Bayesian neural networks

An interesting approach is also the combination of a commercial log D predictor with proprietary descriptors using a Bayesian neural network approach [98]. 

Bate A, et al. 1998. A Bayesian neural network method for adverse drug reaction signal generation. Eur. J. Clin. Pharmacol. 54 315—321. 

D 3D AD ADME ADMET ANN ARD BCI BCUT BNN C4.5 CART ClogP CoMFA CV Two dimensional Three dimensional Applicability domain Absorption, distribution metabolism, and excretion Absorption, distribution metabolism, excretion, and toxicity Artificial neural network Automatic relevance determination Bernard chemical information Burden, CAS, University of Texas descriptors Bayesian neural network Decision trees using information entropy Classification and regression tree Calculated partition coefficient between octanol and water Comparative molecular field analysis Cross-validation... 

Bayesian neural networks (BNNs) are an alternative to the more traditional ANNs. The main advantage with BNNs is that they are less prone to overtraining compared to ANNs. BNNs are based on Bayesian probabilistics for the network training. Network weights are determined by Bayesian inference. BNNs have been successfully used together with automatic relevance determination (ARD) for the selection of relevant descriptors to model aqueous solubility [89]. For a good review on BNNs, see Ref. [90]. 

Artificial neural networks (feed-forward neural networks, self-organizing neural networks, counterpropagation neural networks, Bayesian neural networks)... 

This scoring function was developed by training a Bayesian neural network on a large database of known drugs and nondrug compounds. 

Burden, F.R., Ford, M.G., Whitley, D.C. and Winkler, D.A. (2000) Use of automatic relevance determination in QSAR studies using Bayesian neural networks./. Chem. Inf. Comput. Sci., 40,1423-1430. 

Techniques for the use of existing information and analyzing new data to identify causal relationships between structure, processing and properties is being developed using tools based on distributed data bases [27] and Bayesian neural networks [28], These advanced mathematical techniques permit more effective searches for critical dependencies than traditional methods. In some cases virtual mathematical experiments can be conducted that vary only a single microstructural variable to determine material responses under conditions that are impossible to duplicate in real experiments [29],... 

CNS activity is a complex process, and remains far from fuUy understood. It is difficult to interpret why certain fragments appear to influence CNS activity. Several compounds with a protonated tertiary amine can pass the blood-brain barrier, although they are not CNS-active. Tertiary amines have a pKa around 7 and can easily be protonated. A comparison of five different approaches, including Bayesian neural network and several other methods described here, revealed that substructural analysis gained so far the best prediction accuracy [38]. 

Lansner, A., Holst, A. A higher order Bayesian neural network with spiking units. Int. J. Neural Systems 7(2), 115-128 (1996)... 

A great deal of interest exists in methods for variable selection as well as for model evaluation, which are actually two sides of the same coin. Bayesian neural networks include a procedure called automatic relevance determination (ARD), allowing for the identification of important variables.A -nearest neighbor method for variable selection has been applied successfully to problems of biological activity and metabolic stability. Other... 

Use of Automatic Relevance Determination in QSAR Studies Using Bayesian Neural Networks. 

---