Modeling CRISPR gene drives for suppression of invasive rodents using a supervised machine learning framework.
Modeling CRISPR gene drives for suppression of invasive rodents using a supervised machine learning framework.
复制标题
DOI：
10.1371/journal.pcbi.1009660
复制
发表时间：
2021-12
影响因子：
4.3
通讯作者：
Messer PW
中科院分区：
文献类型：
--
作者：
Champer SE;Oakes N;Sharma R;García-Díaz P;Champer J;Messer PW
入侵啮齿动物种群对全球生物多样性构成威胁。当面对这些入侵者时，独立进化的本土物种往往毫无防御能力。 CRISPR 基因驱动系统可以通过在入侵者中传播转基因从而导致种群崩溃来解决这一问题，并且即使在传统控制方法不切实际或昂贵得令人望而却步的情况下也可以部署。在这里，我们开发了一个岛屿入侵啮齿动物种群的高保真模型，其中包括三种类型的抑制基因驱动系统。基于个体的模型在空间上是明确的，允许世代重叠和种群规模波动，并包括驱动适应度、效率、抗性等位基因形成率以及各种生态参数的变量。评估具有如此多参数的模型的计算负担给全面理解其结果空间带来了巨大障碍。因此，我们为人口模型配备了一个元模型，该元模型利用监督机器学习来高精度地近似基础模型的结果空间。这使我们能够对总体模型进行详尽的调查，包括使用数千万次评估进行基于方差的敏感性分析。我们的结果表明，在广泛的人口统计假设下，功能足够强大的基因驱动系统有可能消除岛屿上的啮齿动物种群，但前提是抵抗力可以保持在最低水平。这项研究强调了监督机器学习在识别决定复杂进化系统种群动态的关键参数和过程方面的力量。入侵的啮齿动物会破坏小岛屿上的生物多样性。这是因为在这些岛屿上进化的许多类型的植物和动物没有针对迅速蔓延的新入侵者的自然防御机制。基因驱动是一项有前途的新技术，除其他应用外，它可能有助于控制入侵啮齿动物种群。精心设计的基因驱动系统可以将工程基因传播到整个啮齿动物种群中，并最终导致种群崩溃。我们开发了一个详细的计算模型，将抑制基因驱动释放到岛屿老鼠种群中，并证明足够有效的驱动确实可以在几年内消灭这样的种群。为了帮助详细分析我们的模型（涉及各种生态和遗传参数），我们还开发了一个机器学习模型来匹配基础种群模型的结果。经过充分的训练后，该机器学习模型与底层模型非常匹配，但运行速度快了数千倍，从而可以对模型的行为进行更详细的分析。我们相信这种新技术可以应用于许多其他复杂进化系统的研究。
Invasive rodent populations pose a threat to biodiversity across the globe. When confronted with these invaders, native species that evolved independently are often defenseless. CRISPR gene drive systems could provide a solution to this problem by spreading transgenes among invaders that induce population collapse, and could be deployed even where traditional control methods are impractical or prohibitively expensive. Here, we develop a high-fidelity model of an island population of invasive rodents that includes three types of suppression gene drive systems. The individual-based model is spatially explicit, allows for overlapping generations and a fluctuating population size, and includes variables for drive fitness, efficiency, resistance allele formation rate, as well as a variety of ecological parameters. The computational burden of evaluating a model with such a high number of parameters presents a substantial barrier to a comprehensive understanding of its outcome space. We therefore accompany our population model with a meta-model that utilizes supervised machine learning to approximate the outcome space of the underlying model with a high degree of accuracy. This enables us to conduct an exhaustive inquiry of the population model, including variance-based sensitivity analyses using tens of millions of evaluations. Our results suggest that sufficiently capable gene drive systems have the potential to eliminate island populations of rodents under a wide range of demographic assumptions, though only if resistance can be kept to a minimal level. This study highlights the power of supervised machine learning to identify the key parameters and processes that determine the population dynamics of a complex evolutionary system. Invasive rodents can devastate biodiversity on small islands. This is because many types of plants and animals that evolved on such islands have no natural defense mechanisms against a rapidly spreading new invader. Gene drive is a promising new technology that, among other applications, may help control invasive rodent populations. A well-designed gene drive system could spread an engineered gene throughout a rodent population and eventually cause the population to collapse. We developed a detailed computational model of the release of a suppression gene drive into an island rat population and demonstrate that an efficient enough drive could indeed eradicate such a population within several years. To assist with a detailed analysis of our model, which involves various ecological and genetic parameters, we also developed a machine learning model to match the outcomes of the underlying population model. After sufficient training, this machine learning model is a close match to the underlying model, but runs thousands of times faster, thereby allowing for a much more detailed analysis of the behavior of the model. We believe that this new technique could be applied to the study of many other complex evolutionary systems.
登录
查看更多内容
影响因子：
1.4
作者：
Beaghton, Andrea;Beaghton, Pantelis John;Burt, Austin
通讯作者：
Burt, Austin
影响因子：
4.1
作者：
Champer J;Champer SE;Kim IK;Clark AG;Messer PW
通讯作者：
Messer PW
影响因子：
3.3
作者：
Champer, Jackson;Wen, Zhaoxin;Clark, Andrew G.
通讯作者：
Clark, Andrew G.
影响因子：
3
作者：
Beaghton, Andrea;Beaghton, Pantelis John;Burt, Austin
通讯作者：
Burt, Austin
DOI：
10.1073/pnas.1720354115
发表时间：
2018-05-22
影响因子：
11.1
作者：
Champer, Jackson;Liu, Jingxian;Messer, Philipp W.
通讯作者：
Messer, Philipp W.