Neural compounds

A challenging task in material science as well as in pharmaceutical research is to custom tailor a compound s properties. George S. Hammond stated that the most fundamental and lasting objective of synthesis is not production of new compounds, but production of properties (Norris Award Lecture, 1968). The molecular structure of an organic or inorganic compound determines its properties. Nevertheless, methods for the direct prediction of a compound s properties based on its molecular structure are usually not available (Figure 8-1). Therefore, the establishment of Quantitative Structure-Property Relationships (QSPRs) and Quantitative Structure-Activity Relationships (QSARs) uses an indirect approach in order to tackle this problem. In the first step, numerical descriptors encoding information about the molecular structure are calculated for a set of compounds. Secondly, statistical and artificial neural network models are used to predict the property or activity of interest based on these descriptors or a suitable subset. 

Therefore the 28 analytes and their enantiomers were encoded by the conformation-dependent chirality code (CDCC) and submitted to a Kohoiien neural network (Figure 8-1 3). They were divided into a test set of six compounds that were chosen to cover a variety of skeletons and were not used for the training. That left a training set containing the remaining 50 compounds. 

Breindl et. al. published a model based on semi-empirical quantum mechanical descriptors and back-propagation neural networks [14]. The training data set consisted of 1085 compounds, and 36 descriptors were derived from AMI and PM3 calculations describing electronic and spatial effects. The best results with a standard deviation of 0.41 were obtained with the AMl-based descriptors and a net architecture 16-25-1, corresponding to 451 adjustable parameters and a ratio of 2.17 to the number of input data. For a test data set a standard deviation of 0.53 was reported, which is quite close to the training model. 

Recently, several QSPR solubility prediction models based on a fairly large and diverse data set were generated. Huuskonen developed the models using MLRA and back-propagation neural networks (BPG) on a data set of 1297 diverse compoimds [22]. The compounds were described by 24 atom-type E-state indices and six other topological indices. For the 413 compoimds in the test set, MLRA gave = 0.88 and s = 0.71 and neural network provided... 

Figure 10.1-3. Predicted versus experimental solubility values of 552 compounds in the test set by a back-propagation neural network with 18 topological descriptors.

Figure 10.1-4. Distribution of compounds from two data sets in the same KNN (Kohonen s self-organizing neural network) map by using 18 topological descriptors as input descriptors, where 1 represents the 1588 compounds in the Merck data set (excluding those compounds that are also in the Huuskonen data set) 2 represents the 799 compounds in the Huuskonen data set (excluding those compounds that are also in the Merck data set), and 3 represents the overlapping part of the Huuskonen data set and the Merck data set.

The models are applicable to large data sets with a rapid calculation speed, a wide range of compounds can be processed. Neural networks provided better models than multilinear regression analysis. 

The objective of this study is to show how data sets of compounds for which dif-ferent biological activities have been determined can be studied. It will be shown how the use of a counter-propagation neural networb can lead to new insights [46]. The cmpha.si.s in this example is placed on the comparison of different network architectures and not on quantitative results. 

The predictive power of the CPG neural network was tested with Icavc-one-out cross-validation. The overall percentage of correct classifications was low, with only 33% correct classifications, so it is clear that there are some major problems regarding the predictive power of this model. First of all one has to remember that the data set is extremely small with only 11 5 compounds, and has a extremely high number of classes with nine different MOAs into which compounds have to be classified. The second task is to compare the cross-validated classifications of each MOA with the impression we already had from looking at the output layers. 

J. Gasteiger, Obtaining the 3D shnacture from inffared spectra of organic compounds lasing neural networks, in SoJiware-EntuHcklung in der Chemie 11, G. Pels, V Schubert (Eds.), Gesellschaft Deutscher Chemiker, Frankfurt/Main, 1997. 

The results presented here imply that a similar approadi can be used for comparing two different hbraries, for determining the degree of overlap between the compounds in these two Hbraries. Examples of the application of artificial neural networks or GA in drug design are given in [57, 58, 84, 85]. 

VR, the inputs correspond to the value of the various parameters and the network is 1 to reproduce the experimentally determined activities. Once trained, the activity of mown compound can be predicted by presenting the network with the relevant eter values. Some encouraging results have been reported using neural networks, have also been applied to a wide range of problems such as predicting the secondary ire of proteins and interpreting NMR spectra. One of their main advantages is an to incorporate non-linearity into the model. However, they do present some problems Hack et al. 1994] for example, if there are too few data values then the network may memorise the data and have no predictive capability. Moreover, it is difficult to the importance of the individual terms, and the networks can require a considerable 1 train. 

Vision is vital for human activities, and eyes are very sensitive to a number of toxic insults induced by chemical compounds. The most serious outcome is permanent eye damage which may be so severe as to cause loss of vision. The eye consists of the cornea and conjunctiva, the choroid, the iris, and the ciliary body. It also contains the retina, which is of neural origin, and the optic nerve. The retina contains photoreceptors, a highly specific light-sensitive type of neural tissue. The eye also contains the lens and a small cerebrospinal fluid system, the aqueous humor system, that is important for the maintenance of the steady state of hydration of the lens and thus the transparency of the eye. 

Viscosities of the siloxanes were predicted over a temperature range of 298-348 K. The semi-log plot of viscosity as a function of temperature was linear for the ring compounds. However, for the chain compounds, the viscosity increased rapidly with an increase in the chain length of the molecule. A simple 2-4-1 neural network architecture was used for the viscosity predictions. The molecular configuration was not considered here because of the direct positive effect of addition of both M and D groups on viscosity. The two input variables, therefore, were the siloxane type and the temperature level. Only one hidden layer with four nodes was used. The predicted variable was the viscosity of the siloxane. 

Several nonlinear QSAR methods have been proposed in recent years. Most of these methods are based on either ANN or machine learning techniques. Both back-propagation (BP-ANN) and counterpropagation (CP-ANN) neural networks [33] were used in these studies. Because optimization of many parameters is involved in these techniques, the speed of the analysis is relatively slow. More recently, Hirst reported a simple and fast nonlinear QSAR method in which the activity surface was generated from the activities of training set compounds based on some predefined mathematical functions [34]. 

The speed with which taste stimulation occurs, coupled with the fact that stimulation with toxic substances does no damage to the receptors, led Beidler to suggest that taste stimulus need not enter the interior of the taste cell in order to initiate excitation. Because a taste cell has been shown to be sensitive to a number of taste qualities, and to a large number of chemical stimuli, he and his coworkers concluded that a number of different sites of adsorption must exist on the surface of the cell. Therefore, they assumed that taste response results from adsorption of chemical stimuli to the surface of the receptor at given receptor sites. This adsorption is described by a monomolecular reaction similar to that assumed by Renqvist, Lasareff, and Hahn, but with a difference. From the fact that each type of chemical-stimulus compound has a unique level of saturation of the taste receptor, it was concluded that the magnitude of the response is dependent on the initial reaction with the receptor, and not on other, subsequent receptor-reactions that are common to all types of receptor stimulation. Therefore, it was assumed that the magnitude of neural response is directly proportional to the number of sites filled, the maximum response occurring when all of the sites are filled. Beidler derived a fundamental... 

Although some neural fibers respond to sweet-stimulus compounds placed on the tongue, others do not. This pattern of sensitivity is often a very complicated one. The fibers often respond to more than one, sometimes even to all, of the four taste modalities. Very rarely does a fiber respond specifically to only sweet or salty substances. Furthermore, other fibers may have an entirely different spectrum of sensitivities, and may respond strongly to one sweetener and very weakly to another. Pfaffmann reported how two fibers, one having one pattern of sensitivity to taste, and the other, a different pattern, can signal two different taste-qualities, even though... 

---