Modeling the red pigment production by Monascus purpureus MTCC 369 by Artificial Neural Network using rice water based medium

Abstract

Abstract The study investigates the application of Artificial Neural Network (ANN) in modeling a Liquid State Fermentation (LSF) for red pigment production by Monascus purpureus MTCC 369. A neural network model was characterized by the three fermentation parameters as neurons in the input layer and pigment yield as one neuron in the output layer. The input neurons included incubation period of 6–18 days, pH of rice water as substrate (3.0–5.0) and concentration of ammonium nitrate as nitrogen source (0.0–2.0%). The model was trained and validated to predict the red pigment yield (abs 500 nm /mg dry fungal biomass). The results showed a good fit between predicted and experimental values for the model. The maximum red pigment yield (20.44 U abs 500 nm /mg dfb) was obtained with substrate pH of 4.0 without any ammonium nitrate as N-source after 12 days of incubation. The developed ANN model can be used to predict the effects of fermentation parameters on red pigment production with a high correlation.