Grain yield (Y) of crops is determined by several Y components which reflect positive or negative effects. Conventionally, the ordinary Y components, like 1000 grains weight, have been screened for highest direct effect on Y using path analysis to be used in next steps of breeding programs. Increasing one component tends to be somewhat counterbalanced by concomitant reduction in other component (s) due to competition for assimilates. Therefore, it has been suggested that components should be manipulated in conjugation with other traits to break the competition-resulted barrier. The objective of this study was to optimize the effective components in conjugation with some participant traits for increased barley Y using artificial neural network (ANN) and genetic algorithm (GA) as an alternative procedure. Two field experiments were separately carried out in Agriculture Research Station located in Gonbade Kavous (37o16' N, 55o12' E and 37 asl), Iran. Ten genotypes were grown in each experiment, and Y and some traits/components were measured. Among components/traits, those with significant direct effect and/or correlation with Y were selected as effective to be used for ANN and GA analysis. The results indicated that remobilization of stored pre-anthesis assimilates to grain (R1), crop height (R2), 1000 grains weight (R3), grain number per ear (R4), vegetative growth duration (R5), grain filling duration (R6), grain filling rate (R7), and tiller number (R8) were effective. R2 for training and test phases was 0.99 and 0.94, respectively, reveals capability of ANN for predicting Y. The optimum values obtained by GA were 14.2%, 104.34 cm, 36.9 g, 41.9, 100 d, 48 d, 1.22 mg seed-1 day-1, and 3.38 plant-1, for R1 to R8, respectively. Optimization increased potential Y to 5791 kg ha-1 which was higher than those were observed for genotypes (3527 to 5163 kg ha-1).

Keywords