Path analysis for quality traits and minerals in common bean based on data from multi-environment experiments

Abstract

The number of experiments to be used in path analysis to enhance the efficacy of indirect selection for fast-cooking common bean (Phaseolus vulgaris L.) lines has not yet been defined. This study sought to break down the correlation coefficient into direct and indirect effects for quality traits and mineral concentration based on data acquired from multi-environment experiments. Additionally, the study proposed to establish the minimum number of experiments required for path analysis aiming at indirect selection for fast cooking in common bean. Four experiments were conducted in which various quality traits and the concentration of seven minerals were analyzed across 25 common bean cultivars. Variance and path analyses were applied to data from individual experiments and combinations of two, three, and four experiments. Significant cultivar × experiment interaction effects were found for most evaluated traits. The traits exerting the greatest direct effects on cooking time varied across the four experiments. Data from individual experiments were highly variable, resulting in low ability to identify promising traits for indirect selection. However, data from two, three, and four experiments had lower variability and therefore provided a greater ability to identify traits with the greatest direct and indirect effects on cooking time. Mass of 100 grains and calcium concentration emerged as promising traits for indirect selection to achieve fast cooking in common bean. Using data from two experiments allows for an effective interpretation of path analysis results for quality traits and mineral concentration in common bean.