Skip to main content
Fig. 6 | Animal Microbiome

Fig. 6

From: An artificial intelligence approach of feature engineering and ensemble methods depicts the rumen microbiome contribution to feed efficiency in dairy cows

Fig. 6

Machine learning prediction using feature selection and Ridge regression in 10-fold cross-validation on the milk protein efficiency (MPE) of lactating dairy cows. 6.A. Model used for the prediction of MPE containing the main energy input and sinks in a lactating dairy cow (DMI, NESec, MBW, and BEC), the effect of parity, the effect of previous treatment plus the effect of the rumen microbiome; 6.B. Prediction of MPE using only the rumen microbiome to explore the overall contribution of the rumen microbes to the trait; 6.C. Prediction of the residual MPE, using only the rumen microbiome composition to depict the contribution of the rumen microbiome to the unexplained variance in the trait; and 6.D. Summary of the residual MPE prediction with the rumen microbiome, from which confusion matrix was derived from comparing extracted observed vs. predicted residuals. The loss function for the AI model was the mean squared error (MSE) and the scoring metric was R2

Back to article page

Animal Microbiome

ISSN: 2524-4671