Modeling of individual lactation curves for milk production in a population of Alpine goats in Cuba

González-Peña, Dianelys; Acosta, José; Guerra, Danilo Dourado; Gonzalez, NC; Acosta, M. J.; Sosa, D.; Torres-Hernández, Glafíro

doi:10.1016/j.livsci.2012.07.026

Cited by 10 publications

(13 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, studies using RRM based on parametric functions are most often found in the literature about goats as a milk-trait analysis. (González-Peña et al, 2012;León et al, 2012).…”

Section: Multiple-trait Random Regressionmentioning

confidence: 99%

“…Under a single-trait RRM, studies about the milk yield of dairy goats using splines (Thepparat et al, 2014;León et al, 2012) and other specific functions to model the lactation curve (González-Peña et al, 2012) have demonstrated equal or better fit than the traditional Legendre polynomial models. In relation to the modeling of milk constituents, the majority of the studies about single-trait RRM have also been based on Legendre polynomials (Andonov et al, 2013), but some studies in cows have shown that splines are better to describe fat and protein yield (Bouallegue et al, 2015).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combining different functions to describe milk, fat, and protein yield in goats using Bayesian multiple-trait random regression models1

Oliveira

Silva

Siqueira

et al. 2016

Journal of Animal Science

View full text Add to dashboard Cite

We proposed multiple-trait random regression models (MTRRM) combining different functions to describe milk yield (MY) and fat (FP) and protein (PP) percentage in dairy goat genetic evaluation by using Bayesian inference. A total of 3,856 MY, FP, and PP test-day records, measured between 2000 and 2014, from 535 first lactations of Saanen and Alpine goats, including their cross, were used in this study. The initial analyses were performed using the following single-trait random regression models (STRRM): third- and fifth-order Legendre polynomials (Leg3 and Leg5), linear B-splines with 3 and 5 knots, the Ali and Schaeffer function (Ali), and Wilmink function. Heterogeneity of residual variances was modeled considering 3 classes. After the selection of the best STRRM to describe each trait on the basis of the deviance information criterion (DIC) and posterior model probabilities (PMP), the functions were combined to compose the MTRRM. All combined MTRRM presented lower DIC values and higher PMP, showing the superiority of these models when compared to other MTRRM based only on the same function assumed for all traits. Among the combined MTRRM, those considering Ali to describe MY and PP and Leg5 to describe FP (Ali_Leg5_Ali model) presented the best fit. From the Ali_Leg5_Ali model, heritability estimates over time for MY, FP. and PP ranged from 0.25 to 0.54, 0.27 to 0.48, and 0.35 to 0.51, respectively. Genetic correlation between MY and FP, MY and PP, and FP and PP ranged from -0.58 to 0.03, -0.46 to 0.12, and 0.37 to 0.64, respectively. We concluded that combining different functions under a MTRRM approach can be a plausible alternative for joint genetic evaluation of milk yield and milk constituents in goats.

show abstract

“…However, studies using RRM based on parametric functions are most often found in the literature about goats as a milk-trait analysis. (González-Peña et al, 2012;León et al, 2012).…”

Section: Multiple-trait Random Regressionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Combining different functions to describe milk, fat, and protein yield in goats using Bayesian multiple-trait random regression models1

Oliveira

Silva

Siqueira

et al. 2016

Journal of Animal Science

View full text Add to dashboard Cite

show abstract

“…This suggests the ALISCH model enables capturing a greater fraction of the variability in the data sample when compared to other models, and the great repercussion that time evolution has on the components being measured. González-Peña et al [ 45 ] and Harder et al [ 46 ] found slightly lower results for the Adj. R 2 of the ALISCH model, which can be attributed to the properties and characteristics of the sample that was used.…”

Section: Discussionmentioning

confidence: 97%

“…According to González-Peña et al [ 45 ], the Ali and Schaeffer model (ALISCH) [ 57 ] and the third-order orthogonal polynomials of Legendre (3ORDLEG) were able to recognize 9 and 14 types of curves. The correlations between the estimated parameter values for ALISCH were greater than those estimated for 3ORDLEG.…”

Section: Discussionmentioning

confidence: 99%

Goat Milk Nutritional Quality Software-Automatized Individual Curve Model Fitting, Shape Parameters Calculation and Bayesian Flexibility Criteria Comparison

Inostroza

González

Landi

et al. 2020

Animals

View full text Add to dashboard Cite

SPSS syntax was described to evaluate the individual performance of 49 linear and non-linear models to fit the milk component evolution curve of 159 Murciano-Granadina does selected for genotyping analyses. Peak and persistence for protein, fat, dry matter, lactose, and somatic cell counts were evaluated using 3107 controls (3.91 ± 2.01 average lactations/goat). Best-fit (adjusted R2) values (0.548, 0.374, 0.429, and 0.624 for protein, fat, dry matter, and lactose content, respectively) were reached by the five-parameter logarithmic model of Ali and Schaeffer (ALISCH), and for the three-parameter model of parabolic yield-density (PARYLDENS) for somatic cell counts (0.481). Cross-validation was performed using the Minimum Mean-Square Error (MMSE). Model comparison was performed using Residual Sum of Squares (RSS), Mean-Squared Prediction Error (MSPE), adjusted R2 and its standard deviation (SD), Akaike (AIC), corrected Akaike (AICc), and Bayesian information criteria (BIC). The adjusted R2 SD across individuals was around 0.2 for all models. Thirty-nine models successfully fitted the individual lactation curve for all components. Parametric and computational complexity promote variability-capturing properties, while model flexibility does not significantly (p > 0.05) improve the predictive and explanatory potential. Conclusively, ALISCH and PARYLDENS can be used to study goat milk composition genetic variability as trustable evaluation models to face future challenges of the goat dairy industry.

show abstract

“…A mixed model was applied to analyze the various factors that affect the traits under study. Next, the individual lactation curves were analyzed using the Spline model, which has proved to be the one that provides the best goodness of fit for the MG breed (R 2 = 0.98, mean squared error (MSE) = 0.0020 without existence of autocorrelation among residuals) than several other common functions (Wood, Cappio-Borlino, Cobby and Le Du, Wilmink and Legendre) [6] and for a better estimation of genetic parameters [10]. The Spline function with one knot used: y t = β 0 + β 1 t + β 2 t 2 for t ≤ X , y t = β 0 + β 1 t + β 2 t 2 + β 3 ( t − X ) 2 for t > X , where y t represents the average daily milk production recorded on day t, while β 0 , β 1 , β 2 and β 3 are the parameters specifically designed to adapt to the shape of the lactation curves due to its flexibility.…”

Section: Methodsmentioning

confidence: 99%

Estimation of Genetic Parameters for Peak Yield, Yield and Persistency Traits in Murciano-Granadina Goats Using Multi-Traits Models

Miranda

León

Pieramati

et al. 2019

Animals

View full text Add to dashboard Cite

This paper studies parameters of a lactation curve such as peak yield (PY) and persistency (P), which do not conform to the usual selection criteria in the Murciano-Granadina (MG) breed, but are considered to be an alternative to benefit animal welfare without reducing production. Using 315,663 production records (of 122,883 animals) over a period of 24 years (1990–2014), genetic parameters were estimated with uni-, bi- and multivariate analysis using multiple trait derivative free restricted maximum likelihood (MTDFREML). The heritability (h2)/repeatability (re) of PY, yield (Y) and P was estimated as 0.13/0.19, 0.16/0.25 and 0.08/0.09 with the uni-trait and h2 of bi- and multi-traits analysis ranging from 0.16 to 0.17 of Y, while that of PY and Y remained constant. Genetic correlations were high between PY–Y (0.94 ± 0.011) but low between PY–P (–0.16 ± 0.054 to –0.17 ± 0.054) and between Y–P (–0.06 ± 0.058 to –0.05 ± 0.058). Estimates of h2/re were low to intermediate. The selection for Y–PY or both can be implemented given the genetic correlation between these traits. PY–P and Y–P showed low to negligible correlation values indicating that if these traits are implemented in the early stages of evaluation, they would not be to the detriment of PY–Y. The combination of estimated breeding values (EBVs) for all traits would be a good criterion for selection.

show abstract

Modeling of individual lactation curves for milk production in a population of Alpine goats in Cuba

Cited by 10 publications

References 42 publications

Combining different functions to describe milk, fat, and protein yield in goats using Bayesian multiple-trait random regression models1

Combining different functions to describe milk, fat, and protein yield in goats using Bayesian multiple-trait random regression models1

Goat Milk Nutritional Quality Software-Automatized Individual Curve Model Fitting, Shape Parameters Calculation and Bayesian Flexibility Criteria Comparison

Estimation of Genetic Parameters for Peak Yield, Yield and Persistency Traits in Murciano-Granadina Goats Using Multi-Traits Models

Contact Info

Product

Resources

About