Computed tomography (CT) images provide fast and accurate non-invasive measurements of anatomy, which are crucial in pig breeding. In the commercial breeding program of Topigs Norsvin, the CT-images are mainly used for estimation of carcass value and for scoring the severity of osteochondrosis lesions in joints. This study presents the first major step towards automated detection of osteochondrosis, using CT images. This involves an anatomic segmentation model that can segment 29 classes of different tissues, like individual bones, muscles, and organs. The algorithm then identifies major joints in the fore- and hindlimbs by detecting the center points of the joints, and this method is validated against manually labeled data. Average distance between labeled and predicted center points was 29 mm (SD = 13mm). The next step will utilize these center points to create bounding boxes for local segmentation models to focus on relevant subsets of voxels, enhancing the detection of lesions in joints. This approach aims to improve the efficiency and accuracy of osteochondrosis quantification in CT scan images of pigs, ultimately -benefiting pig breeding programs by providing detailed and automated phenotypes.

Genetic progress in litter size in pigs have significantly increased in recent decades, resulting in a corresponding rise in farrowing duration (FRD). Prolonged FRD is negatively correlated with stillbirth rates and the colostrum intake of liveborn piglets. However, measuring FRD is time-consuming and, as a result, often unavailable. Consequently, studies exploring the genetic aspects of FRD in pigs remain scarce. The objectives of this study were to estimate the genetic parameters and prediction accuracy for FRD using phenotypic and genotypic data from a large commercial farm. Data were collected from 18,180 litters of 4,967 purebred Large White sows on a single farm in Brazil, between 2020 and 2024, as part of the farm’s routine data collection protocol. The time of birth was recorded at the piglet level, and FRD was recorded at the sow level, defined as the time between the birth of the first and last piglet in a given litter. All sows were genotyped using a custom Illumina 25K SNP chip. Genetic parameters were estimated using an additive linear model with ASReml v3. The model included fixed effects for population mean, litter type (purebred/crossbred), farrowing room, year-week, sow parity, average birth weight of piglets, and litter size. Random effects for permanent environment, additive genetic, and residual effects were also included. Gilts that had their first litter in 2024 were used as the validation set for prediction accuracy analysis. The correlation between their estimated breeding values using the entire dataset (“true breeding value”) during the genetic parameter analyses and their predicted genomic breeding values using only phenotypes collected before 2024 (“predicted breeding values”) was defined as prediction accuracy. The average FRD and litter size for sows in this dataset were 216±110 minutes and 15.0±2.7 piglets (total born), respectively. Results indicated that the population mean, year-week, sow parity, and litter size all significantly affected FRD (P < 0.001). The heritability of FRD was estimated at 0.07±0.01, with the permanent environment effect accounting for 0.03±0.01 of the total phenotypic variance. A clear, linear relationship between litter size and FRD was also observed. The prediction accuracy for FRD based on this dataset was 0.64. In conclusion, this study is the first to investigate the prediction accuracy of FRD in pigs using genomic data. Our findings suggest that while the heritability of FRD is low, the trait is heritable and can be genetically predicted, offering potential for improving FRD through genetic selection.

Pig breeding companies collect data on body weight and feed intake each time boars visit individual automatic feeding stations at boar testing farms. Proper modeling of feed intake and body weight gain is essential for achieving significant genetic progress in feed efficiency and growth. Sigmoidal curves, such as the Gompertz curve, are widely used to model pig growth because they capture the initial rapid exponential growth of pigs until the inflection point. After this point, growth slows until the pig reaches the adult weight at the curve’s plateau. Fitting Gompertz curves also generate, based on the inflection point, an indirect prediction of the onset of puberty (OP). Based on these growth curves, the age at the onset of puberty (AOP) of boars and gilts has been previously determined to be around 5-6 months. The weight at the onset of puberty (WOP) will present significant variation especially depending on the pig line. A Bayesian hierarchical model was previously developed to integrate a whole-genome prediction model using a 25K SNP panel and a pig growth model that fits a Gompertz curve, incorporating a recursive relationship between daily body weight and daily feed intake records. This model allows the prediction of breeding values for the parameters of the growth model. We aimed to apply this model to predict OP-related traits and validate these predictions with reproduction traits. The model was run using data from 25,686 pigs from four commercial lines (Duroc, Landrace, Large-White, and a Synthetic Line). The AOP ranged between 110 and 210 days across different lines, with an average ranging between 146 and 165 days (SD = 8-10 days). The WOP ranged between 75 and 190 kg, with an average ranging between 91 and 134 kilograms (SD = 7-11 kg). The OP in boars is positively correlated, though only low to moderately, with male reproductive performance. Correlations between the OP ranged from 0.13-0.30, 0.15-0.21, and 0.16-0.34 with boars’ trainability (training time until first jump), the age at the first ejaculate collection, and semen quality (sperm motility), respectively, indicating a link between the boar rearing phase and the boar reproductive performance. We are running the model with data (genotypes and phenotypes) from the boar testing farms (training population) to get effect estimates of SNPs on the OP. Based on the SNP effects, we will predict the OP in target populations (with only genotypes) of purebred boars and gilts, and also crossbred gilts. The OP is a trait that holds great potential for pig breeding programs, including the reproduction management of boars and gilts.

Characterization of essential genes across the genome is fundamental to understanding cellular functions at a molecular level. While significant progress has been made in characterizing essential genes in human and mouse models, relatively little is known about essential genes in the porcine genome. Pigs are an important production species and are now emerging as valuable models for studying human diseases due to their physiological similarities to humans. To map essential genes across the porcine genome, we have developed a novel porcine genome-wide CRISPR knockout screening library (pGeCKO) and applied it to two porcine cell lines, PK15 and IPEC-J2. We identified 2,245 essential genes in PK15 cells and 919 essential genes in IPEC-J2 cells, with 683 of these shared between both cell lines. Functional analyses revealed that most essential genes are involved in core cellular processes such as cell cycle regulation, DNA replication, transcription, and translation. Comparative analysis with human essential genes from the DepMap project revealed that over half of the genes are shared with humans and the rest are porcine-specific. These porcine-specific essential genes included genes in core functional pathways related to protein and RNA processing as well as many related to N-glycan biosynthesis, signal transduction, and several long-noncoding RNAs. This work provides a new resource for leveraging porcine models in disease research, enhancing our understanding of porcine genetics and its implications for human health. ### Competing Interest Statement The authors have declared no competing interest.

Minimizing the risk of disease transmission, disseminating superior genetics, and reducing transportation costs are recognized advantages of embryo biotechnologies. These advantages make the development of a minimally invasive and repeatable procedure in pigs enticing, but simultaneously magnify the anatomical constraints. For decades, the swine industry has struggled to establish a universal procedure to collect pre-implantation embryos from pigs due to their long and convoluted uterine horns (UHs). Thus, the objectives were to evaluate the benefits of employing a transitional surgical model by shortening UH tissue using a 40 cm ipsilateral resection and assess the compensatory ovulatory response following an ovariectomy. The surgery was deemed successful as the UH was resected and the contralateral UH was fully ligated. The dam- and sire-line gilts exhibited ovarian hypertrophy between surgery and slaughter on the remaining ovary, illustrated by an increase in the number of corpora lutea (13.4 and 3.0 vs. 27.2 and 12; p < 0.05, respectively) and intact ovary weight (11.9 and 7.7 vs. 25.9 vs. 38.7 g; p < 0.05, respectively). This research is a vital step in assessing whether this interim surgical approach serves as a valuable method to advance the development of non-surgical techniques to collect pre-implantation embryos in pigs.

The aim of this study was to investigate the reference population size required to obtain substantial prediction accuracy within- and across-lines and the effect of using a multi-line reference population for genomic predictions of maternal traits in pigs. The data consisted of two nucleus pig populations, one pure-bred Landrace (L) and one Synthetic (S) Yorkshire/Large White line. All animals were genotyped with up to 30 K animals in each line, and all had records on maternal traits. Prediction accuracy was tested with three different marker data sets: High-density SNP (HD), whole genome sequence (WGS), and markers derived from WGS based on pig combined annotation dependent depletion-score (pCADD). Also, two different genomic prediction methods (GBLUP and Bayes GC) were compared for four maternal traits; total number piglets born (TNB), total number of stillborn piglets (STB), Shoulder Lesion Score and Body Condition Score. The main results from this study showed that a reference population of 3 K-6 K animals for within-line prediction generally was sufficient to achieve high prediction accuracy. However, when the number of animals in the reference population was increased to 30 K, the prediction accuracy significantly increased for the traits TNB and STB. For multi-line prediction accuracy, the accuracy was most dependent on the number of within-line animals in the reference data. The S-line provided a generally higher prediction accuracy compared to the L-line. Using pCADD scores to reduce the number of markers from WGS data in combination with the GBLUP method generally reduced prediction accuracies relative to GBLUP using HD genotypes. The BayesGC method benefited from a large reference population and was less dependent on the different genotype marker datasets to achieve a high prediction accuracy.

Preimplantation genetic testing for aneuploidy (PGT-A) is widespread, but controversial, in humans and improves pregnancy and live birth rates in cattle. In pigs, it presents a possible solution to improve in vitro embryo production (IVP), however, the incidence and origin of chromosomal errors remains under-explored. To address this, we used single nucleotide polymorphism (SNP)-based PGT-A algorithms in 101 in vivo-derived (IVD) and 64 IVP porcine embryos. More errors were observed in IVP vs. IVD blastocysts (79.7% vs. 13.6% p < 0.001). In IVD embryos, fewer errors were found at blastocyst stage compared to cleavage (4-cell) stage (13.6% vs. 40%, p = 0.056). One androgenetic and two parthenogenetic embryos were also identified. Triploidy was the most common error in IVD embryos (15.8%), but only observed at cleavage, not blastocyst stage, followed by whole chromosome aneuploidy (9.9%). In IVP blastocysts, 32.8% were parthenogenetic, 25.0% (hypo-)triploid, 12.5% aneuploid, and 9.4% haploid. Parthenogenetic blastocysts arose from just three out of ten sows, suggesting a possible donor effect. The high incidence of chromosomal abnormalities in general, but in IVP embryos in particular, suggests an explanation for the low success of porcine IVP. The approaches described provide a means of monitoring technical improvements and suggest future application of PGT-A might improve embryo transfer success.

Improved nutrient digestibility is an important trait in genetic improvement in pigs due to global resource scarcity, increased human population and greenhouse gas emissions from pork production. Further, poor nutrient digestibility represents a direct nutrient loss, which affects the profit of the farmer. The aim of this study was to estimate genetic parameters for apparent total tract digestibility of nitrogen (ATTDn), crude fat (ATTDCfat), dry matter (ATTDdm), and organic matter (ATTDom) and to investigate their genetic relationship to other relevant production traits in pigs. Near-infrared spectroscopy was used for prediction of total nitrogen content and crude fat content in feces. The predicted content was used to estimate apparent total tract digestibility of the different nutrients by using an indicator method, where acid insoluble ash was used as an indigestible marker. Average ATTDdm, ATTDom, ATTDn, and ATTDCfat ranged from 61% to 75.3%. Moderate heritabilities was found for all digestibility traits and ranged from 0.15 to 0.22. The genetic correlations among the digestibility traits were high (>0.8), except for ATTDCfat, which had no significant genetic correlation to the other digestibility traits. Significant genetic correlations were found between ATTDn and feed consumption between 40 and 120 kg live weight (F40120) (-0.54 ± 0.11) and ATTDdm and F40120 (-0.35 ± 0.12) and ATTDom and F40120 (-0.28 ± 0.13). No significant genetic correlations were found between digestibility traits and loin depth at 100 kg, nor backfat thickness at 100 kg (BF), except between BF and ATTDn (-0.31 ± 0.14). These results suggested that selection for improved feed efficiency through reduced feed intake within a weight interval, also has led to improved ATTDdm, ATTDom, and ATTDn. Further, the digestibility traits are heritable, but mainly related to feed intake and general function of the intestines, as opposed to allocation of feed resources to different tissues in the body.

Cumulus cells (CCs) are pivotal during oocyte development. This study aimed to identify novel marker genes for porcine oocyte quality by examining the expression of selected genes in CCs and oocytes, employing the model of oocytes from prepubertal animals being of reduced quality compared to those from adult animals. Total RNA was extracted either directly after follicle aspiration or after in vitro maturation, followed by RT-qPCR. Immature gilt CCs accumulated BBOX1 transcripts, involved in L-carnitine biosynthesis, to a 14.8-fold higher level (p < 0.05) relative to sows, while for CPT2, participating in fatty acid oxidation, the level was 0.48 (p < 0.05). While showing no differences between gilt and sow CCs after maturation, CPT2 and BBOX1 levels in oocytes were higher in gilts at both time points. The apparent delayed lipid metabolism and reduced accumulation of ALDOA and G6PD transcripts in gilt CCs after maturation, implying downregulation of glycolysis and the pentose phosphate pathway, suggest gilt cumulus–oocyte complexes have inadequate ATP stores and oxidative stress balance compared to sows at the end of maturation. Reduced expression of BBOX1 and higher expression of CPT2 in CCs before maturation and higher expression of G6PD and ALDOA after maturation are new potential markers of oocyte quality.

Commercial application of embryo transfer in pig breeding is dependent on the storage of embryos. The aim of this study was to assess the embryo quality of in vitro-produced blastocysts after 3 h liquid storage at 37°C in CO2-free medium by evaluating morphology, in vitro developmental capacity and apoptosis. Blastocysts at days 5 and 6 post-fertilization were randomly allocated to the storage group (HEPES-buffered NCSU-23 medium including bovine serum albumin in a portable embryo transport incubator at 37°C) or a control group (porcine blastocyst medium in a conventional culture incubator). Thereafter, blastocysts were evaluated for morphology and stained to assess apoptosis straight after the 3 h storage period or after a further 24 h conventional incubation. There was no significant difference between the storage and control group after 3 h storage and the further 24 h conventional incubation for any of the parameters, nor for apoptosis straight after the 3 h storage. Embryos that reached the blastocyst stage at day 5 showed less apoptosis (6.6% vs 10.9%, P = 0.01) and a trend for a higher rate of developmental capacity (70.6% vs 51.5%, P = 0.089) than embryos reaching the blastocyst stage on day 6. In conclusion, in vitro-produced porcine blastocysts can be stored for 3 h at physiological temperature in transportable incubators using a CO2-independent medium without compromising quality.

Hyperactive sperm motility is important for successful fertilization. In the present study, a proteome profiling approach was performed to identify the differences between Landrace boars with different levels of hyperactive sperm motility in liquid extended semen. Two contrasts were studied: (i) high versus low levels of sperm hyperactivity at semen collection day and (ii) high versus low change in levels of sperm hyperactivity after 96 h semen storage. Testicular samples were analyzed on a Q Exactive mass spectrometer and more than 6000 proteins were identified in the 13 samples. The most significant differentially expressed proteins were mediator complex subunit 28 (MED28), cell division cycle 37 like 1 (CDC37L1), ubiquitin specific peptidase 10 (USP10), zinc finger FYVE-type containing 26 (ZFYVE26), protein kinase C delta (PRKCD), actinin alpha 4 (ACTN4), N(alpha)-acetyltransferase 30 (NAA30), C1q domain-containing (LOC110258309) and uncharacterized LOC100512926. Of the differentially expressed proteins, 11 have previously been identified as differentially expressed at the corresponding mRNA transcript level using the same samples and contrasts. These include sphingosine kinase 1 isoform 2 (SPHK1), serine and arginine rich splicing factor 1 (SRSF1), and tubulin gamma-1 (TUBG1) which are involved in the acrosome reaction and sperm motility. A mass spectrometry approach was applied to investigate the protein profiles of boars with different levels of hyperactive sperm motility. This study identified several proteins previously shown to be involved in sperm motility and quality, but also proteins with no known function for sperm motility. Candidates that are differentially expressed on both mRNA and protein levels are especially relevant as biological markers of semen quality.

The aim of this study was to estimate genetic parameters for apparent total tract digestibility of nitrogen (ATTDn) and dry matter (ATTDdm) and to investigate their genetic relationship to other relevant production traits in pigs. Near infrared spectroscopy was used for prediction of nitrogen content in feces. The estimated heritabilities of ATTDdm and ATTDn were 0.22 and 0.20, respectively. The genetic correlations between ATTDdm and ATTDn and feed intake from 40 to 120 kg were -0.5 and -0.65, respectively. No significant genetic correlations were found between any of the digestibility traits and growth and carcass quality traits. The results suggested that the current selection strategy is improving the digestibility of nitrogen and dry matter, but that selection for improved digestibility would reduce the feed consumption from 40 to 120 kg.

Sperm motility and viability of cryopreserved semen vary between boars and straws, which influences the outcomes of in vitro embryo production (IVEP). However, progressive motility is usually not considered during IVEP. The aim of this study was to assess fertilization with a 500:1 and 250:1 'progressively motile sperm to oocyte' ratio on IVEP outcomes using semen from three Duroc and three Landrace boars. Frozen-thawed sperm was centrifuged through a 45/90% Percoll® density gradient and sperm quality parameters were assessed. In vitro matured oocytes were fertilized at the two ratios, a portion was stained 10-12 h after start of fertilization to analyze fertilization and polyspermy, while the remaining zygotes were cultured up to day 7. The 500:1 ratio resulted in a higher fertilization and blastocyst yield on day 6 compared with the 250:1 ratio, but no effect of ratio was observed for polyspermy, cleavage rate or blastocyst cell number. Individual differences between boars were observed for fertilization, cleavage and blastocyst rates, but not for the other IVEP outcomes. In conclusion, a higher fertilization and blastocyst yield was obtained with the 500:1 ratio compared with the 250:1 ratio, while polyspermy level was consistent across ratios. Differences in IVEP outcomes were still observed between the individual boars although adjusted for progressive motility. Promising blastocyst yields and high total blastocyst cell counts were obtained with sperm from both breeds.

It has been debated whether intensive selection for growth and carcass yield in pig breeding programmes can affect the size of internal organs, and thereby reduce the animal’s ability to handle stress and increase the risk of sudden deaths. To explore the respiratory and circulatory system in pigs, a deep learning based computational pipeline was built to extract the size of lungs and hearts from CT-scan images. This pipeline was applied on CT images from 11,000 boar selection candidates acquired during the last decade. Further, heart and lung volumes were analysed genetically and correlated with production traits. Both heart and lung volumes were heritable, with h 2 estimated to 0.35 and 0.34, respectively, in Landrace, and 0.28 and 0.4 in Duroc. Both volumes were positively correlated with lean meat percentage, and lung volume was negatively genetically correlated with growth ( r g = − 0.48 ± 0.07 for Landrace and r g = − 0.44 ± 0.07 for Duroc). The main findings suggest that the current pig breeding programs could, as an indirect response to selection, affect the size of hearts- and lungs. The presented methods can be used to monitor the development of internal organs in the future.

The aim of this study was to develop a strategy for selecting SNPs that could receive higher weight than other SNPs in a GBLUP approach due to their expected association with important pig phenotypes. In addition, we aimed to investigate if such a strategy yields improved prediction accuracy compared to a traditional GBLUP approach. Four prediction accuracy scenarios were evaluated using three production traits in two pig populations. Our results show that adding extra weight to SNPs that are expected to be (close to) causal variants in the G matrix increases the prediction accuracy of genomic prediction. The advantage of weighted G matrix compared to a traditional one is not very large, but the added value is consistent.

To improve production efficiency and ensure good animal welfare, novel phenotypes are developed for minimizing damaging behaviour in pigs. The aim of this study was therefore to estimate genetic parameters for two novel feeding behaviour phenotypes and investigate their genetic relationship with other traits recorded at the nucleus boar testing station. The genetic analyses were performed using multivariate animal models. The estimated heritability of interrupter (IntER) and interrupted (IntED) at the feeding station was 0.58 and 0.61, respectively, and the genetic correlation between the traits was 0.89. The genetic correlation between the feeding behaviour traits (IntER and IntED) with tail lesions and feed consumption from 40 to 120 kg live weight was not significantly different from zero. The results suggested that the feeding behaviour traits in this study could be treated as the same trait. However, as the frequency of tail lesions in this study was low, the genetic relationship between feeding behaviour and tail lesions remains unclear.

The aim of this study was to compare three methods of genomic prediction: GBLUP, BayesC and BayesGC for genomic prediction of six maternal traits in Landrace sows using a panel of 660 K SNPs. The effects of different priors for the Bayesian methods were also investigated. GBLUP does not take the genetic architecture into account as all SNPs are assumed to have equally sized effects and relies heavily on the relationships between the animals for accurate predictions. Bayesian approaches rely on both fitting SNPs that describe relationships between animals in addition to fitting single SNP effects directly. Both the relationship between the animals and single SNP effects are important for accurate predictions. Maternal traits in sows are often more difficult to record and have lower heritabilities. BayesGC was generally the method with the higher accuracy, although its accuracy was for some traits matched by that of GBLUP and for others by that of BayesC. For piglet mortality within 3 weeks, BayesGC achieved up to 9.2% higher accuracy. For many of the traits, however, the methods did not show significant differences in accuracies.

The aim of this study was to perform genome-wide association analyses for backfat thickness and osteochondrosis in Landrace pigs and to fine map pleiotropic genomic regions. In order to characterise genomic regions, phenotypic information of 5,000 animals with osteochondrosis scored from CT images and 40,000 animals with backfat thickness scored from CT or ultrasound images were analysed. All animals were genotyped with a medium density SNP chip and a subset of them were genotyped with a high-density SNP chip as well, allowing for imputation. Two genomic loci were found in common for osteochondrosis and backfat thickness, one on chromosome 5 and one on chromosome 14. For both regions, an antagonistic relationship was found. Fine mapping using an impact score approach identified the CCND2 gene as the most likely causal gene on chromosome 5, whereas a mutation in CRTAC1 had the highest impact score in the chromosome 14 region.