ANIMAL GENETICS • ORIGINAL PAPER Zebrafish breeding program: genetic parameters estimates for growth traits Vanessa Lewandowski 1 & Cesar Sary 2 & Jaisa Casetta 2 & André Luiz Seccatto Garcia 3 & Carlos Antonio Lopes de Oliveira 2 & Ricardo Pereira Ribeiro 2 & Lauro Daniel Vargas Mendez 2 Received: 22 August 2018 /Revised: 2 April 2019 /Accepted: 8 April 2019 # Institute of Plant Genetics, Polish Academy of Sciences, Poznan 2019 Abstract The objective of this study was to evaluate the genetic parameters of two generations of zebrafish breeding program. The base population was formed by crossing individuals of six commercial stocks of zebrafish, resulting in a nucleus with 60 families. Two generations were evaluated, with a total of 780 and 781 individuals for the first and second generation, respectively. The selection was made based on the mean genetic value of each family, followed by mass selection of the breeders. Mathematical models that considered the fixed (age, density in the larval stage, sex, and generation) and random (animal additive genetics, common to full- sibs, and residual) effects were evaluated using BLUPF90 program family. Weight and total length were used as response variables. Total length was the best selection criterion because it had a higher heritability (0.30) than weight (0.22). There was a high common to full-sib effect, especially in the first generation of animals. For second-generation data, the heritability was 0.26 for total length, as well as a lower common to full-sib effect for length. The best model obtained for this evaluation was considering all effects, being age and density as first and second polynomial, respectively. The genetic and phenotypic correla- tions for weight and length were 0.87 and 0.75, respectively. These results indicate that genetic breeding using total length as the selection criterion may produce a larger and heavier zebrafish strain. Keywords Components of variance . Danio rerio . Total length . Heritability . Weight Introduction The zebrafish ( Danio rerio) is an animal model that is widely used in various fields including neurophysiology, embryology, genet- ics, and toxicology (Grunwald and Eisen 2002; He et al. 2014; Kalueff et al. 2014; Acosta et al. 2016; Monroe et al. 2016). It is a valuable model due to its genetic similarity with humans, its pro- ductive traits, and low cost of maintenance (Vilella et al. 2008; Nasiadka and Clark 2012). Specimens destined for research can have different origins and are either produced in laboratories or acquired from fish farms or animal shops. Among the existing lines, the AB and TU lines stand out. In addition, the development of mutant and transgenic lines has increased in recent years due to the advance of genetic engineering and the transparency of the embryos and larvae (Yakulov and Walz 2015). Among other factors, an ideal animal model should provide reliable and representative results compared with other organisms (Ribas and Piferrer 2014). To achieve this goal, the standardization of individuals is required to increase the reliability of the genetic and phenotypic responses to a specific evaluation. However, the variation in results due to different lines of zebrafish has generated concern in the scientific community, where studies have demon- strated that the phenotypic expression resulting from a specific treatment may differ with animal origin (Meyer et al. 2013; Vignet et al. 2013; Monroe et al. 2016). This problem is exacer- bated when individuals are used from sources lacking information on genetics, behavioral factors, and age. The zebrafish productive cycle is not yet fully understood (Lawrence et al. 2012), which makes it challenging to main- tain fish in laboratories and reduce the efficiency of some Communicated by: Maciej Szydlowski * Vanessa Lewandowski vanessalewandowski@ufgd.edu.br 1 Faculty of Agrarian Sciences, Federal University of Grande Dourados – UFGD, Dourados, Brazil 2 Department of Animal Science, State University of Maringá – UEM, Maringá, Brazil 3 Department of Animal and Dairy Science, University of Georgia, Athens, USA Journal of Applied Genetics https://doi.org/10.1007/s13353-019-00497-9