Developing a new selection index for the Italian Mediterranean buffalo (Bubalus bubalis)

The aim of this work was to develop a new selection index and to estimate the annual genetic response for milk yield (MY), milk components, composite feet and legs (FL) and mammary system (MS) in the Italian Mediterranean Buffalo (IMB) population under different selection indices and breeding objectives. Buffalo milk has excellent nutritional characteristics as a consequence of the high percentage of its constituents. These milk properties are desirable for the dairy industry because mainly used to make Campana buffalo mozzarella PDO cheese. Former breeding objective of the National Association of Buffalo Breeders (ANASB) was based only on the evaluation of kg of mozzarella production (PKM index). Hence, selection by PKM was mainly carried out according to the kg of fat and protein and consequently milk yield. To address the call of mozzarella cheese manufacturers and farmers community, who both asked for a more balanced breeding objective, annual genetic response for the aforementioned traits and correlated response with mozzarella cheese production were estimated using selection index theory. Genetic parameters among traits included in the selection index and traits included in the breeding objectives were estimated using data from 7199 buffalo cows and a pedigree including 19,574 buffalos. A multitrait animal model was fitted using a Bayesian implementation via Gibbs sampling. Three alternative breeding objectives were formulated according to relative weight given to MY, cheese production, FL and MS: 5:45:30:20 (S1), 30:35:15:20 (S2) and 45:45:5:5 (S3). Traits included in the selection index were MY, fat and protein percentage, FL and MS. Cheese production had a negative genetic correlation with MY (-0.54), a null correlation with MS (-0.01), a positive correlation with FL (0.25), fat % (0.15) and protein % (0.28). S1 breeding objective had the largest and positive genetic response in terms of cheese production, fat and protein %. However, MY was strongly penalised, with a negative genetic response. S3 showed the best results with a positive genetic response for all traits included in the breeding objective ranging from 0.02 to 0.05 genetic standard deviation. Relative emphasis of selection criteria in breeding objective S3, namely MY, fat %, protein %, FL and ML were 21%, 15%, 20%, 24% and 20 %, respectively.