Application of ANN for prediction of cellulase and xylanase production by Trichoderma reesei under SSF condition

Abstract

Solid state fermentation is a bioconversion process that involves treatment of biodegradable solid substrate with microorganisms. This technique is widely applied for biotransformation of agricultural waste into industrial enzymes, organic solvents and other biochemicals. It is characterized by the presence of moisture, sufficient to solubilize the nutrients, but avoids leaching and operates at water activity (aw) of 0.85. On account of difference in water binding capacity of different substrates, optimum moisture level needs to be established for various combination of substrates, which involves extensive laborious experimental work. Present investigations were carried out to study the application of Artificial Neural Network as a tool for predicting cellulase and xylanase production by Trichoderma reesei as a function of bagasse content and moisture level in comparison to wheat bran medium. A correlation coefficient > 0.8 and root mean square error < 0.4 indicates ANN as a good prediction tool for such complex biological process.