RESUMO
BACKGROUND: Commonly, several traits are assessed in agronomic experiments to better understand the factors under study. However, it is also common to see that even when several traits are available, researchers opt to follow the easiest way by applying univariate analyses and post-hoc tests for mean comparison for each trait, which arouses the hypothesis that the benefits of a multi-trait framework analysis may have not been fully exploited in this area. RESULTS: In this paper, we extended the theoretical foundations of the multi-trait genotype-ideotype distance index (MGIDI) to analyze multivariate data either in simple experiments (e.g., one-way layout with few treatments and traits) or complex experiments (e.g., with a factorial treatment structure). We proposed an optional weighting process that makes the ranking of treatments that stands out in traits with higher weights more likely. Its application is illustrated using (1) simulated data and (2) real data from a strawberry experiment that aims to select better factor combinations (namely, cultivar, transplant origin, and substrate mixture) based on the desired performance of 22 phenological, productive, physiological, and qualitative traits. Our results show that most of the strawberry traits are influenced by the cultivar, transplant origin, cultivation substrates, as well as by the interaction between cultivar and transplant origin. The MGIDI ranked the Albion cultivar originated from Imported transplants and the Camarosa cultivar originated from National transplants as the better factor combinations. The substrates with burned rice husk as the main component (70%) showed satisfactory physical proprieties, providing higher water use efficiency. The strengths and weakness view provided by the MGIDI revealed that looking for an ideal treatment should direct the efforts on increasing fruit production of Albion transplants from Imported origin. On the other hand, this treatment has strengths related to productive precocity, total soluble solids, and flesh firmness. CONCLUSIONS: Overall, this study opens the door to the use of MGIDI beyond the plant breeding context, providing a unique, practical, robust, and easy-to-handle multi-trait-based framework to analyze multivariate data. There is an exciting possibility for this to open up new avenues of research, mainly because using the MGIDI in future studies will dramatically reduce the number of tables/figures needed, serving as a powerful tool to guide researchers toward better treatment recommendations.
RESUMO
The adoption of a multivariate perspective of selection implies the existence of multivariate adaptive peaks and pervasive correlational selection that promotes co-adaptation between traits. However, to test for the ubiquity of correlational selection in nature, we must first have a sense of how well can we estimate multivariate nonlinear selection (i.e., the γ-matrix) in the face of sampling error. To explore the sampling properties of estimated γ-matrices, we simulated inidividual traits and fitness under a wide range of sample sizes, using different strengths of correlational selection and of stabilizing selection, combined with different number of traits under selection, different amounts of residual variance in fitness, and distinct patterns of selection. We then ran nonlinear regressions with these simulated datasets to simulate γ-matrices after adding random error to individual fitness. To test how well could we detect the imposed pattern of correlational selection at different sample sizes, we measured the similarity between simulated and imposed γ-matrices. We show that detection of the pattern of correlational selection is highly dependent on the total strength of selection on traits and on the amount of residual variance in fitness. Minimum sample size needs to be at least 500 to precisely estimate the pattern of correlational selection. Furthermore, a pattern of selection in which different sets of traits contribute to different functions is the easiest to diagnose, even when using a large number of traits (10 traits), but with sample sizes in the order of 1000 individuals. Consequently, we recommend working with sets of traits from distinct functional complexes and fitness proxies less prone to effects of environmental and demographic stochasticity to test for correlational selection with lower sample sizes.