进化计算中的启发式算法和生物大分子的模块化组织:在定向进化中的应用
Heuristic algorithms in Evolutionary Computations and modular organization of biological macromolecules: applications to directed evolution
摘要:
不久前,进化生物学的思想启发了计算机科学家们发展了一个蓬勃发展的领域,即进化计算(EC),特别是遗传算法(GA)。同时,将生物分子的定向进化(体外进化)合理地解释为遗传算法在生化实验中的实现。遗传算法的理论基础之一,证明进化搜索的有效性,是积木(BB)的概念。在进化生物学和合成生物学中,将这些BB与大分子的结构域和基序相匹配是合理的。计算机科学家已经证明并仔细研究了保留已经发现的BB对于进化搜索的有效性的重要性。为此,已经开发了几十种算法,包括启发式交叉算法。另一方面,定义和保存结构域的实验程序仍然是体外进化技术中一个欠发达的领域。在本文中,我们演示了几个简单的算法保持BB可以提高体外进化的效率,在数值实验中几乎一个数量级。作为测试问题,我们提出并使用这些众所周知的合成生物学的进化搜索强细菌启动子(有几个图案)和搜索多域RNA设备的问题,相比经典的GA测试(皇家路功能)。这些简单算法的初步测试的成功使我们有充分的理由期待更先进和现代EC程序的实施和应用将使效率得到更大的提高。这种搜索效率的提高将显著降低体外进化实验的成本,这将完全覆盖基于这些算法开发新实验程序的成本。
英文摘要:
A while ago, the ideas of evolutionary biology inspired computer scientists to develop a thriving nowadays field of evolutionary computation (EC), in general, and genetic algorithms (GA), in particular. At the same time, the directed evolution of biological molecules (in vitro evolution) is reasonably interpreted as an implementation of GA in biochemical experiments. One of the theoretical foundations of GA, justifying the effectiveness of evolutionary search, is the concept of building blocks (BB). In EC, it is reasonable to match these BBs to domains and motifs of macromolecules in evolutionary and synthetic biology. Computer scientists have shown and carefully studied the importance of preserving already found BBs for the effectiveness of evolutionary search. For this purpose, dozens of algorithms have been developed, including heuristic crossover algorithms. On the other hand, the experimental procedures defining and preserving domains remain a poorly developed area in the techniques of evolution in vitro. In this paper, we demonstrate how several simple algorithms preserving the BBs can increase the efficiency of in vitro evolution in numerical experiments by almost an order of magnitude. As test problems, we propose and use such well-known problems of synthetic biology as the evolutionary search for strong bacterial promoters (with several motifs) and search for multi-domain RNA devices, as compared to the classic GA tests (Royal Road functions). The success of these primary tests with simple algorithms gives us every reason to expect that the implementation and application of more advanced and modern EC procedures will give an even greater increase in efficiency. Such an increase in search efficiency will significantly reduce the cost of in vitro evolution experiments, which will fully cover the costs of developing new experimental procedures based on these algorithms.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。