Parameter estimation and model selection in computational biology.
Parameter estimation and model selection in computational biology.
复制标题
DOI：
10.1371/journal.pcbi.1000696
复制
发表时间：
2010-03-05
影响因子：
4.3
通讯作者：
Khammash M
中科院分区：
文献类型：
--
作者：
Lillacci G;Khammash M
生物系统计算建模的一个核心挑战是模型参数的确定。通常情况下，只有一小部分参数（如动力学速率常数）是实验测量的，而其余的往往是拟合。拟合过程通常基于可观测量的实验时间过程测量，其用于分配参数值，以最小化这些测量与相应的模型预测之间的误差的一些测量。这些测量可以来自免疫印迹分析、荧光标记等，往往是非常嘈杂的，并采取了在有限数量的时间点。在这项工作中，我们提出了一种新的方法来选择生物模型的参数的问题。我们展示了如何使用一个动态递归估计，称为扩展卡尔曼滤波器，到达模型参数的估计。建议的方法如下。首先，我们使用一个变化的卡尔曼滤波器，特别适合于生物应用，以获得未知参数的第一个猜测。其次，我们采用后验可识别性检验来检验估计的可靠性。最后，我们解决了一个优化问题，以完善第一个猜测的情况下，它应该是不够准确。最终估计值保证与测量值在统计上一致。此外，我们展示了如何使用相同的工具来区分相同的生物过程的替代模型。我们证明了这些想法，通过应用我们的方法的两个例子，即在E。大肠杆菌，以及合成基因调控系统的模型。所提出的方法是相当普遍的，可以应用到广泛的一类生物系统中的噪声测量用于参数估计或模型选择。参数估计是系统生物学中的一个关键问题，因为它代表了从生物系统的计算模型获得预测的关键步骤。这个问题通常通过将模型模拟与观察到的实验数据“拟合”来解决。这种方法没有充分考虑测量噪声。我们介绍了一种新的方法，建立在卡尔曼滤波，统计测试和优化技术的组合。该滤波器在控制和估计理论中是众所周知的，并且已经在广泛的领域中找到了应用，例如惯性制导系统、天气预报和经济学。我们展示了如何测量噪声的统计数据可以被最佳地利用，并直接纳入估计算法的设计，以实现更准确的结果，并验证/无效的计算估计。我们还表明，我们的估计的一个显着的优势是，它提供了一个强大的工具，模型选择，允许拒绝或接受竞争模型的基础上可用的噪声测量。这些结果在计算生物学中具有直接的实际应用，虽然我们用两个具体的例子证明了它们的用途，但它们实际上可以用于研究广泛的一类生物系统。
A central challenge in computational modeling of biological systems is the determination of the model parameters. Typically, only a fraction of the parameters (such as kinetic rate constants) are experimentally measured, while the rest are often fitted. The fitting process is usually based on experimental time course measurements of observables, which are used to assign parameter values that minimize some measure of the error between these measurements and the corresponding model prediction. The measurements, which can come from immunoblotting assays, fluorescent markers, etc., tend to be very noisy and taken at a limited number of time points. In this work we present a new approach to the problem of parameter selection of biological models. We show how one can use a dynamic recursive estimator, known as extended Kalman filter, to arrive at estimates of the model parameters. The proposed method follows. First, we use a variation of the Kalman filter that is particularly well suited to biological applications to obtain a first guess for the unknown parameters. Secondly, we employ an a posteriori identifiability test to check the reliability of the estimates. Finally, we solve an optimization problem to refine the first guess in case it should not be accurate enough. The final estimates are guaranteed to be statistically consistent with the measurements. Furthermore, we show how the same tools can be used to discriminate among alternate models of the same biological process. We demonstrate these ideas by applying our methods to two examples, namely a model of the heat shock response in E. coli, and a model of a synthetic gene regulation system. The methods presented are quite general and may be applied to a wide class of biological systems where noisy measurements are used for parameter estimation or model selection. Parameter estimation is a key issue in systems biology, as it represents the crucial step to obtaining predictions from computational models of biological systems. This issue is usually addressed by “fitting” the model simulations to the observed experimental data. Such approach does not take the measurement noise into full consideration. We introduce a new method built on the combination of Kalman filtering, statistical tests, and optimization techniques. The filter is well-known in control and estimation theory and has found application in a wide range of fields, such as inertial guidance systems, weather forecasting, and economics. We show how the statistics of the measurement noise can be optimally exploited and directly incorporated into the design of the estimation algorithm in order to achieve more accurate results, and to validate/invalidate the computed estimates. We also show that a significant advantage of our estimator is that it offers a powerful tool for model selection, allowing rejection or acceptance of competing models based on the available noisy measurements. These results are of immediate practical application in computational biology, and while we demonstrate their use for two specific examples, they can in fact be used to study a wide class of biological systems.
登录
查看更多内容
DOI：
10.1073/pnas.262658899
发表时间：
2002-12-24
影响因子：
11.1
作者：
Battogtokh, D;Asch, DK;Schüttler, HB
通讯作者：
Schüttler, HB
影响因子：
--
作者：
Ashyraliyev, Maksat;Jaeger, Johannes;Blom, Joke G.
通讯作者：
Blom, Joke G.
影响因子：
3.5
作者：
Bortz, D. M.;Nelson, P. W.
通讯作者：
Nelson, P. W.
影响因子：
56.9
作者：
KIRKPATRICK, S;GELATT, CD;VECCHI, MP
通讯作者：
VECCHI, MP
网络图案：结构无法确定功能。
影响因子：
4.4
作者：
Ingram, Piers J;Stumpf, Michael P H;Stark, Jaroslav
通讯作者：
Stark, Jaroslav