Department of Biology, <a href="https://ror.org/0085j8z36">Sacred Heart University</a>, Fairfield, Connecticut 06825, USA.
Department of Physics, <a href="https://ror.org/048sx0r50">University of Houston</a>, Houston, Texas 77204, USA.
Phys Rev E. 2024 Oct;110(4-1):044407. doi: 10.1103/PhysRevE.110.044407.
Advances in microarray and sequencing technologies have made possible the interrogation of biological processes at increasing levels of complexity. The underlying biomolecular networks contain large numbers of nodes, yet interactions within the networks are not known precisely. In the absence of accurate models, one may inquire if it is possible to find relationships between the states of such networks under external changes, and in particular, if such relationships can be model-independent. In this paper we introduce a class of such relationships. The results are based on the observation that changes to the equilibrium state of a network due to an alteration in an external input are "small" compared to the change in the input, a phenomenon we refer to as network modulation. It relies on the stability of the state. One consequence of network modulation is that response surfaces containing expression profiles of different mutants of an organism are low-dimensional linear subspaces. As an example, the expression profile of a double-knockout mutant generally lies close to the plane defined by the expression profiles of the wild-type and those of the two single-knockout mutants. This assertion is validated using experimental data from the sleep-deprivation network of Drosophila and the oxygen-deprivation network of Escherichia coli. The linearity of response surfaces is crucial in the design of a feedback control algorithm to move the underlying network from an initial state to a prespecified target state.
微阵列和测序技术的进步使得在越来越复杂的生物过程中进行询问成为可能。潜在的生物分子网络包含大量的节点，但是网络内部的相互作用并不精确。在没有准确模型的情况下，人们可能会询问是否有可能在外在变化下找到此类网络状态之间的关系，特别是，如果这些关系可以是独立于模型的。在本文中，我们介绍了一类这样的关系。该结果基于这样一种观察结果，即由于外部输入的改变而导致网络平衡状态的改变与输入的改变相比是“小”的，我们将这种现象称为网络调制。它依赖于状态的稳定性。网络调制的一个结果是，包含生物体不同突变体表达谱的响应表面是低维线性子空间。例如，双敲除突变体的表达谱通常位于由野生型和两个单敲除突变体的表达谱定义的平面附近。这一断言使用来自果蝇睡眠剥夺网络和大肠杆菌缺氧剥夺网络的实验数据进行了验证。响应表面的线性在设计反馈控制算法以将基础网络从初始状态移动到预定目标状态方面至关重要。
