Evaluation of growth functions in describing the growth of Japanese quails fed diets deferring in nutrients density compared to NRC recommendation

The Japanese quails are blessed with many desirable characteristics, viz, faster growth, early sexual maturity, high rate of egg production (300 egg/annum), short generation interval (3-4 generations a year), small floor space, less feed requirements (20-25 g/adult bird/day), short incubation period of hatching eggs, less feed cost, and less susceptibility to common chicken diseases (Ernst, 2000). In consequence and due to shortage of available data on nutrient requirements of quails, prediction of biological responses of this type of poultry to nutrients using growth functions seems to be beneficial. Growth modeling are also critical for the understanding and formulation of breeding plans because they shift in response to selection (Marks et al. 1988, Mignon-Grasteau et al. 1999, Aggrey 2003 and Beiki et al. 2011). The representation of biological concepts through the simulation of growth dynamics enables us to better adapt management and nutrition to the requirements of the animals, while taking into account the interaction between genotype, nutrition and environmental conditions [Torres and Ferket 2012]. Eleroğlu et al. (2014) pointed out that the application of mathematical functions to describe the growth of birds was useful in estimating production termination deadline and formulation of appropriate feed mixtures (Michalczuk et al. 2016). Ersoy et al. (2006) claimed the Gompertz and Richards models to be the most appropriate for the characteristics of the growth of chickens, ostriches, turkeys and emus. Conflicting concluding by various authors upon the use of the same functions may result from the use of different genetic groups of birds in experiments. The objective of the present study was 1) to determine the best predictive growth function in describing data from Japanese quails and 2) to investigate the effect of dietary nutrients density on the growth curve parameters estimated by the growth models.
Material and One thousand and fifty one-day-old Japanese quails were randomly divided into 4 dietary treatment groups with 3 replicates of 86 quails in each in order to compare four growth functions (Gompertz, Richards, von Bertalanffy and Lopez) for their predicable abilities in describing growth of Japanese quails. The treatment groups were: 1) group with low dietary nutrient density [95% of nutrients recommended by NRC (1994), -5% NRC], 2) group with medium dietary nutrient density [100% of nutrients recommended by NRC (1994)], 3) group with high dietary nutrient density [105% of nutrients recommended by NRC (1994), % NRC] and 4) group with very high dietary nutrient density [110% of nutrients recommended by NRC (1994), % NRC]. Body weights of the birds were measured weekly over the 56 day of the experimental period. Evaluation on the goodness of fit for the models were made by R2, AIC, BIC, and RMSE criteria.
Evaluation on the goodness of fit for the models using R2, RMSE, AIC, BIC criteria showed advantage for the Richards in describing the growth data of Japanese quails which can be related to the variable point of inflexion in the Richards model and therefore its flexibility. According to the four growth functions considered, estimated final body weights were higher in the high density than the low density diets. The age at point of inflection were earlier in high versus low density diets. As early as 1945, Brody suggested that it was possible to select on the shape of the growth curve. A phenotypic modification of the growth curve was also observed in previous studies. Marks et al. 1988, Mignon-Grasteau et al. (1999), Aggrey 2003 and Beiki et al. (2011) showed that growth curve parameters were heritable. However, data shortage of the nutritional requirement necessary for quail breeding makes the mathematic simulation methods very useful in estimating the biological answer to the food nutrients contribution. The method of mathematic simulation of growth using growth functions can also provide the response of the growth parameters to nutrition (Daren and Marks 1988; Marks 1991; Gebhardt-Henrich and Marks 1993) which has the potential benefits when used for selection.
In summary, it was concluded that data regarding the growth parameters of quails would be best interpreted with the use of the Richards model. Meanwhile, since almost all the growth parameters estimated by the models were affected by the dietary nutrients density, therefore special attention needed will be given to the nutrition when selecting the quails as parental for the next generation through parameters estimated by mathematical simulation using growth functions.