DYNAMIC ULTRAFILTRATION OF PROTEINS - A NEURAL-NETWORK APPROACH

A neural network approach for the prediction of the rate of ultrafiltr ation of proteins has been developed. The approach has been used to pr edict the rate of ultrafiltration of bovine serum albumin as a functio n of pH and ionic strength. This is a very non-linear problem that has previously been best described through sophisticated descriptions of protein-protein interactions within the layer close to the membrane su rface. Networks with a single hidden layer have been used to predict t he dynamic rate of filtration from very few data points. Emphasis has been placed on using a small number of training data points and small networks. Variation of the number of training points and use of differ ent training point selection schemes have shown that it is the quality of training points rather than the quantity that leads to the best pr edictions. The network training process may be optimised by using phys ical insights to select appropriate input variables. Testing of the ne ural network approach showed that it could give excellent agreement wi th experimental results, with average errors less than 2.7%. (C) 1998 Elsevier Science B.V. All rights reserved.