematics)
Abstract
This work develops a fundamental paradox of system design. The more robust protection a system has against perturbations, the more random variability and failure there will be in the system s components. Suppose, for example, that a system can correct errors arising when it copies its internal data. That robust system correction tolerates a greater error rate in the component subsystems that copy the data. In general, as systems improve in their robustness, their components tend to fluctuate more and decay in performance. That duality between system robustness and component variability and decay helps to understand the intrinsic fragilities of system designs and the ways in which complex systems tend to fail. The proposed work develops these theoretical concepts of how system designs tend to evolve and tend to fail. The theory will be applied to the design of biological and human-engineered systems. In biology, common protective mechanisms work to correct internal errors. For example, when an organism detects that it is too hot or too cold, it stimulates physiological mechanisms to correct its temperature. When DNA is copied, many molecular mechanisms check for mistakes and, when mistakes are found, trigger other molecular mechanisms to correct those mistakes. The better an organism becomes at using additional corrective mechanisms to fix mistakes in its underlying DNA copying mechanisms, the less pressure there is on the underlying mechanisms to avoid errors. Correction at the system level allows sloppiness of the underlying components. How does that duality between system correction and component sloppiness actually happen in biological designs? Why do some components vary a lot whereas other components vary only a little? How does the variability or failure of system components lead to disease? To analyze these questions, the proposed work will develop mathematical and computer models of evolutionary process in relation to the dynamics of robust design. This work will use the improved theoretical understanding to enhance how we analyze specific problems in biology. For example, humans have a complex set of layered protective mechanisms to repair DNA and cellular damage, and to limit the growth of cells that are damaged beyond repair. How, through evolutionary history, does such a complex layering of protective mechanisms actually evolve? In theory, each additional protective mechanism reduces the pressure on the efficacy of the existing mechanisms, because the new protective mechanism will correct some of the mistakes that passed through the prior corrective systems. Thus, the layering of each new protective mechanism allows greater variability and failure in the existing mechanisms. The final overall system design is a seemingly overwired and highly variable set of components. And that is exactly how cancer protective mechanisms and much of biological system design actually look. How do fragilities in such overwired systems with weakened components arise, and how do systems tend to fail? This proposal will improve understanding of those questions through explicit theory and through application of the concepts to specific problems, including the complex molecular wiring of cellular control systems and the tendency for some molecular components to vary a lot between cells. This project will also apply the biological theory to problems of systems designed by human engineering, such as computer data storage systems. In addition, social systems in both biology and in human organizations often have layered corrective mechanisms to protect against errors. How do the robust corrective mechanisms in social systems lead to variability and failure in the individual components of the system? How does the duality between system robustness and component decay lead to fragilities in social organizations and in command and control hierarchies?
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jul 09, 2020
- Source ID
- W911NF2010227
Entities
People
- Steven Frank
Organizations
- Army Contracting Command
- United States Army
- University of California, Irvine