Gene Expression Programming and Its Application in Modeling Nitrate Reductase Enzyme Activity under Salinity Conditions

Nitrate reductase enzyme activity can be a good indicator for nitrate uptake in the presence of chloride and plant tolerance to salinity. Gene expression programming is a mathematical model between input and output variables and has the ability of choosing the effective parameters in the model. In this study, ions contents of root (chloride, sodium, nitrate and potassium) in two grape genotypes (Gharashani and GhezelUzum) as the effective parameters in the nitrate reductase activity were used in genetic programming model. The results showed that the nitrate absorption was decreased under salinity, also at higher nitrate concentration this reduction was higher in the salt sensitive genotype. It means that in grape the low affinity transport system (LATS) was more sensitive to salinity compared to the high affinity transport system (HATS). In both genotypes, the nitrate reductase activity increased with increasing the nitrate concentration in root medium. Comparing the results obtained using three function set showed that the highest correlation and lowest root mean square error were achieved for training phase by using the default function set, while for the testing phase, the main operators along with the squared root and power gave the better performance accuracy. Also, according to the suggested model, nitrate, chloride, sodium and potassium as the selected effective variables contribute with a decreasing order in the nitrate reductase activity respectively.