Resumen

MORA, Freddy; OSMERIO, Pupim-Junior  y  SCAPIM, Carlos Alberto. Prediction of cultivar effects on cotton yield in the presence of genotype-environment interaction. Cienc. Inv. Agr. [online]. 2007, vol.34, n.1, pp.13-21. ISSN 0718-1620.  http://dx.doi.org/10.4067/S0718-16202007000100002.

We aimed to obtain, by the Best Linear Unbiased Predictor (BLUP), a prediction of the cultivar effect of 14 cotton cultivars (Gossypium hirsutum) established at different sites in Brazil and Paraguay during the 2003-2004 growing season. The BLUP of cotton yield was compared with classical stability analysis according to the Plaisted and Peterson, Wricke, Annicchiarico and Lin and Binns models. The Independence Chain (IC) algorithm, within a Bayesian framework, was also utilized for comparison. A likelihood ratio test provided evidence of significant differences for cultivar and genotype-environment interaction effects. These results were confirmed by Akaike and Bayesian information criteria. Most Spearman and Pearson correlation values were not significant site-to-site (p>0.05), varying from -0.037±0.307 to 0.565±0.290 and from -0.228±0.336 to 0.604±0.257, respectively. The Plaisted and Peterson and Wricke methods showed no significant correlation with BLUP, IC, Annicchiarico and Lin and Binns. On the other hand, Spearman rank coefficients were high and significant (p<0.01) between BLUP and Annicchiarico (0.859±0.151) and BLUP and Lin and Binns (-0.899±0.099). BLUP and IC, based on mean and median posterior estimates, had identical ranks. CD-406 and CD99-2221 were the cultivars that evidenced superior stability, as confirmed equally by BLUP, IC, Annicchiarico and Lin and Binns, indicating a significant concordance between approaches in relation to genotypic stability. A mixed linear models methodology allows us to get important genotypic information of cotton cultivars with both high productivity and stability in the environments where they will be cultivated by farmers

Palabras clave : BLUP; likelihood; genotype-environment interaction; improvement; cultivar stability.

        · resumen en Español     · texto en Español     · Español ( pdf )