Correlated response in fillet weight and yield to selection for increased harvest weight in genetically improved farmed tilapia (GIFT strain), Oreochromis niloticus Nguyen Hong Nguyen a, ⁎, Raul W. Ponzoni a , Khairul R. Abu-Bakar a , Azhar Hamzah b , Hooi Ling Khaw a , Hong Yip Yee a a WorldFish Center, Jalan Batu Maung, Batu Maung, 11960 Bayan Lepas, Penang, Malaysia b National Prawn Fry Production and Research Center, Kg. P. Sayak, 08500 Kedah, Malaysia abstract article info Article history: Received 24 September 2009 Received in revised form 9 April 2010 Accepted 11 April 2010 Keywords: Nile tilapia Oreochromis niloticus Genetic improvement Selection response Carcass traits A data set consisting of 5532 slaughter records collected over three generations from a selection program for increased harvest body weight in the GIFT strain of Nile tilapia (Oreochromis niloticus) was used to estimate genetic parameters and correlated responses in body (live weight, standard length, body width and body depth) and carcass (fillet weight and fillet yield) traits. A multi-trait model using restricted maximum likelihood method was applied to a full pedigree comprising a total of 18,970 animals. The estimates of heritability for body and carcass traits were of moderate magnitude (0.20 to 0.33). The proportions of variance explained by the maternal and common environmental effects were small, ranging from 4 to 8% for body traits and fillet weight, but were negligible for fillet yield. Genetic correlations among body traits were high (0.78 to 0.95), except between standard length and body width (0.56). High genetic correlations (0.78 to 0.96) were also obtained between body traits and fillet weight, whereas those between body traits and fillet yield were generally low (0.35 to 0.44). Genetic changes were measured as the difference in least squares means between the selection and control lines. The correlated increase in fillet weight was 23% in the latest generation studied or 0.97 phenotypic standard deviation units (σ P ). The correlated response in fillet yield was negligible. In conclusion, selection for high growth significantly increased fillet weight. Strategies for the improvement of fillet yield in the GIFT strain are discussed. © 2010 Elsevier B.V. All rights reserved. 1. Introduction Fillet weight and yield are traits of particular economic importance since marketing systems in major tilapia-producing countries are shifting from payment based on whole-fish live weight to fillet weight. However, genetic improvement for fillet traits has been hampered by a lack of efficient methods of measurements which often require slaughter of live animals after selection decisions are made. Several studies have attempted to develop prediction equations based on measurements of body traits (Rutten et al., 2005; Pires et al., 2006; Sang et al., 2009; Nguyen et al., in preparation). Irrespective of methods and number of traits used, the general conclusion from these studies is that fillet weight can be accurately predicted from body traits. By contrast, prediction equations for fillet yield based on body measurements are not accurate enough. Hence, efficiency of selection based on the prediction equations can be high for fillet weight, but not for fillet yield. Alternatively, improvement of fillet traits can be indirectly made through selection for body traits. Our hypothesis is that selection for high growth rate (Ponzoni et al., 2005; Eknath et al., 2007; Nguyen et al., 2007) can also result in favorable correlated responses in fillet weight, but insignificant change in fillet yield. The genetic correlations between body weight and fillet weight are known to be highly positive, whereas the genetic correlations between body weight and fillet yield are weakly positive (Kause et al., 2007) or negative (Navarro et al., 2009). Therefore, the main aim of the present paper was to estimate realized correlated responses in fillet traits from the long term selection program for high growth rate in the GIFT strain. We also report genetic parameters for fillet traits in relation to body measurements. 2. Materials and methods 2.1. The genetic selection lines The origin of the GIFT strain of Nile tilapia (Oreochromis niloticus) was described in detail by Ponzoni et al. (2005) and Nguyen et al. (2007). The strain was established in Malaysia from 63 full sib families, each represented by 35 fish of the sixth generation of selection for increased harvest weight. In 2001 and 2002, the founder stock was transferred in batches from the GIFT Foundation International Inc., Philippines, to the Aquaculture Extension Center, Department of Fisheries, Jitra, Kedah State, Malaysia, where they were reared to an average body weight of about 250 g before mating was initiated. In 2002, the progeny of the first spawning season was produced in Malaysia, thus creating what we call the base population. Two lines were Aquaculture 305 (2010) 1–5 ⁎ Corresponding author. Tel.: +60 4 6202 189; fax: +60 4 626 5530. E-mail address: N.Nguyen@cgiar.org (N.H. Nguyen). 0044-8486/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.aquaculture.2010.04.007 Contents lists available at ScienceDirect Aquaculture journal homepage: www.elsevier.com/locate/aqua-online