03625naa a2200289 a 450000100080000000500110000800800410001902200140006002400250007410000170009924501490011626000090026550004540027452022570072865300300298565300160301565300160303165300250304765300170307265300250308965300250311465300340313970000160317370000160318970000180320577301120322310635352023-07-20       2022    bl uuuu     u00u1 u    #d  a1836-09397 a10.1071/AN214202DOI1 aPRAVIA, M.I.  aEvaluation of feed efficiency traits in different Hereford populations and their effect on variance component estimation.h[electronic resource]  c2022  aArticle history: Submitted 24 August 2021; Accepted 10 June 2022; Published online 1 August 2022. -- Handling Editor: Sue Hatcher. --  Corresponding author: Pravia, M.I.; Instituto Nacional de Investigación Agropecuaria (INIA), Estación Experimental Las Brujas, Ruta 48 Km. 10, Canelones, Uruguay; email:mpravia@inia.org.uy -- FUNDING: Financial support was provided by the National Agency for Research and Innovation (grant RTS-1-2012-1-3489). --  aABSTRACT.- Context: Residual feed intake is a relevant trait for beef cattle, given the positive impact on reducing feeding costs and greenhouse gas emissions. The lack of large databases is a restriction when estimating accurate genetic parameters for dry matter intake (DMI) and residual feed intake (RFI), and combining different data sets could be an alternative to increase the amount of data and achieve better estimations. Aim: The main objective was to compare Uruguayan data (URY; 780 bulls) and Canadian data (CAN; 1597 bulls), and to assess the adequacy of pooling both data sets (ALL) for the estimation of genetic parameters for DMI and RFI. Methods: Feed intake and growth traits phenotypes in both data sets were measured following the same protocols established by the Beef Improvement Federation. Pedigree connections among data sets existed, but were weak. Performance data were analysed for each data set, and individual partial regression coefficients for each energy sink on DMI were obtained and compared. Univariate and multivariate variance components were estimated by the restricted maximum likelihood (REML) for DMI, RFI and their energy sinks traits (average daily gain, metabolic mid weight and back fat thickness). Key results: There were some differences in phenotypic performance among data (P < 0.01); however, no differences (P > 0.1) were observed for phenotypic values of RFI between sets. Heritability estimates for DMI were 0.42 (URY), 0.41 (CAN) and 0.45 for ALL data, whereas heritability estimates for RFI were 0.34 (URY), 0.20 (CAN) and 0.25 for ALL data. The results obtained indicate selection on reducing RFI could lead to a decrease in DMI, without compromising other performance traits, as genetic correlations between RFI, growth and liveweight were low or close to 0 (-0.12-0.07). Conclusions: As genetic parameters were similar between national data sets (URY, CAN), pooling data (ALL) provided more accurate parameter estimations, as they presented smaller standard deviations, especially in multivariate analysis. Implications: Parameters estimated here may be used in international or national genetic evaluation programs. © 2022 The Author(s) (or their employer(s)). Published by CSIRO Publishing.  aAcross country evaluation  aBeef cattle  aFeed intake  aGenetic correlations  aHeritability  aMultiple trait model  aResidual feed intake  aVariance component estimation1 aNAVAJAS, E.1 aAGUILAR, I.1 aRAVAGNOLO, O.  tAnimal Production Science, 2022, Volume 62, Issue 17, pages 1652-1660. doi: https://doi.org/10.1071/AN21420