Prediction of carcass weight using multiple regression, Bayesian networks and artificial neural networks in Nigerian indigenous chickens based on earlier expressed phenotypic traits

Main Article Content

A.S. Adenaike
O.S. Oloye
M.A. Opoola
H.O. Emmanuel
C.O.N. Ikeobi

Abstract

The prediction potential of carcass weight (CW) in chicken is an important undertaking that can help commercial enterprises make better management decisions. The goal of this study was to examine alternative modeling approaches for predicting CW in meat-type Nigerian indigenous chickens using 19 biometric variables, as well as to discover early expressed traits that may be employed in CW breeding selection. Using multiple linear regression (MLR) and stepwise regression (SWREG), artificial neural networks (ANNs), and Bayesian networks (BN), the biometric traits of 320 chickens were modeled to predict CW. The accuracy of the models was evaluated based on their values of root mean square error (RMSE), mean square error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE), coefficient of determination (R2), and correlation coefficient (r) between the predicted and the observed values of CW. The results showed that the MLR model was the least capable of predicting CW (MAE = 0.608, RMSE = 2.020, and MAPE = 93.244), followed by SWREG (MAE = 0.426, RMSE = 0.855, and MAPE = 77.168) compared to the ANNs and BN models. The estimated values of MAE, RMSE, and MAPE for the ANN1 model were 0.091, 0.201, and 52.891 respectively while that of ANN3 were 0.081, 0.101, and 36.765 respectively. The estimated values of MAE, RMSE, and MAPE for the MMHC model were 0.095, 0.129, and 63.551 respectively while that of RSMAX2 were 0.099, 0.132, and 66.193 respectively. Although it is possible to achieve a higher-performing SWREG model, in this study the SWREG (R2 = 57.84%) cannot be considered an optimum model for predicting CW. Based on statistical parameters (i.e., R2, MAE, r, and MAPE), the result of the study showed that the BN models provided a more powerful tool than the regression models and ANNs for predicting CW. The findings of this study showed that day-old chick weight, hatched weight, live weight, and body weight at 8 weeks are good predictors of CW. This could be used for management decisions in the chicken industry in the determination of CW at an earlier age of chickens.

Article Details

Section
Research Articles

References

Abdipour, M., M. Ebrahimi, A. Izadi-Darbandi, A.M. Mastrangelo, G. Najafian, Y. Arshad and G. Mirniyam. 2016. Association between grain size and shape and quality traits, and path analysis of thousand grain weight in Iranian bread wheat landraces from different geographic regions. Not. Bot. Horti. Agrobo. 44(1): 228–236.

Adenaike, A.S., A.E. Ogundero, N. Taiwo and C.O.N. Ikeobi. 2018. Use of path analysis to investigate association between body weight and body dimensions (body metric traits) in Nigerian locally adapted turkeys. Pertanika J. Trop. Agric. Sci. 41(4): 1865–1874.

Adenaike, A.S., U. Akpan and C.O.N. Ikeobi 2015. Principal components regression of body measurements in five strains of locally adapted chickens in Nigeria. Thai J. Agric. Sci. 48(4): 217–225.

Ahmad, H.A. 2009. Poultry growth modeling using neural networks and simulated data. J. Appl. Poult. Res. 18(3): 440–446.

Basak, J.K., E. Arulmozhi, F. Khan, F.G. Okyere, J. Park, D.H. Lee and H.T. Kim. 2020. Assessment of the influence of environmental variables on pig’s body temperature using ANN and MLR models. Indian J. Anim. Res. 54(9): 1165–1170.

Bishop, C.M. 2006. Pattern Recognition and Machine Learning. Springer, New York, USA.

Bolzan, A.C., R.A.F. Machado and J.C.Z. Piaia. 2008. Egg hatchability prediction by multiple linear regression and artificial neural networks. Braz. J. Poult. Sci. 10(2): 97–102.

Burnham, K.P. and D.R. Anderson. 2002. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. 2nd Edition. Springer Verlag, New York, USA.

Chomchuen, K., V. Tuntiyasawasdikul, V. Chankitisakul and W. Boonkum. 2022. Genetic evaluation of body weights and egg production traits using a multi-trait animal model and selection index in Thai native synthetic chickens (Kaimook e-san2). Animals 12(3): 335.

Correa, M., C. Bielza and J. Pamies-Teixeira. 2009. Comparison of Bayesian networks and artificial neural networks for quality detection in a machining process. Expert Syst. Appl. 36(3): 7270–7279.

Felipe, V.P.S., M.S. Silva, B.D. Valente and G.J.M. Rosa. 2015. Using multiple regression, Bayesian networks and artificial neural networks for prediction of total egg production in European quails based on earlier expressed phenotypes. Poult. Sci. 94(4): 772–780.

Friedman, N., I. Nachman and D. Peér. 1999. Learning Bayesian network structure from massive datasets: the “Sparse Candidate” algorithm, pp. 206–215. In: Proceedings of the 15th Conference on Uncertainty in Artificial Intelligence. California, USA.

Gasse, M., A. Aussem and H. Elghazel. 2014. A hybrid algorithm for Bayesian network structure learning with application to multi-label learning. Expert Syst. Appl. 41(15): 6755–6772.

Ghazanfari, S., K. Nobari and M. Tahmoorespur. 2011. Prediction of egg production using artificial neural network. Iran. J. Appl. Anim. Sci. 1(1): 11–16.

Ghoreishi, M., Y. Hossini and M. Maftoon. 2012. Simple models for predicting leaf area of mango (Mangifera indica L.). J. Biol. Earth Sci. 2(2): B45–B53.

Golkar, P., A. Arzani and A.M. Rezaei. 2011. Determining relationships among seed yield, yield components and morpho-phenological traits using multivariate analyses in safflower (Carthamus tinctorius L.). Ann. Biol. Res. 2(3): 162–169.

Hageman, R.S., M.S. Leduc, R. Korstanje, B. Paigen and G.A. Churchill. 2011. A Bayesian framework for inference of the genotype–phenotype map for segregating populations. Genetics 187(4): 1163–1170.

Hahn, G. and M. Spindler. 2002. Method of dissection of turkey carcasses. Worlds Poult. Sci. J. 58(2): 179–197.

Hastie, T., R. Tibshirani and J. Friedman 2009. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd Edition. Springer, New York, USA.

Heald, C.W., T. Kim, W.M. Sischo, J.B. Cooper and D.R. Wolfgang. 2000. A computerized mastitis decision aid using farm-based records: an artificial neural network approach. J. Dairy Sci. 83(4): 711–720.

Huang, J., X. Wang, X. Li, H. Tian and Z. Pan. 2013. Remotely sensed rice yield prediction using multitemporal NDVI data derived from NOAA’s-AVHRR. PLoS ONE 8(8): e70816.

Mehri, M. 2012. Development of artificial neural network models based on experimental data of response surface methodology to establish the nutritional requirements of digestible lysine, methionine, and threonine in broiler chicks. Poult. Sci. 91(12): 3280–3285.

Morota, G., B.D. Valente, G.J.M. Rosa, K.A. Weigel and D. Gianola. 2012. An assessment of linkage disequilibrium in Holstein cattle using a Bayesian network. J. Anim. Breed. Genet. 129(6): 474–487.

Neto, E.C., M.P. Keller, A.D. Attie and B.S. Yandell. 2010. Causal graphical models in systems genetics: a unified framework for joint inference of causal network and genetic architecture for correlated phenotypes. Ann. Appl. Stat. 4(1): 320–339.

Okpeku, M., D.M. Ogah and M.A. Adeleke. 2019. A review of challenges to genetic improvement of indigenous livestock for improved food production in Nigeria. Afr. J. Food Agric. Nutr. Dev. 19(1): 13959–13978.

Pal, S.K. and S. Mitra. 1992. Multilayer perceptron, fuzzy sets and classification. IEEE Trans. Neural Netw. 3(5): 683–697.

Peñagaricano, F., B.D. Valente, J.P. Steibel, R.O. Bates, C.W. Ernst, H. Khatib and G.J.M. Rosa. 2015. Exploring causal networks underlying fat deposition and muscularity in pigs through the integration of phenotypic, genotypic and transcriptomic data. BMC Syst. Biol. 9: 58.

R Core Team. 2021. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.

Scutari, M. 2010. Learning Bayesian networks with the bnlearn R package. J. Stat. Softw. 35(3): 1–22.

Scutari, M., P. Howell, D.J. Balding and I. Mackay. 2014. Multiple quantitative trait analysis using Bayesian networks. Genetics 198: 129–137.

Scutari, M. and J.B. Denis. 2015. Bayesian Networks with Examples in R. Taylor & Francis Group, New York, USA.

Semsarian, S., M.P.E. Nasab, S. Zarehdaran and A.A. Dehghani. 2013. Prediction of the weight and number of eggs in Mazandaran native fowl using artificial neural network. Int. J. Adv. Biol. Biomed. Res. 1(5): 532–537.

Tsamardinos, I., L.E. Brown and C.F. Aliferis. 2006. The max-min hill-climbing Bayesian network structure learning algorithm. Mach. Learn. 65: 31–78.

Wang, B.Y., S.A. Chen and S.W. Roan. 2012. Comparison of regression and artificial neural network models of egg production. J. Anim. Vet. Adv. 11(14): 2503–2508.

Wang, H. and F.A. van Eeuwijk. 2014. A new method to infer causal phenotype networks using QTL and phenotypic information. PLoS ONE 9: e103997.

Whittingham, M.J., P.A. Stephens, R.B. Bradbury and R.P. Freckleton. 2006. Why do we still use stepwise modelling in ecology and behaviour? J. Anim. Ecol. 75(5): 1182–1189.

Yakubu, A., O.I.A. Oluremi and Z.N. Ibrahim. 2018. Modelling egg production in Sasso dual-purpose birds using linear, quadratic, artificial neural network and classification regression tree methods in the tropics. Livest. Res. Rural Dev. 30(10): 172.