Perinatal biology encompasses the coordinated sequence of parental, placental, fetal, and neonatal processes that collectively determine reproductive success and form the basis for lifelong health of the offspring. Recent advances in molecular biology, reproductive technologies, and systems biology have transformed our understanding of how events occurring before conception and throughout gestation influence pregnancy outcomes across mammalian species. The invited reviews that comprise this Perinatal Biology collection highlight emerging concepts in assisted reproductive technologies, maternal endocrine patterns, placental biology, infectious disease, and gamete physiology. Collectively, these articles demonstrate how successful pregnancy depends upon dynamic communication among maternal and paternal, placental, and fetal compartments, as none of these functions independently. Across species, common biological themes emerge around mitochondrial function and metabolic adaptability, immune and endocrine regulation, angiogenesis, and cellular plasticity. The collection also emphasizes the importance of comparative animal models, multi-omics technologies, and integrative physiology for identifying conserved mechanisms that underlie pregnancy outcomes. Together, this collection illustrates the expanding scope of perinatal biology and establishes a framework for translational One-Health research that benefits human medicine and animal agriculture.

Mortality is one of the most impactful and costly outcomes in feedlot cattle production, yet quantification of the economic value of interventions to reduce it across differing cattle populations and market conditions has been limited. The primary research objective was to characterize distributions of per-animal breakeven costs for a hypothetical mortality-reducing intervention in feedlot cattle using stochastic simulation, evaluated across sexes, arrival weight classes (approximately 249, 340, and 431 kg; 5wt, 7wt, and 9wt), current (baseline) mortality levels (0.5, 1.5, 3, 7%), and intervention effectiveness levels (5, 10, 25, 50% proportional mortality reduction). Breakevens represent maximum per-animal intervention costs that achieve equivalent expected net returns to current management. A secondary objective was to evaluate the relative importance of input variables on breakeven variation. A partial budget framework was used, with animal performance and pricing variables simulated from correlated multivariate distributions, parameterized from commercial feedlot data and industry price reports. Discrete variables (interest rate, mortality timing, and feed intake of mortalities) also were evaluated. Long-run (LR) and short-run (SR) models were constructed using pricing data from 2015 to 2025, and 2025 only, respectively. Breakeven costs scaled heavily with current mortality prevalence and intervention effectiveness; median values ranged from $0.47/animal (7wt steer, 0.5% mortality, 5% reduction, LR) to $65.65/animal (7wt steer, 7% mortality, 50% reduction, LR). Distributions widened as mortality reduction levels increased, reflecting greater economic uncertainty at higher intervention effectiveness. Steers had consistently higher breakeven values than heifers, approximately 6 to 10%, and heavier placement weights yielded higher breakevens. Breakeven costs in the SR model were approximately 40 to 50% higher than LR, reflecting historically strong 2025 cattle markets. Fed cattle price was the dominant driver of breakeven variation within primary factors (median Spearman R² = 0.88, LR), with corn price a distant second (R² = 0.13); remaining variables were of minimal individual importance. These findings provide distributional context for evaluating the economic value of mortality-reducing interventions, enabling stakeholders to assess intervention feasibility across various cattle populations and market conditions.

Liver abscesses (LA) in feedlot cattle are associated with reduced growth performance, increased liver and offal condemnations, and carcass value losses; however, potential animal welfare impacts, including pain, remain poorly understood. Although liver parenchyma lacks pain receptors, stretching or inflammation of the surrounding Glisson's capsule is known to cause pain in humans. As the liver swells in response to LA, expansion of Glisson's capsule may activate nociceptive pathways. Our objective was to compare gene expression in Glisson's capsule from livers with (n = 6) and without LA (n = 9) to identify molecular signatures associated with LA. RNA was extracted from Glisson's capsule tissue, sequenced on an Illumina platform, aligned to the Bos taurus reference genome using HISAT2, and quantified with featureCounts. Differential gene expression was assessed using DESeq2 (FDR ≤ 0.05; |log2 FC| ≥ 1). Sixty genes were differentially expressed; 41 were upregulated and 19 were downregulated in ABS tissue. FOS (+1.6 log2FC) and FOSB (+2.4 log2FC), molecular markers associated with acute and chronic pain, respectively, were upregulated in animals with LA. ENSBTAG00000061094, the bovine ortholog of human MRP4, had the greatest differential expression (-4.5 log2FC) and was downregulated in animals with LA; MRP4 is involved in inflammatory mediator transport and altered inflammatory pain thresholds. SNCA (-2.1 log2FC) has been associated with pain hypersensitivity and was downregulated in animals with LA, whereas ISG12(B) (+3.3 log2FC) indicated heightened innate immune and mitochondrial apoptotic activity and was upregulated in animals with LA. Increased EPHB3 (+1.8 log2FC) was upregulated in animals with LA and is consistent with signaling implicated in inflammatory remodeling and peripheral neuroimmune pain pathways. This study provides the first characterization of gene expression changes in Glisson's capsule associated with bovine LA. Differentially expressed genes related to inflammation, nociceptive signaling, and tissue stress suggest that LA create an inflamed and likely pain-relevant molecular environment within the liver capsule. Further research is needed to confirm whether these molecular changes correspond to perceived pain in cattle.

RGFP966 is a selective inhibitor of histone deacetylase 3 (HDAC3). This study aimed to evaluate whether HDAC3 inhibition could promote in vitro maturation (IVM) and subsequent embryo development in Tan sheep oocytes, and to compare its efficacy with luteinizing hormone (LH). Immunofluorescence analysis revealed that HDAC3 expression was significantly decreased in metaphase II (MII) oocytes compared to the germinal vesicle (GV) stage (P < 0.001). Western blotting verified that RGFP966 suppressed HDAC3 deacetylase activity in Tan sheep granulosa cells, increasing the acetylation of downstream substrates H3K9ac, H3K14ac and H4K5ac (P < 0.01). Treatment with RGFP966 significantly promoted polar body 1 (PB1) extrusion, achieving a maturation rate comparable to the LH-treated group. Mechanistically, RGFP966 treatment significantly enhanced mitochondrial distribution (P < 0.01), upregulated GPX4 expression (P < 0.001), and accordingly reduced intracellular reactive oxygen species (ROS) levels (P < 0.001). Moreover, RGFP966 significantly decreased the rate of spindle/chromosome misalignment (P < 0.001) and reduced early apoptosis (indicated by reduced Annexin-V staining) in MII oocytes (P < 0.001), without inducing additional DNA double strand breaks, as evidenced by unchanged γH2AX levels. Subsequent in vitro fertilization (IVF) demonstrated that the 2-cell cleavage and blastocyst rates of oocytes matured with RGFP966 were comparable to those of the LH group, and the resultant blastocysts exhibited normal expression patterns of lineage markers (CDX2 and SOX2). In conclusion, HDAC3 inhibitor RGFP966 promotes Tan sheep oocyte IVM and subsequent developmental competence with an efficacy equivalent to LH. This process relies on alleviating oxidative stress by enhancing mitochondrial distribution and GPX4 expression, while preserving genomic stability during meiosis. These findings provide a theoretical basis for the potential application of HDAC3 inhibitors as effective additives in ovine IVM systems.

This study aimed to investigate the effects and underlying mechanisms of growth differentiation factor 5 (GDF5) in follicular development of Hu sheep. In the present study, GDF5 was prominently localized in ovarian granulosa cells (GCs), and its expression increased progressively with follicle development. In vitro studies demonstrated that GDF5 knockdown suppressed GC proliferation and promoted apoptosis by downregulating cell cycle-related genes (PCNA, CCND1 and CDK6) and upregulating apoptosis-related gene (BAX). Meanwhile, GDF5 overexpression inhibited GC apoptosis and promoted proliferation by upregulating cell cycle-related genes and downregulating the apoptosis-related genes. RNA-seq revealed that nerve growth factor (NGF) expression was significantly downregulated following GDF5 overexpression. Subsequently, we found that the expression of NGF was significantly higher in > 5 mm follicles than in smaller follicles. Functional experiments indicated that NGF knockdown enhanced GC proliferation and repressed apoptosis by upregulating PCNA and downregulating BAX, whereas NGF overexpression exerted opposite effects. Furthermore, rescue experiments confirmed that NGF knockdown reversed the effects of GDF5 knockdown, thereby restoring GC proliferation and suppressing apoptosis, while NGF overexpression attenuated the pro-proliferation and anti-apoptosis functions of GDF5 overexpression. Altogether, these results demonstrated that GDF5 regulated GC proliferation and apoptosis via the NGF pathway, providing novel insights into the regulatory network governing follicular development.

The rapid growth of scientific literature and other unstructured materials has made agricultural decision support increasingly complex. Students, extension agents, inspectors, and private sector researchers increasingly rely on large language models (LLM). These models can produce incomplete, overly cautious, or erroneous responses. We outline how LLM offers a possible opportunity for the animal science discipline, describe what LLM do well and where they fall short, and explain how retrieval-augmented generation reduces fabrication and improves relevance. We present a case study that builds a compact index from the United States Department of Agriculture Animal and Plant Health Inspection Service (APHIS) inspection narratives for Agricultural Research Service bird facilities and show how those passages ground concise answers for nutritionists in that audience. We propose an integrated system in which frontline decision makers gain access to LLM while accessing sound information through a lens of subject-matter expertise.

Subacute ruminal acidosis (SARA) is a common metabolic disorder observed in intensive dairy production systems and is frequently associated with disruptions in rumen microbial homeostasis. In this study, a SARA model was successfully constructed by increasing the proportion of dietary concentrate and rumen-degradable starch (RDS) to examine the dynamic succession of the rumen microbiota across individuals. Forty-seven dairy goats were selected and classified into a control group (CON), low-RDS susceptible/tolerant groups (LRDSS/LRDST), and high-RDS susceptible/tolerant groups (HRDSS/HRDST) according to diet composition and individual sensitivity to SARA. Analysis of the microbial community revealed that the rumen microbiota of healthy and SARA-tolerant goats exhibited a high degree of structural similarity, whereas that of SARA-susceptible individuals underwent a pronounced compositional shift. These microbial shifts were accompanied by distinct alterations in rumen fermentation patterns, characterized by changes in volatile fatty acid (VFA) profiles, indicating functional divergence between tolerant and susceptible goats during SARA development. Notably, the relative abundance of Candidatus_Saccharimonas exhibited opposing trends between tolerant and susceptible goats and was identified as a key discriminant genus in microbial clustering. Moreover, random forest analysis revealed Candidatus_Saccharimonas as one of the most influential taxa contributing to rumen microbial differentiation, suggesting that it may serve as a potential microbial biomarker for distinguishing dairy goats in terms of SARA susceptibility or tolerance. Collectively, these findings suggest that Candidatus_Saccharimonas plays a crucial role in maintaining rumen microecological stability and enhancing the tolerance mechanisms of ruminants against SARA.

Female cattle engaging in estrous activity exhibit varying levels of higher estrous associated temperatures (H.E.A.T.) that are functionally impactful on fertility important components in the periovulatory follicle. To test this hypothesis, estrus was synchronized in virgin Angus heifers. Follicular fluid hormones and metabolites impacted by estrous activity (categorized as Estrous Active or Inactive), alone or in combination with H.E.A.T. were evaluated. Estrus was defined as the first time a heifer stood to be mounted; treatment was assigned alternately at first mount (comingled with others to remain Estrous Active, n = 13) or prevented from engaging in estrous active behaviors (Estrous Inactive, n = 12) by moving to an individual open rail pen, in sight of others and allowing nose to nose contact. Vaginal temperature was recorded using a Thermochron iButton. Periovulatory follicular fluid was aspirated ∼11.9 ± 0.1 h after first standing mount. Multiple linear regression was performed. Metabolite pathway enrichment analysis was conducted using Metaboanalyst 6.0. Estradiol levels (intrafollicular and serum) were negatively related to H.E.A.T. (P < 0.001). Intrafollicular estradiol was lower in Estrous Active heifers (P < 0.0001). Twelve follicular fluid metabolites were positively related to H.E.A.T. (P < 0.05). Pathways enriched with H.E.A.T. related metabolites (FDR < 0.1) included 'Pyrimidine metabolism'. Abundance of 25 other metabolites was lower in the periovulatory follicular fluid of Estrous Active versus Estrous Inactive heifers (P < 0.05). Pathways enriched with metabolites affected by estrous activity included 'Arginine biosynthesis' and 'Phenylalanine, tyrosine and tryptophan biosynthesis'. Six out of 12 other metabolites interactively influenced by H.E.A.T. and estrous activity are known for their antioxidant properties. In summary, periovulatory follicle progression is more advanced in Estrous Active heifers and in those exhibiting higher levels of H.E.A.T. Intrafollicular metabolites in the more advanced periovulatory follicles are linked to thermoregulation, steroidogenesis, antioxidant defense and energy metabolism. Results provide new insight into pivotal changes in the periovulatory follicle when estrogen dominance is coming to an end, the initial differentiative shift in granulosa cells is occurring, along with other preparatory changes that may be important for eventual ovulation and meiotic progression of the oocyte contained within.

In many areas of animal science, the questions are dominated by dynamic problems: outcomes evolve over time, adapt to interventions, and emerge from interacting biological, managerial, and institutional processes operating across temporal and spatial scales. These systems exhibit accumulation, feedback, nonlinearities, and time delays that can make intuitive reasoning unreliable, and that can cause well-intended policies to underperform or generate unanticipated consequences. Dynamic modeling and simulation provide a rigorous way to make such mechanisms explicit and to test whether observed patterns of behavior can arise endogenously from system structure. This paper positions system dynamics modeling as a foundational approach for feedback-driven explanation and policy design in animal systems. It also highlights agent-based modeling as a natural approach to capture heterogeneity, networks, and localized interactions in system behavior. The paper describes how artificial intelligence can strengthen dynamic modeling workflows by scaling evidence synthesis, supporting transparent and auditable structure development, and expanding feasible calibration and updating methods under uncertainty. Finally, the paper describes how high-performance computing enables ensemble-based inference and robust policy evaluation by making large-scale sensitivity analysis, uncertainty quantification, calibration, and scenario development and testing routine. Across these themes, the paper emphasizes methodological discipline: artificial intelligence and high-performance computing should augment-not replace-critical thinking, causal reasoning, rigorous structure development, thoughtful analysis, and transparent reporting. The combined use of dynamic modeling, artificial intelligence, and high-performance computing can help animal science build interpretable, testable, and updateable models that support learning and timely decision making in complex systems.

L-glutamate (Glu) and L-aspartate (Asp) are functional amino acids that play multifaceted roles in mammalian enterocytes and serve as key energy substrates for intestinal epithelial cells, suggesting potential roles in supporting intestinal resilience and disease resistance in weaned piglets. This study investigated their effects on growth performance, diarrhea incidence, intestinal integrity, and amino acid digestibility in weaned pigs challenged with F18 enterotoxigenic Escherichia coli (ETEC). Weaned piglets (7.59 ± 1.11 kg BW) were randomly assigned to five treatments (n=13) including control (CON), 1% Glu supplementation (GLU), 1% Asp supplementation (ASP), 1% Glu and 1% Asp combination supplementation (GAC), and an antibiotic control (ABX). All piglets were orally challenged with F18 ETEC. The study was conducted from d 0 to 28. In the present study, the differences were observed in pigs fed GLU, ASP, or GAC compared with CON-fed pigs. GLU, ASP, or GAC supplementation alleviated (P < 0.05) the impact of ETEC infection by enhancing (P < 0.05) overall growth performance and feed efficiency, with effects comparable to ABX-fed pigs. GLU or ASP supplementation also reduced (P < 0.05) diarrhea frequency, similar to ABX supplementation. Pigs fed ASP, GAC, or ABX had lower (P < 0.05) fecal β-hemolytic coliform shedding than CON-fed pigs on d 14. ASP supplementation tended to increase (P = 0.072) duodenal villus height on d 12, whereas other intestinal morphological responses varied among treatments, intestinal segments, and time points. On d 28, ASP or GAC supplementation enhanced (P < 0.05) ileal amino acid digestibility of selected amino acids, with some responses comparable to those observed in ABX-fed pigs. At the molecular level, responses in tight junction- and mucin-related gene expression were modest and varied among treatments and time points. Overall, Glu, Asp, or their combination enhanced growth performance, while Asp and the combined treatment improved selected measures of amino acid digestibility, and individual Glu or Asp reduced diarrhea. These findings support Glu and Asp as promising nutritional strategies to support growth and nutrient utilization during enteric challenge, with potential to reduce reliance on in-feed antibiotics.

Mathematical models have long supported nutrition management in livestock systems, progressing from early National Research Council empirical tables to complex mechanistic frameworks such as the Ruminant Nutrition System. What is less documented is how models are operationalized in commercial practice, particularly as data streams from precision livestock technologies, remote sensing, and other Internet of Things technologies are increasingly hybridized into existing modeling frameworks. A prevailing assumption is that greater sensing, automation, or algorithmic complexity will naturally translate into better decisions; however, adoption and impact in commercial systems remain uneven. To examine how models are actually used under real-world constraints, we conducted a literature review and semi-structured interviews (n = 11) with livestock producers, technology developers, consultants, and value-chain actors working at the interface of biology, data, models, and farm decision-making. Six cross-cutting themes emerged: 1) model augmentation of systems, 2) model predictive control (MPC), 3) human centered adoption, 4) technology fit, 5) data bottlenecks, and 6) data enlightened value chain (Figure 1). Collectively, these insights re-center nutrition as the integrating discipline for productivity, welfare, and sustainability, and emphasize MPC and open, interoperable data pipelines as prerequisites for moving beyond pilots toward durable adoption. Interviewees consistently emphasized that meaningful impact depends on aligning biological realism, software engineering, and value-chain incentives so that models support actionable decisions under commercial constraints, rather than pursuing automation for its own sake.

A pig's ability to convert consumed feed into saleable meat is an important component of production efficiency in the swine industry. Given the polygenic and environmentally influenced nature of feed efficiency and body composition traits, conventional selection approaches have limited ability to pinpoint causal genetic factors driving variation. In this study, we performed a single-step genome-wide association study, with whole-genome sequence data (imputed from 50 to 60K SNP arrays), for feed efficiency and body composition traits using a multi-generational crossbred swine population. Overall, the three-generation pedigree consisted of 14,242 pigs, of which 9,336 pigs were phenotyped and 7,339 were genotyped. Whole-genome sequence variants were clustered into 705,923 linkage disequilibrium (LD) blocks, ranging from 0.001 to 9981.80 kb in size. A total of 133 LD blocks were significantly associated with the studied traits, several with potentially large pleiotropic effects. A total of 453 candidate genes were identified in the genomic regions of significant LD blocks. Four functional candidate genes (SIRT2, RYR1, MC4R, and LEPR) were found to harbor moderate to high impact variants, expected to affect protein function.

Precision livestock farming (PLF) increasingly relies on computer vision (CV) to monitor animal behavior, health, and production. However, widespread adoption remains limited by the large quantities of manually labeled data required to train most artificial intelligence (AI) models. This paper reviews AI and CV applications in PLF and presents three data-efficient case studies conducted at a research feedlot at Texas A&M University. The assessment of AI model predictability was conducted with precision, recall, F1-score, and adjusted rand index (ARI). In Case Study 1, a data-efficient Mask R-CNN model pre-trained on the COCO dataset was applied without modification to detect cattle feeding events at individual feed bunks. Across 98,720 frame-bunk classifications, the system achieved 87.2% precision, 89.1% recall, and an 88.2% F1 score compared to physical bunk sensor ground truth. In Case Study 2, transfer learning on a ResNet backbone was used to classify bunk feed grade across six customized levels (S00-S40) using only 183 labelled images. The model achieved 91% accuracy on an independent verification set of 7,905 images. In Case Study 3, statistical tools, i.e., Density-Based Spatial Clustering of Applications with Noise (DBSCAN) and k-means clustering, were applied to identify individual animals at the bunk without manual animal-level annotation. This method and achieved an ARI of 0.96 against expert-verified ground truth, successfully differentiating animals. Integrating all three components, the framework produced a continuous record of individual animal bunk visits and estimated feed intake from a one-hour video. Key remaining challenges include long-term individual animal re-identification under changing appearance and environmental conditions, incorporation of spatiotemporal information to better characterize feeding behavior, and the limited availability of large-scale open datasets for PLF research. Future research will integrate multispectral imaging and audio analytics to improve model robustness and advance individual animal identification for cattle health and nutritional management.

Bayesian statistical modeling provides a principled framework for integrating biological knowledge, process understanding, and observed data in animal science research. By treating all unobserved quantities as random variables and explicitly defining probability distributions for both processes and observations, Bayesian methods allow researchers to quantify uncertainty, incorporate prior information, and evaluate full distributions of plausible outcomes. This manuscript introduces the foundational concepts of Bayes' theorem, probability, probability distributions, and Markov chain Monte Carlo, and demonstrates how these components form coherent, process-based statistical models. Two case studies illustrate the application of Bayesian regression to questions central to beef cattle research: the influence of day-to-day variation in dry matter intake on average daily gain, and the prediction of intake from biometric variables collected using precision livestock technologies. Together, these examples highlight how Bayesian approaches enhance model transparency, improve interpretation, and support decision-making in complex biological systems. This framework equips animal scientists with the tools needed to ask clearer questions, build biologically grounded models, and leverage emerging data streams to advance research and management.

First egg weight is an economically important trait in indigenous chickens, but its molecular regulation remains unclear. We tested whether a missense variant in quiescin Q6 sulfhydryl oxidase 1 (QSOX1; g.8_5992660 T > C, p. Ser291Gly) was associated with egg traits and whether QSOX1 influenced oviduct epithelial function. Among 450 Huainan Partridge hens, 375 were TT and 75 were TC; the C-allele frequency was 0.08 and the population conformed to Hardy-Weinberg equilibrium (P = 0.16). TC hens had heavier first eggs, greater albumen and thick-albumen weights, and greater eggshell weight than TT hens (P < 0.05). QSOX1 expression was greatest in the isthmus and was higher in TC than TT hens at the mRNA, protein, and immunohistochemical levels (P < 0.05). In primary isthmus epithelial cells, QSOX1 overexpression increased CCK-8 absorbance at 24 h, altered viability-associated transcripts, and reduced apoptosis, whereas QSOX1 knockdown produced the opposite responses. In a transwell co-culture model, QSOX1 overexpression in isthmus cells increased ovalbumin protein accumulation and upregulated COL6A1 and SEC31A expression. These findings support the hypothesis that QSOX1 contributes to first egg weight by maintaining isthmus epithelial viability and favoring egg-white protein accumulation.