Dealing with multiple traits in maize: A new approach for selecting progenies

• Published in: Crop science 2020-11, Vol.60 (6), p.3151-3165
• Main Authors: Crispim‐Filho, Ailton José, dos Santos, Flávio Pereira, Pinto, Jefferson Fernando Naves, Melo, Patrícia Guimarães Santos, dos Reis, Edésio Fialho, Mendes‐Resende, Marcela Pedroso
• Format: Article
• Language: English
• ISSN: 0011-183X; 1435-0653
• DOI: 10.1002/csc2.20292

Many maize (Zea mays L.) breeding programs select genotypes considering just grain yield. However, this strategy may not be efficient due to undesirable associations between grain yield and other traits. The breeder's challenge is to combine good characteristics in one genotype. The aim of this study was to propose a new approach to select maize progenies considering multiple traits. One hundred forty‐one half‐sib progenies were evaluated for 14 agronomic traits in a 12 by 12 triple lattice design in two environments. Four progeny selection strategies were performed to increase yield. Strategy 1, considering just yield; Strategy 2, considering all traits in a selection index; Strategy 3, considering only traits that have effect in yield, based on path analysis; and Strategy 4, similar to Strategy 3, but disregarding any trait showing undesirable correlation with other traits. Strategies 2, 3, and 4 considered Smith and Hazel index. The progenies selected by Strategy 4 showed more balanced predicted gains considering all traits. Its predicted gain in all traits related to yield was higher than the of Strategies 2 and 3, and it led to smaller gains in height and flowering time, which breeders want to decrease its means. Strategy 1 showed the best gains in yield, as expected, but obtained undesirable phenotypes for the other traits. The use of path analysis to delineate the Smith and Hazel index is a very appealing approach for selecting maize progenies and the Strategy 4 was the most efficient at accomplishing balanced gains for multiple traits.