Linear and Nonlinear QSAR models on Effective Biological and chemical Parameters of water in Abundance Oligochaetes (P. hammoniensis, O. serpentin and B. sowerbyi) in the Anzali International Wetland

Subject Areas : International Journal of Bio-Inorganic Hybrid Nanomaterials

Received: 2021-09-28 Accepted : 2021-11-30 Published : 2022-01-01

Keywords: QSAR, MLR, Genetic algoritm, Oligochaetes,

Abstract :

Quantitative structure-activity relationship (QSAR) of the Biological and chemical properties of water in the Anzali International Wetland was estimated by means of multiple linear regression (MLR), artificial neural network (ANN), simulated annealing (SA) and genetic algorithm (GA) techniques. The obtained results from MLR-MLR, MLR-SA, SA-ANN and GA-ANN approaches were compared and GA-ANN combination showed the best performance according to its correlation coefficient (R2) and mean sum square errors (RMSE). A high predictive ability was observed for the GA-ANN model; with the root mean sum square errors (RMSE) of 0.0054 in P. hammoniensis and 0.0020 in B. sowerbyi and 0.001 in O. serpentina, respectively. From GA-ANN simulations, it was found that the total soluble solids (TSS), dissolved oxygen (DO), total nitrogen (TN), NH3 concentration, Sodum chloride (Sali), Nitrat (NO3), Total dissolved solids (TDS), total organic material (TOM), are the most important Biological and chemical parameters of water that affect abundance of the P. hammoniensis, O. serpentin, B. sowerbyi.