A hybrid model based on general regression neural network and fruit fly optimization algorithm for forecasting and optimizing paclitaxel biosynthesis in Corylus avellana cell culture.

摘要: Paclitaxel is a well-known chemotherapeutic agent widely applied as therapy for various types of cancers. In vitro culture Corylus avellana has been named promising and low-cost strategy paclitaxel production. Fungal elicitors have reported an impressive improving biosynthesis in cell suspension (CSC) C. avellana. The objectives this research were to forecast optimize growth based on four input variables including extract (CE) filtrate (CF) concentration levels, elicitor adding day CSC harvesting time culture, case study, using general regression neural network-fruit fly optimization algorithm (GRNN-FOA) via data mining approach the first time. GRNN-FOA models (0.88–0.97) showed superior prediction performances compared (0.57–0.86). Comparative analysis multilayer perceptron-genetic (MLP-GA) very slight difference between two for dry weight (DW), intracellular extracellular testing subset, unseen data. However, MLP-GA was slightly more accurate total portion subset. observed maximum optimized by FOA GA. developed that optimal CE [4.29% (v/v)] CF [5.38% (17) (88 h 19 min) can lead highest (372.89 µg l−1). Great accordance predicted values DW, intracellular, yield paclitaxel, also support excellent performance models. Overall, new mathematical tool may pave way forecasting optimizing secondary metabolite production plant culture.