Linking laboratory findings to field fertility: a comparative study of frozen semen from dairy and beef bulls

Artificial insemination in cattle relies on laboratory semen evaluation, yet field fertility often diverges from laboratory-based expectations. This study compared spermatological and computer-assisted sperm analysis (CASA) kinematic traits of commercially frozen-thawed semen from dairy and beef bulls and examined their associations with pregnancy outcome under commercial conditions. Frozen semen from 11 bulls (beef: Angus, <i>n</i> = 5; dairy: Holstein, <i>n</i> = 6) was obtained through the Amasya Cattle Breeders' Association; six straws from the same batch per bull were analysed and averaged at the bull level. After thawing (37 °C, 30 s), motility, sperm concentration, and kinematic parameters were assessed using CASA under standardized settings. Field inseminations were performed by a single veterinarian in healthy multiparous cows showing spontaneous estrus. A total of 413 inseminations were conducted (beef semen: 194; dairy semen: 219); pregnancy diagnosis was available for 409 inseminations at day 60. Mean daily milk yield (days 5-305) was derived from association records (beef females, <i>n</i> = 194; dairy females, <i>n</i> = 219). Semen from beef bulls showed higher total motility (83.59 ± 2.08% vs. 56.24 ± 2.42%; <i>p</i> < 0.0001) and higher progressive motility (58.54 ± 3.01% vs. 48.25 ± 9.57%; <i>p</i> = 0.0475) than semen from dairy bulls. Dairy bulls had higher STR and LIN and lower ALH, whereas velocity descriptors differed modestly and did not show consistent between-group separation. Mean bull-level pregnancy rate was numerically higher for beef than dairy bulls (68.06% vs. 63.71%), without a significant between-group difference at the insemination level. Daily milk yield differed markedly between female groups (18.47 ± 0.38 vs. 6.21 ± 0.19 kg/d; <i>p</i> < 0.001). Within-group Pearson correlations indicated that motility was positively associated with pregnancy rate in both production types, while several kinematic descriptors exhibited production-type-specific correlation directions. These findings suggest that production context may modulate how CASA phenotypes relate to field fertility, supporting cautious interpretation of single-parameter semen metrics in commercial AI systems.