Evaluation of the efficiency of artificial neural networks for genetic value prediction

Comparative Studies of RSM, RSM–GA and ANFILS for Modeling and Optimization of Naphthalene Adsorption on Chitosan–CTAB–Sodium Bentonite Clay Matrix

The aim of this article was to compare the predictive abilities of the optimization techniques of response surface methodology (RSM), the hybrid of RSM–genetic algorithm (RSM–GA) and adaptive neuro-fuzzy interference logic system (ANFILS) for design responses of % removal of naphthalene and adsorption capacity of the synthesized composite nanoparticles of chitosan–cetyltrimethylammonium bromide (CTAB)–sodium bentonite clay.  The process variables considered were surfactant concentration, , activation time, ,  activation temperature, , and chitosan dosage, .  The ANFILS models showed better modeling abilities of the adsorption data on the synthesized composite adsorbent than those of ANN for reason of lower % mean absolute deviation, lower % error value, higher coefficient of determination, , amongst others and lower error functions’ values than those obtained using ANN for both responses.  When applied RSM, the hybrid of RSM–genetic algorithm (RSM–GA) and ANFILS 3–D surface pot optimization technique to determine the optimal conditions for both responses, ANFILS was adjudged the best.  The ANFILS predicted optimal conditions were = 116.00 mg/L, = 2.06 h, = 81.2oC and = 5.20 g.  Excellent agreements were achieved between the predicted responses of 99.055% removal of naphthalene and 248.6375 mg/g adsorption capacity and their corresponding experimental values of 99.020% and 248.86 mg/g with % errors of -0.0353 and 0.0894 respectively.  Hence, in this study, ANFILS has been successfully used to model and optimize the conditions for the treatment of industrial wastewater containing polycyclic aromatic compounds, especially naphthalene and is hereby recommended for such and similar studies.

Prediction of the importance of auxiliary traits using computational intelligence and machine learning: A simulation study

The present study evaluated the importance of auxiliary traits of a principal trait based on phenotypic information and previously known genetic structure using computational intelligence and machine learning to develop predictive tools for plant breeding. Data of an F₂ population represented by 500 individuals, obtained from a cross between contrasting homozygous parents, were simulated. Phenotypic traits were simulated based on previously established means and heritability estimates (30%, 50%, and 80%); traits were distributed in a genome with 10 linkage groups, considering two alleles per marker. Four different scenarios were considered. For the principal trait, heritability was 50%, and 40 control loci were distributed in five linkage groups. Another phenotypic control trait with the same complexity as the principal trait but without any genetic relationship with it and without pleiotropy or a factorial link between the control loci for both traits was simulated. These traits shared a large number of control loci with the principal trait, but could be distinguished by the differential action of the environment on them, as reflected in heritability estimates (30%, 50%, and 80%). The coefficient of determination were considered to evaluate the proposed methodologies. Multiple regression, computational intelligence, and machine learning were used to predict the importance of the tested traits. Computational intelligence and machine learning were superior in extracting nonlinear information from model inputs and quantifying the relative contributions of phenotypic traits. The R² values ranged from 44.0% - 83.0% and 79.0% - 94.0%, for computational intelligence and machine learning, respectively. In conclusion, the relative contributions of auxiliary traits in different scenarios in plant breeding programs can be efficiently predicted using computational intelligence and machine learning.