This study was conducted to define health traits and to estimate their genetic parameters by using on-farm management data in dairy cattle.

In this study, the health event records were collected by extracting from farm management software from 46 large-scale dairy farms in Northern China. Totally, 1 326 kinds of health events were grouped into five categories by standardizing the acronyms of on-farm records, 18 kinds of most frequent health events were selected from five categories. According to whether a health event occurred at least once within a lactation, 18 binary individual health traits corresponding to the 18 selected health events were defined (observations being 0 or 1). Furthermore, in order to define cow's general resistance to certain type of health problem, five binary general health traits were defined according to whether health problems within a health category occurs at least once in a lactation (observations being 0 or 1). The single trait and two traits linear animal models were used to estimate the genetic parameters for 23 newly defined health traits.

The estimated heritabilities for 23 health traits ranged from 0 (rumen acidosis) to 0.03 (milk fever). Udder health, reproductive disorders and metabolic disorders had the highest heritability (approximately 0.02) among five general health traits, whereas digestive disorders and hoof health had relatively low heritability (less than 0.01). Clustering health events into categories resulted in higher heritability for reproductive disorders and digestive disorders, while heritabilities of udder health, metabolic disorders and hoof health were lower than that of single health traits with the highest heritability in their respective category. The low genetic correlations between different health category traits were found; however, the moderate genetic correlations among some health traits from same category were observed. The health traits within the hoof health had the high genetic correlations with each other, ranged from 0.63 (Laminitis and Footrot) to 0.99 (Laminitis and Sole ulcer). For reproductive disorders, retained placenta had medium genetic correlations with metritis (0.47) and endometritis (0.46), respectively. For digestive disorders, relatively high genetic correlations were found between diarrhea and enteritis (0.94) as well as dyspepsia and antony of forestomachs (0.80). The ketosis and abomasum displacement had a genetic correlation of 0.42.

Based on the current data quality of health records in Chinese dairy farm and the results from the current study, it was suggested that five individual health traits (clinical mastitis, metritis, ketosis, abomasum displacement and milk fever) and two general health traits (digestive disorders and hoof health) could be considered as target traits, and they should be intensively considered in research and breeding practice for improving health traits genetically in Chinese Holstein population. This study proved the feasibility of defining health traits using on-farm management data in Chinese dairy cattle. The results from the current study provided a reference for research and genetic selection of disease resistance in dairy cattle, and could help to promote balanced breeding in Chinese dairy cattle population.

Milk production and longevity are important traits in dairy cattle breeding. In recent years, the high milk yield has been pursued in dairy cattle breeding, while longevity has been declining. With the development of balanced breeding, the longevity traits were gradually included into the comprehensive selection index of dairy cows in developed countries. The purpose of this study was to explore the relationship between milk yield and longevity of dairy cows, and to provide the reference for Xinjiang region to develop the comprehensive selection index of Holstein dairy cows in China, so as to achieve balanced breeding and high-quality healthy development of dairy industry.

The production records of three large-scale dairy farms in Xinjiang from 1997 to 2020 were collected, and the herd life and production life were calculated. The longevity records of 7 206 dairy cows and first parity 305 d milk yield of 15 218 first fetuses were obtained. A total of 18 183 pieces of pedigree information from three generations of individuals were collected, including 903 bulls and 20 883 cows. Firstly, the GLM process of SAS 9.2 software was used to analyze the effects of field, birth year, birth season and age at first calving on the herd days, productive life and first parity 305 d milk yield of Chinese Holstein cattle in Xinjiang region, and the least square mean values of the herd days, productive life and the first parity 305 d milk were calculated. SPSS 19.0 software was used to calculate the Pearson correlation coefficient between the number of herd days, production life and first parity 305 d milk yield of Chinese Holstein cattle in Xinjiang. Secondly, the genetic variance and covariance of herd days, productive life and first parity 305 d milk yield of Holstein dairy cows in Xinjiang were estimated by using AI-REML in DMU software combined with EM algorithm and multi-trait animal model, and the genetic correlation between herd days, productive life and first parity 305 d milk yield was calculated. Finally, the breeding values of the herd days, productive life and first parity 305 d milk yield of Chinese Holstein cattle in Xinjiang region were estimated by animal model BLUP method, and the genetic trend chart was drawn.

The results of basic statistical analysis on the days in the herd days, productive life and first parity 305 d milk yield of Chinese Holstein cows in Xinjiang showed that the average herd days of Chinese Holstein cows in Xinjiang was 1 754.7 days, the average production life was 937.33 days, and the average first parity 305 d milk yield was 9 362.94 kg. GLM program analysis showed that the effects of different fields, birth year, birth season and age at first calving had significant effects on herd days, productive life and first parity 305 d milk yield (P<0.01). The heritability of herd days, productive life and first parity 305 d milk yield of Chinese Holstein cattle in Xinjiang was 0.11 (0.03), 0.11 (0.03) and 0.33 (0.03), respectively. The genetic and phenotypic correlations between herd days span and productive life span were 0.99 and 0.98, respectively. There was a positive correlation between first parity 305 d milk yield and the phenotypes of herd days and production life, with correlation coefficients of 0.079 and 0.077, respectively, while genetic correlation was negative, with correlation coefficients of -0.18 and -0.20, respectively. The genetic trend chart of the herd days, the productive life and the first parity 305 d milk yield of Holstein in Xinjiang showed that the breeding values of the herd days and the productive life changed greatly with the increase of the number of cattle tested, and the overall trend decreased. The first parity 305 d milk yield of Holstein cattle in the first foetus also changed greatly with the increase of the number of cattle tested, but showed an overall upward trend.

Therefore, the longevity of dairy cows should be considered while improving milk yield, so as to avoid the decline of longevity of dairy cows with the increase of milk yield.

The study was to explore population characteristics and influencing factors of heat index and activity peak based on cattle collar data and system and to estimate its genetic parameters, and to identify the genetic markers related to the estrus indicators of cow, the genetic analysis and GWAS were performed for heat index and activity peak, so as to provide useful information for improving dairy cow fertility by formulating breeding strategies from the perspective of genetic breeding.

The heat index and activity peak records of 2 074 milking cows from a large-scale dairy farm in Beijing area were collected from July 2017 to May 2020. The MIXED model in SAS 9.4 was used to analyze the impacts of various factors on heat index and activity peak in Holstein cattle, and the DMUAI module of DMU software was used to estimate the heritability and genetic correlations of heat index and activity peak. The GWAS for heat index and activity peak was performed based on De-regressed Proof (DRP) using FarmCPU software.

In Chinese Holstein cattle, the average heat index was (71.77±19.13) au, and the activity peak was (105.94±30.73) au/2 h. The heritability of heat index and activity peak was 0.04±0.01 and 0.19±0.04, respectively. The genetic correlation between heat index and activity peak was 0.45±0.03. The approximate heat index between the age at first calving in heifer, interval from first to last insemination, and interval from first to last insemination was 0.37, -0.41, and -0.55, respectively. The GWAS for heat index and activity peak identified seven significant single nucleotide polymorphism loci (SNP) at the genome level, and 31 genes nearby the significant SNP (within 300 kb) were found.

The heat index and activity peak obtained by collar system were heritable, and there was a moderate genetic correlation between them. The activity peak was of low heritability and could reflect the heat index of cows to some extent. The activity peak and heat index provided a new direction for studying and breeding the estrus behavior of cows and improving the fecundity of cows. The function of PTGS1 (Prostaglandin-Endoperoxide Synthase1) and NDUFA8 (Ubiquinone Oxidoreductase Subunit A8) genes identified by GWAS was related to the total number of donor embryos, and the function of VEGFA (Vascular Endothelial Growth Factor A) gene was related to the reproduction of cows.