OPUS: Robustness and complexity: how evolution builds precise traits from sloppy components

生物体经常修复其组织的损伤并纠正其生理上的错误。同样，人类经常试图让他们的工程系统强劲地运行，从损坏和错误中恢复过来。这个项目研究了自然生物过程和人类工程如何建立稳健运行的系统的异同。了解生物体赖以良好运转的原理有助于洞察事情有时是如何出错的，以及疾病是如何发展的。例如，我们的身体有许多不同的机制来预防癌症。在生物学和工程学上，具有多重保护的系统发生故障的时间遵循相同的模式。通过理解这些健壮和失败的一般特征，我们更好地理解了生物有机体是如何构建的，它们是如何失败的，以及这些原则如何也适用于人类工程系统。这位研究人员和许多其他科学家之前的研究已经形成了对健壮性和失败性原则的基本理解。然而，这些洞察力如何应用于生物学中的许多问题还没有完全发展起来。这个项目将把过去的不同研究结合起来，形成一个更完整的综合。这种综合将帮助我们将我们所知道的应用于更广泛的问题。这种合成还将帮助我们更清楚地看到我们不理解的东西，确定需要进一步研究的挑战。生物体实际上是如何从固有的随机和通常不可靠的生物成分中获得相对精确的特征的？马虎如何产生精确度？我们知道这个谜题的许多部分。冗余和额外容量提供备份。健壮性使生物体保持在正轨上。修复修复损坏。信号处理和控制遵循经典的工程原理。这些不同的调节特征的机制对进化动力学有着重要的影响。例如，当系统纠正组件级别的错误时，组件性能的直接选择压力会下降。对组件的选择性压力减弱可能会导致这些组件变得不那么可靠，保持更多的遗传可变性，或者在设计上保持中性漂移，为新形式的生物复杂性创造基础。为了理解这些拼图是如何拼凑在一起的，工程控制理论、计算机器学习和人工智能的最新进展增强了进化理论，提供了广泛的概念基础。这个项目建立在现代概念的基础上，以综合不同的生物学理论并统一各种生物学观测。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Organisms often repair damage to their tissues and correct errors in their physiology. Similarly, humans often try to make their engineered systems perform robustly, recovering from damage and errors. This project develops the similarities and differences in how natural biological processes and human engineering build robustly performing systems. Understanding the principles by which organisms are built to perform well provides insight into how things sometimes go wrong and how disease develops. For example, our bodies have many different mechanisms to protect against cancer. The time to failure for systems with multiple protections follows the same pattern in both biology and engineering. By understanding these general characteristics of robustness and failure, we understand better how biological organisms are built, how they fail, and how such principles apply also to human engineered systems. Prior studies by this researcher and many other scientists have developed basic understanding of the principles of robustness and failure. However, the ways in which these insights apply to many problems in biology have not been developed completely. This project will bring together the separate studies of the past into a more complete synthesis. That synthesis will help us to apply what we know to a wider range of problems. The synthesis will also help us to see more clearly what we do not understand, identifying the challenges that need further study.How do organisms actually make relatively precise traits from inherently stochastic and often unreliable biological components? How does precision arise from sloppiness? We know many pieces of the puzzle. Redundancy and extra capacity provide backup. Robustness keeps organisms on track. Repair fixes damage. Signal processing and control follow classic engineering principles. These various mechanisms for regulating traits have important consequences for evolutionary dynamics. For example, when a system corrects component-level errors, the direct selective pressure on component performance declines. Weakened selective pressure on components may cause those components to become less reliable, maintain more genetic variability, or drift neutrally in design, creating the basis for new forms of organismal complexity. To understand how these pieces of the puzzle fit together, recent advances in engineering control theory, computational machine learning, and artificial intelligence augment evolutionary theory to provide a broad conceptual foundation. This project builds on that modern conceptual foundation to synthesize diverse biological theories and to unify a wide variety of biological observations.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.