The feed intake (FI) and feed efficiency are economically important traits in ducks. To obtain insight into this economically important trait, we designed an experiment based on the residual feed intake (RFI) and feed conversion ratio (FCR) of a random population Pekin duck.
Two thousand and twenty pedigreed random population Pekin ducks were established from 90 males mated to 450 females in two hatches. Traits analyzed in the study were body weight at the 42th day (BW42), 15 to 42 days average daily gain (ADG), 15 to 42 days FI, 15 to 42 days FCR, and 15 to 42 days RFI to assess their genetic inter-relationships. The genetic parameters for feed efficiency traits were estimated using restricted maximum likelihood (REML) methodology applied to a sire-dam model for all traits using the ASREML software.
Estimates heritability of BW42, ADG, FI, FCR, and RFI were 0.39, 0.38, 0.33, 0.38, and 0.41, respectively. The genetic correlation was high between RFI and FI (0.77) and moderate between RFI and FCR (0.54). The genetic correlation was high and moderate between FCR and ADG (−0.80), and between FCR and BW42 (−0.64), and between FCR and FI (0.49), respectively.
Feed costs contribute a major portion to the total cost of livestock production. Improvement in feed efficiency would reduce the amount of feed required for growth, the production cost and the amount of nitrogenous waste. Feed efficiency is usually expressed as the amount of feed intake (FI) per body weight gain (BWG), which is also known as feed conversion ratio (FCR). Feed efficiency has traditionally been improved through selection for decreased FI and increased growth. Feed efficiency is difficult to measure with high labor and equipment costs because the ducks are raised in groups on the floor. But an improvement in feed efficiency is important because it may reduce the production costs in ducks, lower their feed requirements for growth was well decreasing nitrogenous waste production .
Koch et al.  was the first to introduce the concept of adjusting FI for performance traits with the intention of evaluating different measures of efficiency on indirect feed utilization and efficiency. Residual feed intake (RFI) is defined as the difference between observed feed intake and predicted feed intake based on estimates of maintenance and production requirements. Animals that require less feed than expected for maintenance and production purposes have a negative RFI value and are desirable in animal production as they reduce feed costs and improve profitability of production. This trait was first investigated in poultry for the laying hen , and thereafter more commonly in mammals including; cattle , pigs [5,6], and rabbits .
Genetic variability in RFI has been investigated in chickens , beef cattle [9–13], and pigs [5,6,14,15]. To date, there are no studies on RFI heritabilities and genetic correlation between RFI and FCR in Pekin ducks.
The objectives of this study were to estimate genetic parameters for RFI, FCR, average daily gain (ADG), BW42 and FI in a Pekin duck population, and to ascertain the genetic relationships among the traits studied in Pekin ducks.
MATERIALS AND METHODS
Experimental population and animal husbandry
The care and use of all ducks in this experiment were approved by the Animal Care and Use Committee of the Institute of Animal Sciences, Chinese Academy of Agricultural Science (CAAS). The study included a total of 2,020 Pekin ducks from a random mating population which was raised in the Institute of Animal Science, CAAS. These ducks were pedigreed by mating 90 males with 450 female in two hatches between April 2014 and May 2014. In the first two weeks, ducks were placed in pens (0.08 m2/bird). All ducks were fasted for 12 hours at 14th day of the experiment and transferred to their individual metabolic cages with a dimension of 50 cm (width)×50 cm (length)×30 cm (height) at the same day where they remained caged until 42 days of age. During the first three days after hatching, lighting was continuous 24 L:0 D (that is to say 24 h of light and 0 h of dark); then the lighting regimen of 20 L:4 D was assigned to the ducks from 4 days to 42 days. Ingredient composition and chemical composition of feedstuffs used in this study is presented in Table 1.
Traits under the study and statistical analysis
Body weight was recorded at the beginning (15 days old) and the end (42 days old) of the test, and the BWG and the ADG were computed for the test period. Traits analyzed in the study were body weight at the 42th day, metabolic body weight at the 42th day (BW420.75), 15 to 42 days ADG, 15 to 42 days FI, 15 to 42 days FCR, and 15 to 42 days RFI to assess their genetic inter-relationships. FCR was calculated as FI divided by BWG. RFI was calculated according to Aggrey et al.  (2010), RFI = FI− (a+b1×BW420.75+b2×BWG).
Where “a” is the intercept and b1 and b2 are of partial regression coefficients of FI on BW420.75 and BWG, respectively. RFI values were obtained using the PROC REG procedure in SAS software package .
The genetic parameters for feed efficiency traits were estimated using restricted maximum likelihood (REML) methodology applied to a sire-dam model for all traits using the ASREML software .
Where Yijk is the record of the kth duck from the ith Sex and jth hatch; Sexi = fixed effect of sex (I = 1, 2 male/female), Hatchj = fixed effect of hatch (i = 1, 2); ak = random direct additive genetic effect of individual k, and eijk = random residual error.
At first the variance components were estimated for each trait using bivariate analyses including the selection criteria to properly account for the effect of selection . To estimate the genetic correlations between all pairs of feed efficiency growth traits, 3-trait analyses were performed, including the selection criteria as for bivariate analysis. The pedigree contained one generation. From the variance component estimations, the heritability of each trait i was estimated as h2i = 4×σ2i/σ2pi, where σ2i is the additive variance and σ2pi is the phenotypic variance for trait i. Genetic correlations rgij between traits i and j were estimated as , where covij is the covariance between traits i and j.
The means and standard deviations (SD) of the traits in the study are presented in Table 2. It should be noted that, for FCR and RFI, lower values indicate greater efficiency. During the test period, the average RFI was null 0 (SD, 229), the average FCR was 2.53 (SD, 0.18), and ADG was 65.59 g/d (SD, 5.23 g/d). The heritability estimates for all the traits were ranged from 0.32±0.11 for FI to 0.41±0.12 for BW420.75. The heritability of RFI was 0.41±0.18. The estimate of heritability for FCR (0.38±0.15) was of similar magnitude as the estimate of ADG (0.38±0.20). The estimate of heritability for BW was 0.39±0.12.
Phenotypic correlations between the feed efficiencies traits are presented in Table 3. The phenotypic correlation between RFI and FCR was 0.55±0.04; RFI was strongly correlated with FI (0.82±0.03); RFI was not found to be phenotypically correlated with BW42 and ADG. The phenotypic correlation between FCR and ADG was strongly negative (−0.80±0.04). The correlation was also negative between FCR and BW42 (0.65±0.02). The phenotypic correlation between FCR and FI was 0.44±0.01. ADG was strongly correlated with BW (0.92±0.01). The phenotypic correlation between ADG and FI was 0.48±0.02. The phenotypic correlation between BW and RI was 0.56±0.02.
Table 3 shows the genetic correlations between the feed efficiency traits. The genetic correlation between RFI and FCR or FI were positive (0.54±0.05 and 0.77±0.17), but not with the ADG and the BW. FCR was negatively correlated with ADG (0.80±0.11) and BW42 (0.64±0.14). ADG was strongly correlated with BW42 (0.92±0.08).
The FCR is a ratio trait that is subject to skewness and kurtosis as a result of the changes in the coefficient of variation of BWG (denominator) and subsequently affects SD, covariances and correlations . This is due to the disproportionate fashion by which selection pressure is exerted on the component traits. The RFI is a linear index the weights of the component traits in which are determined by only biological variances. The selection response based on a linear index would be higher than that of the response from a direct selection. With an understanding of the biological significance of RFI, RFI has increasingly become a preferred choice for measuring feed efficiency and is considered to be one of the target traits in animal breeding programs . But in the literature on ducks the data on RFI is scant.
The heritability estimates for FCR and RFI were 0.38 and 0.41 respectively, in our study population of Pekin duck. They were within the limits of published data in beef cattle, pigs and turkeys [5,6,8,21,22]; selection for low RFI will improve feed efficiency with an expected correlated response in FI. Based on the genetic correlation between RFI and metabolic body weight as obtained, it could be mentioned that this would favor obtaining a lower maintenance energy requirement in ducks. In the current study, the genetic interrelationships among the feed efficiency parameters were estimated from 14 days to 42 days of feeding. Cai et al.  and Hoque et al.  have also reported a positive genetic correlation between RFI and ADG in pigs selected for low RFI, which is similar to the genetic correlation in this study. However, a lack of genetic correlation was reported in cattle and pigs [5,9,24,25]. In pigs, RFI is negatively correlated to dressing percentage and positively correlated with back fat thickness . It is possible that the internal allocation of resources above maintenance into protein accretion and fat deposition among other animals could contribute towards the different interrelationships between factors that affect RFI at different periods .
This study’s conclusions and implications apply to growing ducks. The results have shown that the selection against RFI may improve the FCR trait in the young growing ducks, and their efficiencies to maintenance energy expenditures. But feed efficiency is a compound trait, and measured over a long period. Therefore, it is affected by many factors, and the genetic relationships among feed efficiency parameters are also vary with these factors.
This paper was supported by the Program of CARS-43 in China, and Program of 2011BAD28B03.
Stage 1: Cu 10 mg (CuSO4·5H2O), Fe 60 mg (FeSO4·7H2O), Zn 60 mg (ZnO), Mn 80 mg (MnSO4·H2O), Se 0.3 mg (NaSeO3), I 0.2 mg (KI), Cr 0.15 mg (Cr2O3), choline chloride, 1,000 mg; vitamin A 10,000 IU (retinyl acetate), vitamin D3 3,000 IU (cholecalciferol), vitamin E 20 IU (DL-α-tocopheryl acetate), vitamin K3 2 mg (menadione sodium bisulfate), thiamin 2 mg (thiamin mononitrate), riboflavin 8 mg, pyridoxine hydrochloride 4 mg, cobalamin 0.02 mg, calcium-D-pantothenate- 20 mg, nicotinic acid 50 mg, folic acid 1 mg, biotin 0.2 mg.
Stage 2: Cu 10 mg (CuSO4·5H2O), Fe 60 mg (FeSO4·7H2O), Zn 60 mg (ZnO), Mn 80 mg (MnSO4·H2O), Se 0.3 mg (NaSeO3), I 0.2 mg (KI), Cr 0.15 mg (Cr2O3), choline chloride, 750 mg; vitamin A 8,000 IU (retinyl acetate), vitamin D3 3,000 IU (cholecalciferol), vitamin E 20 IU (DL-α-tocopheryl acetate), vitamin K3 2 mg (menadione sodium bisulfate), thiamin 1.5 mg (thiamin mononitrate), riboflavin 8 mg, pyridoxine hydrochloride 3 mg, cobalamin 0.02 mg, calcium-D-pantothenate- 10 mg, nicotinic acid 50 mg, folic acid 1 mg, biotin 0.2 mg.
1. Zhang W., & Aggrey S. Genetic variation in feed utilization efficiency of meat-type chickens. World’s Poult Sci J. 2003; 59:328–39.
2. Koch RM, Swiger L, Chambers D., & Gregory K. Efficiency of feed use in beef cattle. J Anim Sci. 1963; 22:486–94.
3. Bordas A., & Merat P. Genetic variation and phenotypic correlations of food consumption of laying hens corrected for body weight and production. Br Poult Sci. 1981; 22:25–33.
4. Archer JA, Pitchford WS, Hughes TE., & Parnell PF. Genetic and phenotypic relationships between food intake, growth, efficiency and body composition of mice post weaning and at maturity. Anim Sci. 1998; 67:171–82.
5. Gilbert H, Bidanel J-P., & Gruand J. , et alGenetic parameters for residual feed intake in growing pigs, with emphasis on genetic relationships with carcass and meat quality traits. J Anim Sci. 2007; 85:3182–8.
6. Cai W, Casey D., & Dekkers J. Selection response and genetic parameters for residual feed intake in Yorkshire swine. J Anim Sci. 2008; 86:287–98.
7. Drouilhet L, Gilbert H., & Balmisse E. , et alGenetic parameters for two selection criteria for feed efficiency in rabbits. J Anim Sci. 2013; 91:3121–8.
8. Aggrey SE, Karnuah AB, Sebastian B., & Anthony NB. Genetic properties of feed efficiency parameters in meat-type chickens. Genet Sel Evol. 2010; 42:1–5.
9. Arthur P, Archer J, Johnston D, Herd R, Richardson E., & Parnell P. Genetic and phenotypic variance and covariance components for feed intake, feed efficiency, and other postweaning traits in Angus cattle. J Anim Sci. 2001; 79:2805–11.
10. Schenkel F, Miller S., & Wilton J. Genetic parameters and breed differences for feed efficiency, growth, and body composition traits of young beef bulls. Can J Anim Sci. 2004; 84:177–85.
11. Cruz GD, Rodríguez-Sánchez JA, Oltjen JW., & Sainz RD. Performance, residual feed intake, digestibility, carcass traits, and profitability of Angus-Hereford steers housed in individual or group pens. J Anim Sci. 2010; 88:324–9.
12. Durunna ON, Wang Z, Basarab JA, Okine EK., & Moore SS. Phenotypic and genetic relationships among feeding behavior traits, feed intake, and residual feed intake in steers fed grower and finisher diets. J Anim Sci. 2011; 89:3401–9.
13. Durunna ON, Colazo MG, Ambrose DJ, Mccartney D, Baron VS., & Basarab JA. Evidence of residual feed intake reranking in crossbred replacement heifers. J Anim Sci. 2012; 90:734–41.
14. Mrode R., & Kennedy B. Genetic variation in measures of food efficiency in pigs and their genetic relationships with growth rate and backfat. Anim Prod. 1993; 56:225–32.
15. Le NT, Le FHN, Louveau I, Gilbert H., & Gondret F. Metabolic changes and tissue responses to selection on residual feed intake in growing pigs. J Anim Sci. 2012; 90:4771–80.
16. SAS (Statistical Analysis System) Insitute Inc. SAS User’s Guide: Version 8.12. Cary, NC, USA: SAS Institue Inc.;1998.
17. Gilmour AR, Gogel BJ, Cullis BR., & Thompson R. ASREML User Guide. Release 3.0. Hemel Hempstead, UK: VSN International Ltd;2009.
18. Hofer A. Variance component estimation in animal breeding: a review. J Anim Breed Genet. 1998; 115:247–65.
19. Atchley WR, Gaskins CT., & Anderson D. Statistical properties of ratios. I. Empirical results. Syst Biol. 1976; 25:137–48.
20. Fan B, Lkhagvadorj S., & Cai W. , et alIdentification of genetic markers associated with residual feed intake and meat quality traits in the pig. Meat Sci. 2010; 84:645–50.
21. Aggrey SE, Karnuah AB, Sebastian B., & Anthony NB. Research Genetic properties of feed efficiency parameters in meat-type chickens. Genet Sel Evol. 2010; 42:25
22. Case LA, Wood BJ., & Miller SP. The genetic parameters of feed efficiency and its component traits in the turkey (Meleagris gallopavo). Genet Sel Evol. 2012; 44:2
23. Hoque M, Kadowaki H, Shibata T, Oikawa T., & Suzuki K. Genetic parameters for measures of residual feed intake and growth traits in seven generations of Duroc pigs. Livest Sci. 2009; 121:45–9.
24. Herd R., & Bishop S. Genetic variation in residual feed intake and its association with other production traits in British Hereford cattle. Livest Prod Sci. 2000; 63:111–9.