The use of machine learning methods to predict sperm quality in Holstein bulls.

The aim of this study was to develop prediction models for total sperm motility, morphological abnormalities and sperm output based on 1,551 ejaculate records of 58 Holstein bulls. The data was collected from September 2019 to November 2020 in a single artificial insemination (AI) center located in Eastern Germany. Factors considered for the prediction models include barn climate conditions, semen collector, number of false mounts, libido, semen collection frequency, breed and age (10-74 months). In this study, the prediction models Lasso, Group Lasso and Gradient Boosting were evaluated. The best model for each sperm quality parameter was chosen using cross validation. The models were estimated with five algorithms for sperm motility and sperm morphology and three algorithms for the number of total sperm per ejaculate (sperm output). For sperm motility and morphology a binary classification algorithm was applied, reaching an accuracy of over 80% for all models. For sperm output, no such classification was used and the only variable selected by all three algorithms was age. Furthermore, for sperm morphology, climate variables were frequently selected. Additionally, network diagrams from Group Lasso show the interdependencies between the major variable groups influencing sperm motility and morphology. In conclusion, the implementation of such prediction tools could help AI centers to optimize management factors and stabilize bull semen production in the future.

Development of predictive models for boar semen quality.

In this study a prognosis model is developed that predicts sperm quality characteristics based on external factors such as barn climate conditions, seasonality, semen collection frequency, age and breed of artificial insemination (AI) boars. For this a k-fold cross validation framework is used to test the prediction accuracy of a wide range of regression models that are based on different functional forms (linear, log-linear) and estimation techniques (ordinary least squares, seemingly unrelated regression, two-stage least squares estimation and three-stage least squares estimation). The dataset includes 241 boars from three barns within one boar stud located in Southern Germany, consisting of 7455 ejaculates collected during one year. The winner model predicts sperm motility with little error (Mean Absolute Percentage Error (MAPE): 4.35%), but is of limited use to predict sperm output (MAPE: 23.92%) and especially morphologically abnormal spermatozoa (MAPE: 44.67%). An estimation of marginal effects shows, that once confounding variables are controlled for, the considered barn climate variables do not have a measurable effect on sperm quality. Other factors have a more significant effect on sperm quality, like morphology-motility linkages, sperm concentration, interval between semen collections and to a lesser extent age and breed of the AI boar.