Trajectory-oriented Bayesian experiment design versus Fisher A-optimal design: an in depth comparison study.

Trajectory-oriented Bayesian experiment design versus Fisher A-optimal design: an in depth comparison study.
复制标题

DOI:
10.1093/bioinformatics/bts377
复制
发表时间:
2012-09-15
期刊:
Bioinformatics (Oxford, England)
影响因子:
--
通讯作者:
Radde N
Radde N
中科院分区:
其他
文献类型:
--
作者:
Weber P;Kramer A;Dingler C;Radde N

文献摘要

参考文献

被引文献

相似文献

动机:以参数估计或模型判别为目的的生物医学模型的实验设计策略是当前研究的热点。实验的限制,如稀疏和嘈杂的数据导致无法识别的参数和渲染相关的设计任务具有挑战性的问题。通常,数据的时间分辨率是一个限制因素,可能的实验干预的数量是有限的。为了解决这个问题,我们提出了一个贝叶斯实验设计算法,以最大限度地减少预测的不确定性为一组给定的实验,并比较它与传统的A-最优设计。结果如下:在一个深入的数值研究中,涉及一个常微分方程模型的trans-Golgi网络与12个部分不可识别的参数,我们最大限度地减少了预测的不确定性有效地为预定义的场景。该方法的预测精度是相同数量的A-最优设计实验的两倍,同时引入了一个有用的停止准则。因此,算法的主要设计步骤的模拟强度是合理的。与Fisher设计相比,除了预测轨迹的方差更小之外,我们还可以实现更小的参数后验分布熵,使得该方法在参数空间中也上级A-最优Fisher设计。提供情况:补充材料中提供了必要的软件/工具箱信息。包括示例数据的项目脚本可以从http://www.ist.uni-stuttgart.de/%7eweber/BayesFisher2012下载。联系方式:patrick.weber @ ist.uni-stuttgart.de补充信息:补充数据可在生物信息学在线获得。
Motivation: Experiment design strategies for biomedical models with the purpose of parameter estimation or model discrimination are in the focus of intense research. Experimental limitations such as sparse and noisy data result in unidentifiable parameters and render-related design tasks challenging problems. Often, the temporal resolution of data is a limiting factor and the amount of possible experimental interventions is finite. To address this issue, we propose a Bayesian experiment design algorithm to minimize the prediction uncertainty for a given set of experiments and compare it to traditional A-optimal design. Results: In an in depth numerical study involving an ordinary differential equation model of the trans-Golgi network with 12 partly non-identifiable parameters, we minimized the prediction uncertainty efficiently for predefined scenarios. The introduced method results in twice the prediction precision as the same amount of A-optimal designed experiments while introducing a useful stopping criterion. The simulation intensity of the algorithm's major design step is thereby reasonably affordable. Besides smaller variances in the predicted trajectories compared with Fisher design, we could also achieve smaller parameter posterior distribution entropies, rendering this method superior to A-optimal Fisher design also in the parameter space. Availability: Necessary software/toolbox information are available in the supplementary material. The project script including example data can be downloaded from http://www.ist.uni-stuttgart.de/%7eweber/BayesFisher2012. Contact: patrick.weber@ist.uni-stuttgart.de Supplementary Information: Supplementary data are available at Bioinformatics online.
DOI: 10.1016/j.jbiotec.2008.03.008
发表时间: 2008-06-01
影响因子: 4.1
作者:
Becker, Eric;Florin, Lore;Kaufmann, Hitto
通讯作者: Kaufmann, Hitto
DOI: 10.1049/iet-syb:20060065
发表时间: 2007-05-01
影响因子: 2.3
作者:
Casey, F. P.;Baird, D.;Sethna, J. P.
通讯作者: Sethna, J. P.
DOI: 10.1002/elps.200800720
发表时间: 2009-06-01
期刊: ELECTROPHORESIS
影响因子: 2.9
作者:
Gassmann, Max;Grenacher, Beat;Vogel, Johannes
通讯作者: Vogel, Johannes
DOI: 10.1016/j.chemolab.2011.06.008
发表时间: 2012-01-15
影响因子: 3.9
作者:
Donckels, Brecht M. R.;De Pauw, Dirk J. W.;De Baets, Bernard
通讯作者: De Baets, Bernard
DOI: 10.1186/1752-0509-4-92
发表时间: 2010-06-29
影响因子: --
作者:
Li C;Donizelli M;Rodriguez N;Dharuri H;Endler L;Chelliah V;Li L;He E;Henry A;Stefan MI;Snoep JL;Hucka M;Le Novère N;Laibe C
通讯作者: Laibe C