Recent biotechnology requires implementation of new modelling methods based on knowledge principles and learning structures, comprised in fuzzy knowledgebased systems (FKBS), neural networks (NN) and different hybrid methods. The intelligent modelling approaches solve suf®ciently a very important problem ± processing of scarce, uncertainty and incomplete numerical and linguistic information about multivariate non-linear and non-stationary systems as well as biotechnological processes. The paper deals with prediction of an enzyme oxidizing uric acid to alantoin ± the uricase, produced by Candida utilis 90-12 employing neuro-fuzzy knowledgebased approach. The implemented predictive technique exploits the fact that the fuzzy model can be seen as a network structure, similar to arti®cial NN, which on computational level assure a high model accuracy. The predictors implemented are four different by nature variables. The developed predictive model shows that best predictors of uricase production are biomass and limiting substrate concentrations.

The lactose utilization and biomass production by C. blankii 35 and C.pseudotropicalis

The lactose utilization and biomass production by C. blankii 35 and C.pseudotropicalis 11 in batch fermentation was investigated. It was found that the yield of biomass produced by oxidative C.blankii 35 was better with 31.2% than that ofC.pseudotropicalis 11. The fuzzy rulebased non-linear regression ana~ysis was implemented by using a software program NEFRIT (Nonlinear Effective Fuzzy Regression Identification Tool) for modeling of the specific growth rate f..J (h-l)ofthe studied processes. NEFRIT is based on a sub-optimal identification procedure for building a non-linear model involving fuzzy regression coefficients.