Modeling neural networks with ordinary differential equations systems is a sensible approach, but also very difficult. This paper describes a new algorithm based on linear genetic programming which call be used to rev...
详细信息
Modeling neural networks with ordinary differential equations systems is a sensible approach, but also very difficult. This paper describes a new algorithm based on linear genetic programming which call be used to reverse engineer neural networks. The RODES algorithim automatically discovers the structure of the network, including neural connections, their signs and strengths, estimates its parameters, and can even be used to identify the biophysical mechanisms involved. The algorithm is tested on simulated time series data, generated using a realistic model of the subthalamopallidal network of basal ganglia. The resulting ODE system is highly accurate, and results are obtained in a matter of minutes. This is because the problem of reverseengineering a system of coupled differential equations is reduced to one of reverseengineering individual algebraic equations. The algorithm allows the incorporation of common domain knowledge to restrict the solution space. To our knowledge, this is the first time a realistic reverse engineering algorithm based on linear genetic programming has been applied to neural networks. (C) 2008 Elsevier Ltd. All rights reserved.
Advances in transcriptional analysis offer great opportunities to delineate the structure and hierarchy of regulatory networks in biochemical systems. We present an approach based on Boolean analysis to reconstruct a ...
详细信息
Advances in transcriptional analysis offer great opportunities to delineate the structure and hierarchy of regulatory networks in biochemical systems. We present an approach based on Boolean analysis to reconstruct a set of parsimonious networks from gene disruption and over expression data. Our algorithms, Causal Predictor (CP) and Relaxed Causal Predictor (RCP) distinguish the direct and indirect causality relations from the non-causal interactions, thus significantly reducing the number of miss-predicted edges. The algorithms also yield substantially fewer plausible networks. This greatly reduces the number of experiments required to deduce a unique network from the plausible network structures. Computational simulations are presented to substantiate these results. The algorithms are also applied to reconstruct the entire network of galactose utilization pathway in Saccharomyces cerevisiae. These algorithms will greatly facilitate the elucidation of regulatory networks using large scale gene expression profile data. (C) 2004 Elsevier Inc. All rights reserved.
暂无评论