Genetic analysis of quantitative traits in potato breeding: partitioning additive and non-additive effects for optimizing clonal and parental selection

Abstract

Solanum tuberosum L. is one of the world's main food crops, playing an essential role in global food security due to its high nutritional value, versatility and adaptability to different environments. However, crop productivity is still below its potential, mainly due to the occurrence of severe bacterial diseases such as soft rot, caused by bacteria of the genera Pectobacterium spp. and Dickeya. spp. These diseases are difficult to control, cause significant economic losses both in the field and in storage, and do not respond efficiently to chemical management methods, requiring alternative strategies, such as genetic improvement. Considering the quantitative and complex nature of traits such as tuber yield (TTY), dry matter content (SG) and resistance to soft rot, it is essential to use robust genetic approaches that allow understanding the architecture of these traits. Partitioning genetic variance into additive and non-additive components allows optimizing clonal and parental selection strategies, especially in autotetraploid species such as potato, which presents high heterozygosity. Mixed models, such as the animal model proposed by Oakey et al. (2007), offer effective tools for estimating such effects based on pedigree information. In this context, the objective of the first article was to investigate the implications of integrating additive and non-additive genetic effects in a linear mixed model, with emphasis on the partitioning of genotypic variance and the estimation of selective precision for parents and clones in early stages of a tetraploid potato breeding program, for direct use or for crossing. The second article aimed to dissect the genetic effects into additive and non-additive components for the traits TTY, SG and resistance to soft rot; to perform simultaneous selection of clones and parents based on multiple traits; and to evaluate the efficiency of including non-additive effects in mixed models applied to tetraploid potato breeding. The first article provided analyses of two populations (POP1 and POP2), while the second used only one population. Five models were tested in article one (M1, M2wE, M2pE, M3wE and M3pE) and three in article two (M1, M2 and M3). The results showed that non- additive effects were more relevant for TTY, indicating greater potential for gain through clonal selection. For SG, additive effects predominated, suggesting greater genetic stability and efficiency in both clone selection and parental selection. A negative correlation was also observed between SG and soft rot severity, demonstrating that clones with higher dry matter content had smaller lesion diameter, suggesting greater resistance to the disease without compromising productivity. This study reinforces the importance of detailed genetic modeling to improve early selection in potato breeding programs, contributing to the development