Adsa Logo White Adsa Title White

Predictions of daily total sperm production of Holstein and Jersey bulls using health and management data.

A. Quick

Events

06-24-2020

Join A. Quick on this page for a live text chat!
6:00 PM - 8:00 PM GMT

Abstract:

W14
Predictions of daily total sperm production of Holstein and Jersey bulls using health and management data.
A. Quick*1, J. Meronek2, K. Weigel1. 1University of Wisconsin, Department of Dairy Science, College of Agricultural & Life Sciences Madison, WI, 2ABS Global Inc DeForest, WI.

Selection of elite young bulls using genomic data has shortened the generation interval and increased pressure to collect large quantities of semen at an early age. The objective of this study was to develop a prediction model for daily total sperm (TS) production from collection history, health status, and management factors. Data were obtained from a commercial AI company and included 47,199 daily collection records from 1,118 Holstein and Jersey bulls between 10 and 30 mo of age. Potential explanatory variables included: year and season of collection, barn location at collection, days since last collection, number of collections per week, number of ejaculates per collection day, breed, scrotal circumference (SC) at 10—11 mo of age, prior health events (binary incidences in last 1, 3, or 6 mo), and age at arrival, first collection, and current collection. Four analyses were carried out based on availability of prior TS data, each split into 75% training data and 25% testing data: no prior collection requirement (Lag0; n = 47,199), at least one prior collection (Lag1, n = 46,081), at least 2 prior collections (Lag2, n = 44,987), and at least 3 prior collections (Lag3, n = 43,918). Adjusted R-squared for linear regression models (R v3.5.1) ranged from 0.42 for Lag0 to 0.54 for Lag3. The most significant explanatory variables were collection age, days since last collection, number of collections per week, and number of ejaculates per collection day (P < 2 � 10−16). Inclusion of previous TS values in the lag analyses (Lag1, Lag2, Lag3) was highly significant (P < 2 � 10−16), with TS second prior having most influence. In all 4 analyses, SC maximized TS when SC is 32.5 to 34.5 cm (P < 6.12 � 10−7) and minimized when SC <28.5 cm (P < 0.05). Age at first collection and incidence of hoof trimming (last 6 mo), dehorning, and respiratory system (last 3 mo) events were significant (P < 0.05) in all 4 analyses. Jerseys (P < 4.62 � 10−14) produced more TS than Holsteins.

Keywords: sperm production, management factors, prediction.

Biography: Allison Quick is a PhD student in Dairy Science at University of Wisconsin-Madison. Quick's research focuses on data analytics and quantitative genetics, specifically forecasting sperm production in dairy bulls, using management, health, and genotype data.