Skip to main content
ARS Home » Plains Area » Clay Center, Nebraska » U.S. Meat Animal Research Center » Genetics and Animal Breeding » Research » Research Project #433845

Research Project: Identifying Genomic Solutions to Improve Efficiency of Swine Production

Location: Genetics and Animal Breeding

2022 Annual Report

Objective 1: Utilize next-generation sequencing technologies to improve the contiguity of the swine genome assembly and better characterize genomic variation in pigs. Subobjective 1.A: Utilize segregation analysis to improve the porcine genome assembly. Subobjective 1.B: Develop more comprehensive gene models for the swine genome. Subobjective 1.C: Develop an electronic warehouse of genomic variants that can be utilized by the swine genomics research community. Objective 2: Develop genotyping products for commercial swine producers to increase the efficiency of swine production. Subobjective 2.A: Identify predictive genetic markers for traits associated with production efficiency in commercial swine populations. Subobjective 2.B: Develop strategies for inclusion of predictive markers in selection programs.

The principal goals of this project are to enhance our understanding of the biological processes important to swine production and provide the U.S. swine industry with genetic tools that will ensure that it remains the global leader in providing safe, nutritious, and economic pork products. The swine industry has been faced with significant challenges, many of which revolve around the production and performance of feeder pigs. The environment in which females are housed is continually evolving, and the increasing cost of feed has resulted in continuous shifts in the utilization of feed stuffs. Each new challenge requires an assessment of potential solutions. Genetic selection can be used to address many production issues. If DNA variants associated with changes in phenotype can be identified, then marker assisted selection can be implemented to expedite genetic progress. Predictive genetic markers need to be transferred to commercial entities that will rapidly evaluate and adopt them. The increasing improvements to the porcine genome, better annotations of genes from model organisms, and enhanced bioinformatics technologies provide researchers with the necessary tools to identify functional genetic variants. Objective 1 focuses on improving the porcine genome assembly and detecting polymorphisms from data generated by next-generation sequencing. Objective 2 will strive to effectively transfer the results of the research from Objective 1 to producers. Development of marker panels along with economical genotyping platforms will be essential. Our research will focus on the evaluation of genetic markers based on their predicted effects on gene products to discover causal genetic variants of phenotypic variation. This will lead to the development of marker panels and economical genotyping platforms for industry applications.

Progress Report
This is a final report for our project titled “Identifying Genomic Solutions to Improve Efficiency of Swine Production” to be replaced by 3040-31000-106-000D. In 5 years we accomplished many research goals and transferred our findings to the commercial swine industry. We discovered genetic variants in swine populations, evaluated transcriptomes of a variety of pig tissues (embryos, fetal tissues, brain, muscle, liver, milk, blood and intestines), identified important candidate genes associated with economic traits (puberty, feed intake, meat quality, vertebrae number and teat number) and continued to develop genetic tools for the U.S. swine industry to remain competitive in a global economy. We developed new methodologies to incorporate multiple ‘-omics’ data-sets into a combined analysis to improve the power to detect genetic variants altering animal performance. Research findings were released to industry and academia, and we co-developed an available genotyping platform containing 676 putative loss of function variants, as well as a dense representation of immunological genes. Overall, we made significant discoveries and plan to continue our research in the next project plan. Our progress in fiscal year (FY) 2022 was substantial. Several projects conducted have targeted development of comprehensive gene models (Objective 1B). Metabolomic and transcriptomic data were generated from peripheral blood samples on 174 and 326 animals in two separate case-control experiments evaluating sow structural soundness. Cases for this study are lame animals in the breeding herd and control animals are contemporary sows who remained sound throughout production. In the first experiment, samples were collected from mature females, RNA-Seq libraries from peripheral blood of 174 sows (87 cases and 87 controls) were paired-end sequenced on an Illumina NextSeq2000 instrument, resulting in nearly 15 billion sequence reads generated. After quality control, sequence reads were mapped to the Sscrofa 11.1 genome where an average of 91.82% reads were mapped per library. A second experiment sampled animals at 150 days of age, 326 RNA-Seq libraries were generated, and sequencing is in progress. Global transcriptome analysis of gene expression using RNA-Seq with high temporal resolution between spherical, ovoid, and tubular stage blastocysts was performed (Objective 1B). A total of 54 libraries were sequenced, with an average of 97.9 million reads per library. After RNA-Seq analysis, significant differentially expressed genes (DEG) and pathways were identified between distinct sequential development stages. Overall, 1,898 significant DEG’s were identified between spherical and ovoid morphologies and 15 DEG’s were identified between ovoid and tubular morphologies. Analyses to identify and characterize novel elements, such as long non-coding RNA, in these blastocyst transcriptomes are underway. Whole blood samples were taken at days 0 and 42 from 178 pigs monitored for individual feed intake and body weight gain. Blood samples were analyzed for blood cell parameters including white blood cell, neutrophil, lymphocyte, monocyte, eosinophil, basophil, red blood cell and platelet counts, hemoglobin, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin and mean platelet volume. Feed efficiency parameters were predicted using a model including fixed effects of farrowing group and pen and individual hematology parameters at day 0, day 42, or their change as covariates. Changes in red blood cell parameters, especially hematocrit, were found to be associated with feed efficiency measures. RNA-Seq data are being used to characterize the blood transcriptome, and transcriptomic profiles are being investigated for predictive signatures of red blood cell parameters and/or feed efficiency phenotypes (Objective 1B). We continue to collaborate with the Farm Animal Genotype-Tissue Expression (FarmGTEx) and Pig Genotype-Tissue Expression (PigGTEx) consortium. White blood cell RNA-Seq and single nucleotide polymorphism (SNP) genotype data for U.S. Meat Animal Research Center (USMARC) pigs have been processed and used to validate a transcript-wide association study (TWAS) conducted by PigGTEx researchers. Progress to impute genome-level SNP genotypes from commercial array SNP platforms was made (Objective 2A). A total of 16,895 USMARC pigs have been genotyped. Our plan was to optimize imputation protocols by evaluating eight haplotype phasing/imputation software pipelines. Through discussions with colleagues at Michigan State University, we learned that similar analyses had been conducted. Thus, we are utilizing their research results to select phasing and imputation software. The last step in optimizing our bioinformatics pipeline is to determine the optimal reference panel to be used for imputation. Researchers at the University of Edinburgh provided us access to the PigGTEx SNP reference panel, comprised of more than 42 million SNP’s from 1,600 animals, including 187 USMARC animals. We are currently testing six different methodologies for selecting animals to be included in the reference panel. In addition to the imputation mentioned, we also worked with GenCove to collect genome-level SNP genotypes from low coverage genome sequencing. We studied 192 samples of a diverse set of industry animals and an additional 256 animals that broadly represent the USMARC swine population. These genotypes have a high concordance with SNP array data (> 98%) and we are comparing them to genotypes retrieved from the Michigan State University imputation pipeline. To identify genetic markers associated with milk composition to increase production efficiency, we collected data from 160 dams across eight farrowing groups (Objective 2A). Sow body condition (post-farrowing and weaning) and piglet growth from 157 of the dams, including 65 sows used in a previous milk transcriptome study (FY 2020), were evaluated. Additionally, we have completed pathogen analyses on piglet fecal samples. In collaboration with the University of Gothenburg, we are generating milk oligosaccharide data from the same litters. Data collection for a sow lameness study funded by the National Pork Board was completed and analyses have begun (Objective 2A). Several electronically captured measures are predictive of gilt retention for breeding, indicating these measures coincide with observations of a trained swine herdsman. Important predictive measures relate to uniformity of each hoof’s pressure and duration of time each hoof is on the floor as the animal walks. Video data identifying the distance an animal walks and number of rotations they make each day are also predictive of structural soundness. Importantly, some electronic measures were useful to separate retained gilts into those that will become lame in the future from retained gilts that will not. Additional analyses are being conducted. About 5% of USMARC gilts exhibit behavioral anestrus, when gilts have a silent ovulation failing to show estrous behavior. Biological reasons for estrous failure are unclear and hormonal treatments to mitigate this problem are expensive, labor intensive and not wholly effective. To identify biological pathways involved in behavioral anestrus, we performed a case-control genome-wide association study (GWAS) on 2,421 normal cycling gilts and 515 behavioral anestrus gilts. A total of 81 significant SNP associations were found for additive, dominant and recessive models in >20 candidate genes. Genomic heritability was estimated to be 0.35 indicating that genetic variation exists for behavioral anestrus and it will respond to selection. This is the first GWAS performed for behavioral anestrus in pigs and identified biological mechanisms and functional variants to reduce pre-breeding anestrus. An analysis identified genomic regions associated with the number of vertebrae and teats (Objective 2A). SNP genotypes were imputed to a common set of informative markers, genomic estimated breeding values were predicted and marker effects back-solved. The analysis resulted in 43 regions significantly associated with at least one phenotype. Utilizing genome-level SNP from a reference panel provided by Michigan State University, we were able to show the dramatic impact that a PRE-1 repetitive element inserted in the first intron of vertnin has on all traits. We also observed that a mutation in the promotor region of vertnin has a modest effect on vertebrae number and a pair of missense mutations in ABCD4 impacts number of ribs. Research aimed at developing an electronic measure for mothering ability to improve sow productivity (Objective 2A) has produced interesting results. Geophone sensors (sensors that record vibrations in structural materials) were placed under flooring material in farrowing pens. We utilized video images and accelerometers placed on animals to record activities. Collaborators at Stanford University and University of Michigan were able to develop machine learning models to predict the posture and activity level of a sow, as well as nursing behavior of piglets. Sow activity is important as sows are more active just prior to giving birth. This system may be able to notify producers about particular sows that require assistance. A tensor decomposition approach reported in FY 2019 identified 36 SNP associated with swine feed efficiency. As a follow up to this study, and part of the effort to identify predictive markers (Objective 2A), a genotyping-by-sequencing (GBS) panel, featuring those 36 SNP, was developed in FY 2020. A total of 556 USMARC phenotyped animals were genotyped using this panel. Concordance of genotypes from GBS with SNP array was over 90%. After filtering, 16 SNP’s were used for feed efficiency GWAS. No SNP reached statistical significance, but three of the SNP’s approached significance.

1. Embryo elongation plays a critical role for successful pregnancy outcomes in the pig. Alterations in cellular signaling can result in delayed embryo elongation, leading to reduced embryonic survival. To identify genes expressed during the initiation of embryo elongation, ARS scientists at Clay Center, Nebraska, sequenced RNA of embryos at different stages of elongation (spherical, ovoid or tubular) from litters where all embryos were synchronized versus litters where multiple stages were observed. This study demonstrated dramatic changes in genes expressed from these distinct populations of embryos, particularly during the initial transition of the embryo from spherical to ovoid morphologies. Specifically, this study highlights the importance of changes in expression of genes associated with structural remodeling within the embryo and hormone signaling. The information gained from this study can be used to understand mechanisms essential to successful pig embryo elongation and increased embryo survival resulting in improved reproductive efficiency and piglet survival.

2. Geophone sensors and machine-learning predict sow postures and onset of farrowing. Due to labor constraints in swine production, many sows are unattended during parturition contributing to newborn piglet losses of 10-15%. Geophone sensors detect vibration in building materials. ARS scientists at Clay Center, Nebraska, teamed with scientists from the University of Nebraska, Stanford University, and University of Michigan to collect vibration data of flooring material in a sow's farrowing pen. Based on a machine-learning model, researchers were able to predict sow posture and posture changes that were validated with recorded video data and accelerometers placed on animals. As farrowing begins, sows become restless and change their posture frequently. These postural changes were detected by the computer model. As a result, a system can now be developed where information from geophone sensors is processed by a computer in real-time and then alerts are sent to caretakers to focus their attention on specific sows, which should reduce piglet mortality.

3. Brain expression profiles differ between normal and prebreeding anestrus gilts. Approximately 12% of gilts selected for entry into swine breeding herds fail to produce a litter because they do not display sexual behavior necessary for breeding when exposed to boars; a condition called prebreeding anestrus. To better understand how to solve this problem, ARS scientists at Clay Center, Nebraska, studied how genes in the hippocampus and the amygdala, two regions of the brain involved in sexual behavior, differed in gilts with prebreeding anestrus compared to control gilts with estrus. Both tissues showed dynamic changes in gene expression at different stages of the estrous cycle. Several of the genes expressed differently between anestrus gilts and control gilts regulate gamma-aminobutyric acid, a major inhibitory neurotransmitter. Consequently, correcting these regulatory pathways found in this study may reduce the incidence of prebreeding anestrus in commercial swine and improve production efficiency.

4. Genes responsible for ham halo defect discovered. Uniform pink color is a primary determinant of consumer acceptance of cured ham products. Recently, industry has received an increase in consumer concerns about non-uniformity of ham color, primarily lighter color in the periphery of muscles, termed “ham halo.” This effect is seen in both fresh and processed hams. The phenomenon is not caused by manufacturing procedures, rather it is associated with lower myoglobin concentration, pH, and muscle fiber variation. ARS scientists at Clay Center, Nebraska, compared gene expression profiles from light and normal colored hams to identify differentially expressed genes associated with the ham halo defect and meat color traits. Hundreds of genes were identified that provide biological knowledge of phenotypic differences resulting in the ham halo defect and insights into metabolic pathways responsible for biological differences in the lean color of pork. These pathways could provide key interventions to prevent the ham halo defect, increasing consumer satisfaction and demand for pork products.

Review Publications
Nonneman, D.J., Keel-Mercer, B.N., Lindholm-Perry, A.K., Rohrer, G.A., Wheeler, T.L., Shackelford, S.D., King, D.A. 2022. Transcriptomic analysis for pork color – The ham halo effect in biceps femoris. Meat and Muscle Biology. 6(1):1-8. Article 13050.
Rempel, L.A., Keel, B.N., Oliver, W.T., Wells, J.E., Lents, C.A., Nonneman, D.J., Rohrer, G.A. 2022. Dam parity structure and body condition during lactation influence piglet growth and gilt sexual maturation through pre-finishing. Journal of Animal Science. 100(4):1-9. Article skac031.
Walsh, S.C., Miles, J.R., Keel, B.N., Rempel, L.A., Wright-Johnson, E.C., Lindholm-Perry, A.K., Oliver, W.T., Pannier, A.K. 2022. Global analysis of differential gene expression within the porcine conceptus transcriptome as it transitions through spherical, ovoid, and tubular morphologies during the initiation of elongation. Molecular Reproduction and Development. Article 23553.
Wijesena, H.R., Nonneman, D.J., Keel, B.N., Lents, C.A. 2022. Gene expression in the amygdala and hippocampus of cyclic and acyclic gilts. Journal of Animal Science. Article skab372.
Lindholm-Perry, A.K., Kuehn, L.A., Wells, J., Rempel, L.A., Chitko-McKown, C.G., Keel, B.N., Oliver, W.T. 2021. Hematology parameters as potential indicators of feed efficiency in pigs. Translational Animal Science. 5(4). Article txab219.
Keel, B.N., Lindholm-Perry, A.K., Oliver, W.T., Wells, J.E., Jones, S.A., Rempel, L.A. 2021. Characterization and comparative analysis of transcriptional profiles of porcine colostrum and mature milk at different parities. BMC Genomic Data. 22. Article 25.
Nonneman, D.J., Lents, C.A. 2022. Functional genomics of reproduction in pigs: Are we there yet?. Molecular Reproduction and Development. 90(7): 436-444.
Lindholm-Perry, A.K., Meyer, A.M., Kern-Lunbery, R.J., Cunningham-Hollinger, H.C., Funk, T.H., Keel, B.N. 2022. Genes involved in feed efficiency identified in a meta-analysis of rumen tissue from two populations of beef steers. Animals. 12(12). Article 1514.