A predictive model of gene expression reveals the role of network motifs in the mating response of yeast.

Cells use signaling pathways to receive and process information about their environment. These systems are nonlinear, relying on feedback and feedforward regulation to respond appropriately to changing environmental conditions. Mathematical models developed to describe signaling pathways often fail to show predictive power, because the models are not trained on data that probe the diverse time scales on which feedforward and feedback regulation operate. We addressed this limitation using microfluidics to expose cells to a broad range of dynamic environmental conditions. In particular, we focus on the well-characterized mating response pathway of S. cerevisiae (yeast). This pathway is activated by mating pheromone and initiates the transcriptional changes required for mating. Although much is known about the molecular components of the mating response pathway, less is known about how these components function as a dynamical system. Our experimental data revealed that pheromone-induced transcription persists following removal of pheromone and that long-term adaptation of the transcriptional response occurs when pheromone exposure is sustained. We developed a model of the regulatory network that captured both persistence and long-term adaptation of the mating response. We fit this model to experimental data using an evolutionary algorithm and used the parameterized model to predict scenarios for which it was not trained, including different temporal stimulus profiles and genetic perturbations to pathway components. Our model allowed us to establish the role of four regulatory motifs in coordinating pathway response to persistent and dynamic stimulation.

细胞利用信号通路来接收和处理有关其环境的信息。这些系统是非线性的，依靠反馈和前馈调节来对不断变化的环境条件做出适当反应。为描述信号通路而建立的数学模型往往缺乏预测能力，因为这些模型没有基于探测前馈和反馈调节所作用的不同时间尺度的数据进行训练。我们利用微流体技术让细胞暴露于广泛的动态环境条件中来解决这一局限。特别是，我们聚焦于酿酒酵母（酵母）特征明确的交配反应通路。该通路由交配信息素激活，并启动交配所需的转录变化。尽管人们对交配反应通路的分子成分了解很多，但对这些成分如何作为一个动态系统发挥作用却知之甚少。我们的实验数据显示，在信息素去除后，信息素诱导的转录仍会持续，并且当信息素持续存在时，转录反应会发生长期适应。我们开发了一个调控网络模型，该模型捕捉到了交配反应的持续性和长期适应性。我们使用一种进化算法将该模型与实验数据进行拟合，并使用参数化模型来预测它未经过训练的情况，包括不同的时间刺激模式以及对通路成分的基因扰动。我们的模型使我们能够确定四种调控基序在协调通路对持续和动态刺激的反应中的作用。