Machine Learning to Predict Pregnancy in Dairy Cows: An Approach Integrating Automated Activity Monitoring and On-Farm Data.

Automated activity monitoring (AAM) systems are critical in the dairy industry for detecting estrus and optimizing the timing of artificial insemination (AI), thus enhancing pregnancy success rates in cows. This study developed a predictive model to improve pregnancy success by integrating AAM data with cow-specific and environmental factors. Utilizing data from 1,054 cows, this study compared the pregnancy outcomes between two AI timings-8 or 10 h post-AAM alarm. Variables such as age, parity, body condition, locomotion, and vaginal discharge scores, peripartum diseases, the breeding program, the bull used for AI, milk production at the time of AI, and environmental conditions (season, relative humidity, and temperature-humidity index) were considered alongside the AAM data on rumination, activity, and estrus intensity. Six predictive models were assessed to determine their efficacy in predicting pregnancy success: logistic regression, Bagged AdaBoost algorithm, linear discriminant, random forest, support vector machine, and Bagged Classification Tree. Integrating the on-farm data with AAM significantly enhanced the pregnancy prediction accuracy at AI compared to using AAM data alone. The random forest models showed a superior performance, with the highest Kappa statistic and lowest false positive rates. The linear discriminant and logistic regression models demonstrated the best accuracy, minimal false negatives, and the highest area under the curve. These findings suggest that combining on-farm and AAM data can significantly improve reproductive management in the dairy industry.

Selection signatures for heat tolerance in Brazilian horse breeds.

Since domestication, horse breeds have adapted to their environments and differentiated from one another. This paper uses two methods to detect selection signatures in 23 horse breeds, eight of which are Brazilian (610 animals), both cold-blooded and warm-blooded, from temperate and tropical regions. These animals were genotyped using the GGP Equine BeadChip and we analysed the data by Principal Component Analysis (PCA). The samples were separated into groups based on their geographical area of origin and PCA results studied. The genomic regions under selection were detected by hapFLK and PCAdapt methodologies, identifying six regions under selection with at least one Brazilian horse breed. These regions contain genes associated with heat tolerance, skin colour, body size, energy production/metabolism, genes involved in protein degradation/turnover/DNA repair, genes reducing the impact of oxidative stress/cellular repair, and transcriptional regulation. This work confirmed LCORL and NCAPG gene regions in previous studies associated with body size on Equine Chromosome Autosome 3 (ECA3). On the same ECA3, a region implicating genes linked to coat colour was identified, also previously related to heat stress. Regions with genes coding heat shock proteins were found on ECA1 and 2, and many candidate genes for oxidation-reduction which are a natural response to heat stress. However, a larger sample size and whole-genome SNPs are needed to understand better and identify new candidate regions as well as their functional relation with heat tolerance.

Fatty acid profile and carcass characteristics in castrated and uncastrated hair lambs.

The aim of this study was to investigate the effects of castration on carcass characteristics and fatty acid profile of visceral fat and meat from lambs. Eighteen six-month-old Santa Inês male lambs (18.9 ± 2.4 kg of body weight (BW)) were used. Animals were assigned to two treatments according to their sexual condition: uncastrated (U) or castrated (C). During a 98-day experimental period, animals were kept on an Andropogon gayanus grass pasture area of 1 ha and supplemented with 200 g/animal/day of concentrate mixture. Water and mineral salt were available ad libitum. The lambs were weighed fortnightly, and at the last day of the trial, they were slaughtered for evaluation of carcass characteristics and fatty acids profile of perirenal fat and longissimus lumborum muscle samples. Castrated lambs showed higher BW than U during most part of the experiment (p < 0.05). Fat deposition was higher in C lambs as evidenced by their increased carcass fat cover. Meat from U lambs showed lower content of C16:0 and higher polyunsaturated fatty acids (PUFA) (p < 0.05) than that from C (U, 14.3 and C, 10.5%). Conjugated linoleic acid (CLA) content was not affected by castration (p > 0.05) (U, 0.74 and C, 0.76%). The cis-9, cis-12 C18:2n-6 (U, 10.6 and C, 6.86%) fatty acid and the PUFA:SFA (saturated fatty acids) ratio (U, 0.36 and C, 0.25) were higher in the muscle of U lambs (p < 0.05), indicating that the meat from U animals may provide more benefits to human health than that of C.

Selection methods for resistance to and tolerance of helminths in livestock.

Helminthiases are among the most important livestock diseases worldwide, in particular for small ruminants, which are the focus of this review. Resource Allocation Theory implies that high-productivity farm animals proportionate insufficient resources for adequate coping with stressful conditions. Significant differences between breeds and within breeds are seen, as well as genotype vs. environment interactions. With improvement of genetic host resistance to infection, transmission of infection will be impacted. On the other hand, genetic improvement of resilience can lead to a reduction in clinical signs of disease, but not necessarily reduce transmission of infection to other animals. Faecal egg count (FEC) is the main measurement used to evaluate helminthiasis load, despite the fact that the protocols and analytical methods can affect the results, and the FEC data frequently shows aggregative, negative skewed distribution, and a high coefficient of variation. Mass selection where heritability is generally medium to low generally produces slow results and low economic returns. Many studies have been published linking resistance to nematodes in livestock to Quantitative Trait Loci and most studies have concentrated on chromosomes where the major histocompatibility complex region is located. Nevertheless, these complex traits have been seen to be affected by thousands of variants that each has a small effect. More recent studies have shown that genome-wide selection strategies can be useful in selecting animals for improved production and resistance traits in this case.

Production traits in F1 and F2 crosses with naturalized hair breed Santa Inês ewes.

The once bred ewe slaughter method proposes the use of female lamb to produce a lamb and then both are slaughtered, increasing income and high quality meat production. Thus, this study evaluated the growth and reproduction performance of ewe lamb from Santa Inês (SI), a naturalized genetic resource, and their crosses (Dorper x Santa Inês (DOR), Texel x Santa Inês (TEX), Ile de France x Santa Inês (ILE)), as well as the survivability and development of their offspring. The animals were weighed monthly from birth to 12-months age. Samples of milk were collected on approximately 30 days of lactation. The physical-chemical analysis of milk was performed. SI females (2.94 kg) had significantly lower birth weight than DOR (3.80 kg) and TEX (3.87 kg). ILE females had higher weaning weight and weight at 12 months than SI females, which reflected in higher daily weight gain (ADG) (108.46 g/day) than TEX and SI. The pregnancy rates at 12 months were ILE (57.14%), TEX (25%), DOR (50%), and SI (28.57%), with TEX and SI differing of ILE and DOR (p = 0.03). Therefore, in semi-confinement and in a once-bred ewe production system using crossbreeding and allying meat production and reproduction, we recommend the use of Dorper and Ile de France breeds for crossbreeding with Santa Inês females. These results demonstrated the useful of a local genetic resource in productive system aiming a low cost meat production.

Thermographic evaluation of climatic conditions on lambs from different genetic groups.

In production systems the characterization of genetic resources in relation to their capacity to respond to environmental conditions is necessary. The objective of this study was to evaluate the use of infrared thermography for separation of animals from different genetic groups and determine which phenotypic traits are important for climatic adaptation. A total of 126 suckling lambs from four different genetic groups (Santa Inês--SI, Bergamasca--B, Bergamasca X Santa Inês--BS, and Ile de France X Santa Inês--IL) were used. The animals were divided into two groups, one housed and another in an outside paddock. Thermograph photographs were taken at four-hour intervals over three full days. Temperatures of the nose, skull, neck, fore and rear flanks and rump were measured, as well as coat depth, the density and length of hairs, reflectance and color. The daily temperature range during the experimental period was more than 20°C, with animals experiencing heat (12 h to 15 h) and cold (24 h to 4 h) stress. The three main phenotypic traits that influenced genetic group separation were hair density, height of coat, and length of hairs. Thermograph temperatures were able to detect different responses of the genetic groups to the environment. Therefore, infrared thermography is a promising technique to evaluate the response of animals to the environment and to differentiate between genetic groups.

Mastitis detection in sheep by infrared thermography.

This study aims to evaluate the use of an infrared thermograph for mastitis diagnosis in sheep. Thirty-seven Santa Inês ewes were evaluated weekly through infrared images obtained with thermograph FLIR System Series-i®. Milk was collected for somatic cell count and milk compound level determination. The clinical mastitis group had the highest fat and protein level, as well as the lowest lactose level. The udder temperatures were higher for subclinical mastitis group. The udder temperature data was able to correctly classify the animals into the mastitis groups and the canonical analysis showed that these temperatures clearly differentiated the subclinical mastitis groups from the others. Therefore, this study showed that udder infrared temperatures can be used as diagnostic method to mastitis in sheep.