Document Type : Review Paper
Authors
1
Department of Plant Production and Genetics, Faculty of Agriculture, University of Saravan, Saravan, Iran.
2
Department of Plant Production and Genetics, Faculty of Agriculture, Urmia University, Urmia, Iran
3
Department of Plant Production and Genetics, Faculty of Agricultural Sciences, University of Guilan, Rasht, Iran.
4
Department of Plant Production and Genetics, Faculty of Agriculture, Urmia University, Urmia, Iran.
5
Urmia University
10.22124/cr.2026.33177.1894
Abstract
Abstract
Introduction: Selection, as a fundamental stage in crop breeding programs, aims to increase the frequency of desirable genotypes regarding agronomic traits, biotic and abiotic stress resistance, and superior product quality. Given the polygenic and complex nature of traits and the interplay between genetic and environmental factors, choosing an appropriate selection method plays a decisive role in the success of breeding programs. This paper provides a comprehensive and systematic review of classical and modern selection methods, their advantages and limitations, and their application in the genetic improvement of field crops.
Materials and Methods: This review study was conducted through a systematic analysis of authoritative scientific literature and published papers in the field of crop breeding. In this study, selection methods are categorized into two groups: classical and modern. Classical methods, based on phenotypic evaluation, include direct phenotypic selection (such as mass selection, pure-line selection, pedigree selection, and bulk selection), progeny testing-based selection, and indirect selection using statistical models. In contrast, modern methods, based on the integration of phenotypic and genotypic evaluations, include marker-assisted selection and genomic selection, which have significantly increased the accuracy and speed of the selection process. Furthermore, simultaneous selection for multiple traits using selection indices is comprehensively reviewed in this study.
Results: Classical methods such as mass and pedigree selection are mainly effective for traits with high heritability, although the rate of genetic gain in these methods is relatively slow. In contrast, modern technologies such as marker-assisted selection and genomic selection enable the high-accuracy identification of desirable genotypes at early growth stages. This feature makes these methods ideal for improving complex quantitative traits with low heritability. Particularly, genomic selection, utilizing advanced prediction models such as RR-BLUP and Bayesian methods, is capable of estimating the genetic merit of genotypes with remarkable accuracy and effectively reducing breeding cycle duration. On the other hand, the application of artificial neural networks in processing complex phenotypic-genotypic datasets has enhanced the prediction accuracy of yield and other related traits. Additionally, the development of multivariate indices such as MGIDI and MTSI has facilitated the simultaneous evaluation of multiple traits and the selection of superior genotypes.
Conclusion: The intelligent combination of classical and modern selection methods can significantly enhance the efficiency of crop breeding programs. Classical methods, despite their limitations, remain efficient for improving simple traits with high heritability. Conversely, modern technologies such as marker-assisted selection and genomic selection are essential for breeding quantitative traits with low heritability. The deployment of advanced computational approaches, including artificial neural networks and selection indices, not only improves the accuracy and speed of the selection process but also enables the development of high-yielding and stable cultivars. This study provides a scientific framework for the optimal choice of selection methods designed based on breeding objectives, environmental conditions, and the nature of the target traits.
Keywords
Main Subjects
)