1. A whole-genome transmission disequilibrium test of fetal autolysis in a Porcine Reproductive and Respiratory Syndrome Virus challenge


Tianfu, Y., Wang, Z., Plastow, G.

Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB


High-density genomic data provide a powerful tool to detect genomic regions that are associated with traits of interest in pig production. For some important disease-related traits, however, genomic data may be inaccessible for extremely sick animals because of the difficulty in getting valid DNA samples. This study aims to use pedigree information to exploit records from these animals through a whole-genome transmission disequilibrium test, to identify candidate genomic regions that are associated with fetal autolysis after Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) challenge. Genomic data were obtained from 105 litters from an experimental pregnant gilt challenge model [1]. Parents and 873 non-autolysed fetuses were genotyped at a medium density (60k SNPs), and 435 autolysed fetuses did not have genotype due to the low quality of their DNA samples. For each SNP, a Chi-square test was conducted to detect distortion in genotypic frequencies from expectation in the non-autolysed offspring. Obvious deviation from Mendelian transmission rules implied a possible association between the genotype of the SNP and the autolysis rate in the offspring. Two SNPs showed distortion in transmission. They both had a significantly different frequency of heterozygotes from expectation in non-autolysed offspring. An ongoing related transcriptomic analysis of fetal response to PRRSV infection [2] provided additional support for these results. The two genomic regions around the SNPs harbour differentially expressed genes that are involved in pathways related to immunoglobulin and viral response.


[1] Ladinig et al., Variation in fetal outcome, viral load and ORF5 sequence mutations in a large scale study of phenotypic responses to late gestation exposure to type 2 porcine reproductive and respiratory syndrome virus. PLoS One 9, e96104 (2014).

[2] Wilkinson et al., unpublished data.




2. Identifying genomic predictors of vaccine response in swine


Wilkinson, J. 1, May, N. 1, Meng Y. 1, Napper, S. 2, and Plastow, G. 1

1Department of Agricultural, Food, and Nutritional Science, University of Alberta, Edmonton, Canada; 2Vaccine and Infectious Disease Organization, Saskatoon, Canada.

Vaccination is a widely used disease-prevention strategy, but response to vaccination is highly variable between individuals [1]. The aims of this study are to investigate the genomic factors that influence this trait in swine, and identify gene expression and protein phosphorylation biomarkers that are predictive of a strong response to vaccination.

Blood samples for transcriptomic and kinomic analyses were taken from 117 healthy piglets just prior to (Day 0) and in the first week following (Days 2 and 6) vaccination with an anti-Mycoplasma hyopneumoniae vaccine, Respisure One. Anti-Mycoplasma antibodies were quantified at Day 35 post vaccination, for the purpose of identifying biomarkers from the earlier time-points that correlate with a strong response to vaccination.

Antibody titers varied over 100-fold in magnitude between strong and weak responders. Genomic and kinomic analyses are currently underway, but preliminary kinome data from each of the first two groups revealed 10 peptides that were differentially phosphorylated between strong and weak responders. Several of these peptides were consistently identified across both groups. Interestingly, some differences are evident even prior to vaccination, which indicates that it may be possible to select individual animals for this trait without the requirement for vaccination.

The next steps are to complete the identification of biomarkers, and to validate their usefulness in a genetically distinct pig population. The long-term objectives of this research are to improve the response of swine to vaccination, at a time when the industry is under pressure to reduce its reliance on antibiotics for disease prevention.

[1] Haining, W.N. and Pulendran, B. (2012). Identifying gnostic predictors of vaccine response. Curr Opin Immunol 24: 332-336.

3. Genetic and genomic identification of dairy cattle based on immune response traits to improve dairy health


Wagter-Lesperance, L. 1, Atalla, H. 1, Cartwright, S. 1, Emam, S. 1, Fleming, K. 1 ,Gallo, N. 1, Hodgins, D. 1, Miglior, F. 2, Paibomesai, M. 1, Read, L. 1, Ross, M. 1, Sargolzaei, M. 2, Schmied, J.1, Thompson-Crispi, K. 1 and Mallard, B.1


1Department of Pathobiology, 2 CGIL, Department of Animal and Poultry Science, University of Guelph, Guelph, Ontario


Dairy cattle can be classified based on estimated breeding values (EBV) of heritable (h2~0.25) immune responses (IR) using a patented method known as the High Immune Response (HIR)TM technology. High responders have genetically superior immunity and enhanced disease resistance compared to average or low responders. This provides a novel breeding and management tool to improve herd health. Genome-wide association studies using the Illumina 50k chip have found unique genetic profiles related to antibody (AMIR) and cell-mediated response (CMIR) making broader application of HIR possible for the dairy industry.  Epigenetic, as well as genetic, influences on bovine IR have been noted. Additionally, immunologically characterised cattle provide an effective means to investigate the specific mechanisms associated with disease resistance. The main pathways of IR from recognition and initiation to mounting a protective response are under investigation in this lab.  For example, we have found that at the recognition stage, high responders express greater amounts of toll-like receptor 2 (TLR2) on blood mononuclear cells, which bind and respond to relevant pathogens.  At the next step, high responders have a greater ability to phagocytose bacterial particles and produce bactericidal responses such as nitric oxide. Further down the pathway when adaptive IR is mounted, other differences were observed. For example, cows with high EBV for AMIR have more total IgG, specific antibody, and β-lactoglobulin in their colostrum. Finally, we are investigating if small regulating microRNAs within exosomes of colostrum and milk of high and low responders differ and whether this affects animal and human health.




4. Plasma levels of volatile fatty acids as potential predictors of feed intake in beef cattle


Karisa, B. K.1, Basarab, J. H.2, Moore, S. S.3, Plastow, G.1

1Department of Agricultural Food and Nutritional Science, University of Alberta, Canada; 2 Alberta Agriculture and Rural Development, Lacombe Research Centre; 3 Centre for Animal Science, Queensland Alliance for Agriculture & Food Innovation, The University of Queensland, St Lucia


In this study, we show results from three consecutive studies performed in diverse populations of beef cattle to identify indicators of feed efficiency. Plasma levels of volatile fatty acids (VFAs) were associated with feed intake across all the three studies (P <0.05). These results indicate that acetate and other volatile fatty acids have the potential to be used as predictors of feed intake in beef cattle. This debate, on the relationship between plasma levels of volatile fatty acids and feed intake in ruminants, has been ongoing for several decades without a clear understanding of the exact mechanism influencing the relationship. Earlier reports showed that ruminants match their intake and rate of absorption of VFAs from the rumen with their utilization in cellular metabolism. Therefore the rate of acetate clearance from plasma had a significant influence on feed intake. It was known that animals needed to maintain the levels of acetate in blood within specific physiological limits and as the levels reduce the animal would develop appetite to replenish them. Recent reports on the effect of plasma acetate on appetite in humans have gone further to show that the mechanism of regulation does not only function at the digestive system. In fact, acetate was shown to cross the blood-brain barrier and directly impact appetite at the central nervous system. This knowledge of the mechanisms regulating feed intake may create new perspectives on the underlying causes of variation in feed efficiency and result in identification of new markers associated with feed efficiency.




5. Genome wide mRNA analysis of rumen papillae revealing distinct response during rapid grain adaption in beef cattle


Zhao, K. [1], Wang, D., Liang, G., Chen, Y., Guan, L[2].


Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB T6G2P5, Canada.


High grain diet usually causes ruminal acidosis, which can impact on animal productivity and health. However, little is known about the mechanism regulating this process, especially at transcriptional level in rumen. In this study, we conducted a genome wide mRNA comparison of rumen papillae under high grain challenge. Beef heifers were randomly assigned to a rapid grain adaptation (RGA) treatment (n = 15, from 3% to 92% grain) over 29 days. Rumen papillae were collected for RNA-seq analysis when the heifers were fed 3%, 75%, and 92% grain diets. The mean ruminal pH change was strongly dependent on the animal and followed three patterns: continually decreased as the proportion of grain increased (DD group, n = 5), decreased when the diet transitioned from 3% to 75% grain with little change thereafter (DB group, n = 5), and decreased only when the diet transitioned from 3% to 75% grain and then increased (DU group, n = 5). It’s found that 7, 9 and 308 genes correlated with the pH trend and were conserved in DD, DB and DU group, respectively. Among them, the expression of ADA (adenosine deaminase), an immune activation factor, was consistently increased in DD group, suggesting the elevated immune function may contribute to the less resistance of acidosis. Retinoblastoma-binding protein 8 (RBBP8) and lysine-specific demethylase b (Kdm1b) genes were up-regulated with increasing grain content in DB group (P < 0.05), which might stimulate the cell proliferation. Over all, our results showed that the distinct immune response and cell renewal might be related to rapid grain adaption in beef cattle.



6. Genomic prediction of pig feed efficiency component traits using 80K SNP chip


Zhang, C.1, Kemp, R.2, Boddicker, N.2, Wang, Z.1 and Plastow, G.1*


1 Dept. Agricultural, Food & Nutritional Sciences, University of Alberta, Edmonton, AB T6G 2P5; 2 Genesus Inc., Oakville, MB, Canada

* Corresponding author, E-mail: graham.plastow@ales.ualberta.ca;

This study evaluated the genetic parameters and accuracy of genomic prediction for pig feed efficiency component traits, including average daily feed intake (ADFI) and average daily gain (ADG) during the test period (~ 75 kg – 130 kg), ultrasound backfat depth (BFD) and loin depth (LD) at the end of test. A total of 1,359 pure Duroc boars with both genotypic and phenotypic records were split into training (n = 1,160) and validation datasets (n = 199) according to their birthdate before and after June 10th 2014, respectively. Genomic prediction was conducted using Bayes B in GenSel and included 51,998 SNPs with minor allele frequency > 0.001. Accuracy was evaluated through the correlation of GEBV and corrected phenotype divided by square root of heritability. Variance components were estimated by an animal model in ASREML. Heritabilities from pedigree analysis were moderate to high for the studied traits ranging from 0.22 to 0.52. Comparatively, heritability estimates using marker information were about half of those from pedigree analysis, suggesting “missing heritability”. Positive and high genetic correlations were observed between ADFI and traits of ADG and BFD. Accuracies of genomic prediction for ADFI, ADG, BFD and LD were 0.34, 0.14, 0.71 and 0.21, which were related to how much genetic variance was captured by the SNPs used in the predictions. These results provide new insights into pig breeding and selection programs using genomic information. However, higher SNP densities and/or increased number of training animals may help to resolve the missing heritability issue and consequently improve the accuracy of GEBV.


7. Genomic selection for disease resilient pigs using immune traits from blood samples


Lim. J. 1,2, Field C. J. 2, Plastow. G.S.1,2

1Livestock Gentec Centre, 2Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Canada.



Disease resilience is defined here as the ability of pigs to respond to infection in a way that minimizes the impact of disease. Many studies have been done in the past to identify immune traits that are responsible for orchestrating response to disease in pigs. However, immune traits identified in an artificial challenge trial with a specific pathogen may not be useful in breeding selection: as this does not completely reflect “real-life” in commercial farms which may harbour many pathogens at any one time. Our aim is to identify genes involved in orchestrating immune traits which are phenotypically different between healthy and sick pigs. A total of 3,600 high health nursery crossbred barrows of Yorkshire and Landrace lines will be introduced into a challenge model established in a test station using seeder pigs from commercial farms with different diseases to simulate the commercial situation. Peripheral blood samples will be collected before, and at different time points after, introduction to the test station. Complete blood count and serological tests will be done for all samples. Samples with phenotypically different immune traits between healthy and sick groups will then be selected for gene expression analysis for both whole blood and cell subsets to identify genes that are associated with health outcomes in pigs. In conclusion, we hypothesize that the interactions between subpopulations of immune cells will explain overall immune response, subsequently determining the degree of disease resilience of the pigs.


8. Characterization of innate inflammatory responses against Salmonella typhimurium in a swine peritoneal model


More Bayona J. A.1, Gao. Y.2, Benjamin, P. W.2, Barreda, D. R.1,2


1Department of Biological Sciences, University of Alberta, Edmonton, AB; 2Department of Agricultural, Food and Nutritional Sciences, University of Alberta, Edmonton, AB


The efficient induction and regulation of innate inflammatory responses is central to the protection against pathogens, control of unnecessary tissue damage and efficient energy utilization by an animal host. In this study, we examined the kinetics and absolute levels of innate inflammatory responses against Salmonella typhimurium using an in vivo peritonitis model. Our goal was to define key cellular and molecular events during this immune response. We injected 109 CFU of heat-killed Salmonella typhimurium in the peritoneal cavity of 7weeks-old pigs and the inflammatory response was evaluated at different time points (8, 12, 18, 24, 36 and 48 hours post challenge). The absolute numbers and identity of various leukocyte populations were analyzed using conventional and imaging flow cytometry. In addition, we examined changes in cellular viability and the expression of pro- and anti-inflammatory cytokine genes using quantitative PCR. Overall, we identifed the times post challenge for peak induction and control of immune inflammatory responses. These novel tools offer an improved capacity to assess modulation of swine immune responses and should prove indispensable for our understanding and prevention of swine diseases.



9. Characterization of bacterial colonization in large intestine of pre-weaned dairy calves using quantitative real time-PCR


Song, Y.1, Malmuthuge, N., Steele, M., Guan, L2


Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, ABT6G 2P5, Canada.


The mammalian hindgut is populated with a dynamic bacterial population, which has been suggested to associate with gut functions and host health. However, prior to this experiment, the microbiota composition in the hindgut of pre-weaned calves had not been studied. This study aimed to investigate population dynamics of total bacteria, Bifidobacterium, E. coli, Faecalibacterium prausnitzii and Clostridium cluster XIVa in the calf large intestine during the first 6 weeks of life. Tissues and digesta samples were collected from the cecum, colon and rectum of 18 Holstein bull calves at day 7 (n=6), day 21 (n=6), and day 42 (n=6), following euthanasia. Quantitative real time-PCR was used to estimate the copy number of 16S rRNA gene of total bacteria and selected bacterial groups associated with tissue and digesta. The density of large intestinal digesta-associated E. coli exhibited two and four-fold higher copy numbers on day 7 compared to that of day 21, and day 42, respectively. Similarly, the density of E. coli attached to cecum and colon tissue at day 7 was higher compared to day 21, and day 42. In contrast, the densities of butyrate producing beneficial bacteria Clostridium cluster XIVa and F. prausnitzii in colon tissue were highest on day 21. This study revealed that the prevalence of bacteria colonized in the large intestinal communities varied depending on calf age and intestinal region. Thus, detailed understanding of microbial colonization and their functions in the large intestine requires future studies based on sample type and region. 


10. Testing Production Performance and Foraging Behavior of Cattle on Pasture with Divergent Molecular Breeding Values for Residual Feed Intake

C. Moore1,2, N. Lansink1,2, J. Basarab2,3, C. Fitzsimmons2,4, S. Nielsen5, G. Plastow2, E. Bork1

1Rangeland Research Institute, 410 Agriculture/Forestry Center, University of Alberta, Edmonton, AB 2Livestock Gentec at University of Alberta, 1400 College Plaza, 8215 112st Edmonton, AB; 3Alberta Agriculture and Forestry, Lacombe Research Centre, Lacombe, AB; 4Agriculture and Agri-Food Canada; 5Dept. of Renewable Resources, University of Alberta, Edmonton, AB


Residual feed intake (RFI) is a heritable trait that can be used to measure feed efficiency in cattle and serve as a tool for managing costs in beef operations. Measures of RFI however have only been evaluated under drylot conditions, where animals are on a standardized diet and foraging behaviour is eliminated. This study is exploring the utility of RFI molecular breeding values (MBVs) by relating these to cattle performance while grazing under free-range conditions within heterogeneous pastures. We are also using novel technologies (GPS collars and leg mounted pedometers) to determine if there are relationships between cattle habitat selection and activity budgets with associated MBVs for RFI. This work was conducted at the University of Alberta Mattheis Research Ranch, just north of Brooks, Alberta in the Mixedgrass Prairie. A subset of commercial cows estimated to have high and low MBVs for RFI were evaluated for industry-relevant production metrics such as calf growth, cow rebreeding interval and cow weight gain and body condition recovery, during summer 2015. A subset of these cows were fitted with Lotek 3300LR GPS collars and AfiAct II pedometers to track their location and movement during the grazing season. Behavioural data for each animal and treatment group will be used to develop resource selection functions for beef cattle with divergent RFI scores. Methods of data collection throughout the summer will be discussed in the poster.


11. Optimizing feed intake recording to increase the rate of improvement for feed efficiency in beef cattle


Manafiazar, G.1,3, Basarab, J.2, McKeown, L.1, Stewart-Smith, J.3, Baron, V. 4, MacNeil, M. 5, and Plastow, G1


1Agricultural Food and Nutritional Science Department, University of Alberta, Edmonton AB; 2Alberta Agriculture and Forestry, Lacombe Research Centre, Lacombe, AB; 3BeefBooster Inc. Calgary, AB; 4Agriculture and Agri-Food Canada, Lacombe Research Centre, Lacombe, AB; 5DeltaG, Miles city, Montana, USA


Measuring individual feed intake is a primary limiting factor for genetic studies on feed efficiency and for implementing effective genetic improvement programs by the beef industry. Currently, daily feed intake is recorded on growing animals for periods of 76-120 days during their growth phase to test for feed efficiency. It is proposed that test duration for feed intake could be decreased to 35-40 days without significantly compromising accuracy of the test. In this research, data from 1710 bulls in 37 contemporary groups, and 477 replacement heifers in 8 contemporary groups were used to test whether short period (30-40 days) can be used to predict regular test period (60-85 days) feed intake. Observations more than three standard deviations from the mean within contemporary groups were excluded from analysis. The results showed that average short and regular period feed intakes were highly correlated in both of bulls and replacement heifers (0.96 and 0.98, respectively). Regular feed intake is predicted from linear regression of short period feed intake with accuracy of 0.93 in bulls and 0.95 in replacement heifers. It is concluded that shortening the feed intake period to 40 days from 76-85 days would double the number of animals tested for feed intake and reduce costs with the current infrastructure, while only compromising accuracy by 5-7%.

12. Methane and carbon dioxide emissions from high and low residual feed intake beef heifers

Manafiazar, G.1, McKeown, L.1, Baron, V.2, Plastow, G.1, Ominski, K.3, and Basarab, J4

1Agricultural Food and Nutritional Science Department, University of Alberta, Edmonton AB; 2Agriculture and Agri-Food Canada, Lacombe Research Centre, Lacombe, AB; 3Department of Animal Sciences, University of Manitoba, Winnipeg, MB; 4Alberta Agriculture and Forestry, Lacombe Research Centre, Lacombe, AB 

Selection for low residual feed intake is one of the three registered protocols for reducing greenhouse gas emissions from beef cattle production in Alberta. In this research, 86 replacement heifers were test for residual feed intake adjusted for back-fat (RFIfat) in three dry-lot pens fitted with the GrowSafe System feeding bunks at Lacombe Research Center. In one pen, animals (n=26) were noninvasively and automatically monitored for daily CH4 and CO2 emissions using a GreenFeed Emission Monitoring system in conjunction with feed efficiency testing. Animals in this pen at the end of test consisted of 12 low (-0.32 ± 0.14) and 14 high RFIfat (0.34 ± 0.21) heifers with an average start body weight of 345.4 kg (SD = 31.4). Low RFIfat heifers consumed 7.1% less DMI (8.09 vs. 8.71 kg DMI/day) compared with high RFIfat heifers, but had similar on-test growth rate (1.20 ±0.11 g/day). Low RFIfat heifers compared with their high RFIfat pen mates emitted 6.5% lower CH4 (196.4 vs. 210.0 g/day, respectively) and 5.6% lower CO2 (6302.9 vs. 6680.2 g/day, respectively). However, efficient heifers had 2.7% higher CH4 yield (25.15 vs. 24.49 g/kg DMI, respectively) and 4.5% higher CO2 yield (819.29 vs. 784.07 g/kg DMI, respectively) compared with inefficient heifers. Thus, selection for low RFIfat would decrease greenhouse gas intensity (kg CO2e/kg beef) from the beef industry; though a portion of this mitigation was moderated by the slightly higher methane yield per kg of feed intake in efficient animals. Further research is required to phenotype and understand the biological processes of individual beef cattle that are both efficient and have lower CH4 and CO2 yield. 

13. Improving feed efficiency and reducing methane emissions from dairy cows using milk Mid-infrared spectroscopy to support “green Alberta milk”

Hailemariam, D.1, Manafiazar, G.1, Yang, T.1, Liu, J.J.1, Miglior, F.2, Basarab, J.1,3, Plastow, G..1, Grelet, C. 4, Petreny, N5 and Wang, Z.1

1Department of Agricultural, Food and Nutritional Science, University of Alberta; 2Canadian Dairy Network, 660 Speedvale Avenue West, Suite 102, Guelph, Ontario N1K 1E5; 3Alberta Agriculture and Forestry, Lacombe Research Centre, 6000 C & E Trail, Lacombe, AB, Canada, T4L 1W1; 4Walloon Agricultural Research Centre, Rue de Liroux 9, B-5030 Gembloux, Belgium; 5CanWest DHI, 660 Speedvale Avenue West, Suite 101, Guelph, Ontario N1K 1E5

Genetic variation in residual feed intake (RFI) and Methane has been reported in dairy cattle. However,  RFI and CH4 has been missed in current dairy breeding goals mainly due to the expense and difficulty in obtaining sufficient data to generate accurate estimates of genetic merit for selection. Thus, we hypothesized that identifying low-cost and easy-to-measure milk indicator traits would assist to perform genetic selection. We conducted trait recording (daily feed intake, monthly body weight & body condition score and CH4 emission) on 75 lactating dairy cows at the University of Alberta. Milk samples were collected once weekly at consecutive evening and morning milking and analysed at Dairy Herd Improvement (DHI) to determine milk composition traits. The GreenFeed system was used to estimate daily CH4 emissions from 30 lactating dairy cows in two batches. The methane trial involved adaptation (4-7d), test (14d) and validation (2-3d) periods. Measurements were conducted at 2 and 8 time points in the 24 h cycle during test and validation periods respectively. The preliminary result from the CH4 trial indicated that the correlation between the test and validation period was 0.65 and 0.72 for the first and second batch respectively. The average daily methane emission in test and validation periods was 388±63 and 340±93 g/day respectively. The mean daily methane emission per kg of DMI decreased with lactation despite increasing DMI. Taken together, the comfortable degree of correlation we observed between the test and validation periods verified the robustness of the methodology used to generate the methane data.



14. Improvement of cow feed efficiency and the production of consistent quality beef using molecular breeding values for RFI and carcass traits -The “Kinsella Breeding Project”

Ekine-Dzivenu, C. 2Plastow, G.1,2, Basarab, J.2,3, Fitzsimmons, C.1,2,4, Li, C.1,2,4, Chen, L.2, McKeown, L.1,3, Irving, B.1, Lynch-Staunton, T. 2, Baker, L. 4, Vinsky, M. 4, Miller, S. 5, Wang, Z. 1,2, Crowley, J. 2, Colazo, M. 3, Ambrose, D. 3, Juarez, M. 4, BruceH. 1,2

1Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB, Canada, T6G 2P5; 2Livestock Gentec, 1400 College Plaza, 8215 112 St, Edmonton, AB, Canada, T6G 2C8; 3Alberta Agriculture and Rural Development, Lacombe Research Centre, 6000 C&E Trail Lacombe, AB, Canada, T4L 1W1; 4Agriculture and Agri-Food Canada, Lacombe Research Centre, 6000 C&E Trail, Lacombe, AB, Canada T4L 1W1; 5 AgResearch, Invermay, New Zealand


The Kinsella Breeding Project was designed to show how genomics can be used by beef producers to make better breeding and management decisions by demonstrating the use of molecular breeding values (MBVs) in improving cow feed efficiency (residual feed intake (RFI)) and sorting feeder cattle into more uniform groups for slaughter.

For this project, the University of Alberta Kinsella Composite (KC) cattle population was split into equal groups named “efficient” (-RFI) and “control” (+RFI) herds in spring 2013. The purebred Angus and Charolais herds at the Kinsella ranch are also part of the project. 

Preliminary results from the KC population for 2014 born calves show that at P < 0.05, least square means (LSM±SE) for RFI and dry matter intake (DMI) of efficient (n= 75, RFI= - 0.1593 ± 0.09, DMI= 10.41 ± 0.13) and control (n= 82, RFI= 0.1362 ± 0.09, DMI= 10.72 ± 0.12) steers were significantly different. For heifers, differences in RFI and DMI values of efficient (n= 67, RFI= - 0.0255 ± 0.06, DMI=9.74 ± 0.09) and control (n= 73, RFI= 0.0374 ± 0.06, DMI= 9.68 ± 0.08) groups did not reach statistical significance. However, average daily gain (ADG) and weaning weight (WWT) were statistically different between heifer groups (Efficient, ADG= 1.28 ± 0.02, WWT= 488.64 ± 5.24; Control, ADG= 1.21 ± 0.02, WWT= 502.86 ± 5.06) but were not different between steer groups (Efficient, ADG= 1.65 ± 0.03, WWT= 483.93 ± 8.33; Control, ADG= 1.62 ± 0.03, WWT= 503.42 ± 8.05). Birth weight (BWT) was not significantly different between RFI groups for heifers and steers.

These initial results seem promising. They show response on selection for RFI.

Preliminary results from the feeder cattle sort program show potentials for using MBV to sort cattle that meet specifications for quality and yield.


15. Genome wide association studies (GWAS) for milk production traits in Mediterranean water buffalo

Liu, J.J.1, 3, Liang, A.X.1, Plastow, G.3, Zhang, C.Y. 3, Wang, Z.Q.3, Salzano A.2, Gasparrini, B.2, Canmpanile, G.2, and Yang, L.G.1

1 College of Animal Science and Technology, Huazhong Agricultural University, Wuhan, China; 2 Department of Veterinary Medicine and Animal Productions, University of Naples "Federico II", Naples, Italy; 3 Department of Agricultural, Food & Nutritional Sciences, University of Alberta, Edmonton, AB

This study aimed to identify genomic regions and/or markers associated with milk production in Mediterranean water buffalo. Six traits including peak milk yield per lactation (PM), 270-days total milk yield (MY270), 270-days fat yield (FY270) and fat percentage (FP270), 270-days protein yield (PY270) and protein percentage (PP270) were collected from 474 individuals with 1377 lactations. Estimated breeding values (EBVs) calculated through the linear mixed model in ASReml by accounting for fixed effects of herd, year, season, parity, calf gender and age, were considered as the adjusted phenotype for GWAS. Genotyping was conducted at Delta Genomics using the Axiom® Buffalo 90K Genotyping Array. After quality control, a total of 60,387 SNPs and 397 animals remained for the analysis which was performed by ridge regression BLUP (RR-BLUP). Genome wide multiple test of P-value was adjusted by Bonferroni’s correction. Four SNPs were detected associated with at least one of the studied traits (P < 0.05). Referring to cattle genomic map, these SNPs were located on cattle chromosomes (BTA) 3, 12 and 14, which are homologous to buffalo chromosomes 6, 13 and 15, respectively. According to this study, the significant gene on BTA 14 is unlikely to be DGAT which influences these traits in dairy cattle. These findings reveal useful information about the genomic regions affecting buffalo milk production, which may provide new insights into breeding programs using genomic information. However, a more precise buffalo genomic map is required to identify the genes affecting the traits, and a larger sample size may increase the detection power and help validate these results


16. Acute phase proteins: potential biomarkers to predict pig resilience


Yang, Z. and Plastow, G.

Department of Agricultural, Food, and Nutritional Science, University of Alberta, Edmonton, AB

Acute phase proteins (APP) are proteins whose level changes very rapidly after infection or exposure to other stressors. They are not only an early systemic sign of animal disease, but also can indicate the health status of animals indicating level of exposure and, in apparently healthy pigs, subclinical infections. Most of the APPs are regarded as non-specific markers. However, there are some APPs that appear to be associated with specific types of pathogen. For example, increased levels of serum haptoglobin, but not alpha-1 acid glycoprotein, were observed from 7 to 21 days post-inoculation after experimental porcine reproductive and respiratory syndrome (PRRS) challenge.[1] Besides their indicator health status, APPs may also be used as an index for assessing productive performance. In particular, haptoglobin was found to be negatively correlated with growth rate in pigs fed a diet supplemented with β-glucans as growth promoter.[4] Genetic variation in plasma alpha-1 acid glycoprotein (AGP) level was discovered between different pig lines selected for lean growth under restricted feeding[2] and also between Meishan and Large White pig breeds.[3] Albumin appears to be a negative acute phase protein related to nutritional status and little is known in terms of its relationship with inflammatory status of individuals. Resilience is related to both health status and productive performance, so that an index of acute phase proteins may have potential for use as biomarkers to predict pig resilience. This is being investigated with samples from natural challenge studies.

[1] Asai, T., Mori, M., Okada, M., Uruno, K., Yazawa, S., & Shibata, I. (1999). Elevated serum haptoglobin in pigs infected with porcine reproductive and respiratory syndrome virus. Veterinary immunology and immunopathology, 70(1), 143-148.

[2] Clapperton, M., Bishop, S. C., Cameron, N. D., & Glass, E. J. (2005). Associations of acute phase protein levels with growth performance and with selection for growth performance in Large White pigs. Animal Science, 81(02), 213-220.

[3] Clapperton, M., Bishop, S. C., Pineiro, M., Campbell, F. M., & Glass, E. J. (2007). The association between plasma levels of acute phase proteins, haptoglobin, alpha-1 acid glycoprotein (AGP), Pig-MAP, transthyretin and serum amyloid A (SAA) in Large White and Meishan pigs. Veterinary immunology and immunopathology, 119(3), 303-309.

[4] Hiss, S., & Sauerwein, H. (2003). Influence of dietary ßglucan on growth performance, lymphocyte proliferation, specific immune response and haptoglobin plasma concentrations in pigs. Journal of animal physiology and animal nutrition, 87(12), 2-11.



17. Accuracy of genomic prediction of breed composition and heterosis in beef cattle using 50K SNP genotype


Akanno, E.C.1, Chen, L.1, Karisa, B.1, Basarab, J. A.1, 2, and Plastow, G.S.1


1Livestock Gentec, Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB, Canada; 2Lacombe Research Centre, Alberta Agriculture and Rural Development, 6000 C & E Trail, Lacombe, AB, Canada


The knowledge of breed composition has applications in beef cattle production including assessment of heterosis and accounting for population stratification in genetic evaluation and genome-wide association studies. This study examined the feasibility and accuracy of using Illumina BovineSNP50 (50K) array (Illumina Inc., San Diego, CA) to estimate individual cattle breed composition and heterosis over estimates from pedigree. First, pedigree information was used to compute breed fractions for 1,124 crossbred cattle from the Lacombe Research Station, Lacombe, AB. Given the breed composition of sires and dams, retained heterosis was computed for all individuals. Similarly, retained heterozygosity was assessed using individual’s breed fractions. Second, all animals’ 50K genotypes in AB format were screened for quality control and recoded as 0, 1, 2 based on number of B alleles. The admixture software [1] was used to compute individual’s genomic breed fractions. A cross-validation method was applied to find the optimal number of ancestral breeds. Genome-based heterozygosity was then computed for all individuals. Lastly, accuracy was assessed as Pearson’s correlation between pedigree- and genome-based predictions. The average breed composition for Angus, Charolais and Hereford were 0.53, 0.23 and 0.25 for pedigree-based and 0.47, 0.26, 0.28 for genome-based predictions, respectively. Accuracy of predicting breed composition ranged from 0.94 – 0.96. Additionally, genome-based heterozygosity predicted well the retained heterosis from pedigree with an accuracy of 0.96. The genomic prediction approach can aid analyses that depend on knowledge of breed composition and serve as a reliable method to predict heterosis in order to improve the efficiency of commercial crossbreeding schemes. 


[1] Alexander, D. H., Novembre, J. and Lange, K (2009). Fast model-based estimation of ancestry in unrelated individuals. Genome Research, 19:1655-1664.



18. Use of genomic tools to improve feed efficiency in purebred Hereford cattle


Crowley, J.1,2, McDonald, T.3, French, N.3, Scott, S.4, Torres, W.5, Lynch-Staunton, T.1, McKeown, L.1, Basarab, J.1,6, Plastow, G.1


1Livestock Gentec at University of Alberta, 1400 College Plaza, 8215 112st Edmonton, AB; 2Canadian Beef Breeds Council, 165-6815 8 St NE, Calgary, AB T2E 7H7; 3Olds College, 4500-50th St. Olds, AB, T4H 1R6; 4Canadian Hereford Association, 5160 Skyline Way NE, Calgary, AB T2E 6V1; 5Cattleland Feedyards Ltd, P.O. Box 2265, Strathmore, Alberta T1P 1K2; 6Alberta Agriculture and Forestry, Lacombe Research Centre, Lacombe, AB, Canada


A new partnership of Olds College, the University of Alberta and their industry partners formed to develop resources for the Canadian Hereford Association (CHA) to include feed efficiency to its selection programs. Feed intake was recorded using GrowSafe Systems on 1197 Hereford bulls at Olds College and Cattleland Feedyards. Expected progeny differences (EPDs) for residual feed (RFI) intake and residual intake and gain (RIG) were calculated using both phenotypic and genomic information, achieving accuracies of >0.35 on animals with a phenotype. Both traits were evaluated singly using a linear animal model with 5,654 animals in the pedigree file to account for relationships among animals. Fixed effects included in the model were age of dam, age of test and contemporary group (period, location, pen). The heritability of RFI and RIG were 0.36 and 0.31, respectively. The data was also used in the Canadian Cattle Genome Project to test and validate panels of single nucleotide polymorphisms (SNPs) and genomic predictions. Most recently it has helped predict EPDs for Hereford cattle that are part of a trial of feed efficiency and grazing behaviour at the Mattheis ranch. Identifying low RFI breeding stock has the potential to cut feeding costs, the number one non-capital cost faced by cattle producers. Additional benefit will accrue from reduced environmental footprint from reduced methane and manure output of efficient cattle. Improved characterization of efficiency and production of associated EPDs will increase the competitiveness and sustainability of the Canadian beef industry through reduced feed costs to the primary producers.


19. Genomics and the Hays Converter breed


Crowley, J.1,2, Akanno, E.1, Khorshidi, R.1, Hays, D.3, Trautman, D.1, Fleming, A.4, Miller, S.5, Basarab, J.1,6, Plastow, G.1


1Livestock Gentec at University of Alberta, 1400 College Plaza, 8215 112st Edmonton, AB; 2Canadian Beef Breeds Council, 165-6815 8 St NE, Calgary, AB T2E 7H7 ;3Red Bow Ranching, Calgary, AB

4The Centre for Genetic Improvement of Livestock, University of Guelph, 50 Stone Road East, Guelph, Ontario, N1G 2W1; 5AgResearch, New Zealand; 6Alberta Agriculture and Forestry, Lacombe Research Centre, Lacombe, AB, Canada


The Hays Converter (HC) was the first Canadian breed to be recognized in terms of the Canadian Livestock Pedigree Act. It combined the genetics of the Hereford, Holstein and Brown Swiss breeds. The breed was established to have a higher percentage of lean meat, but most importantly, to convert feed to gain as efficiently as possible and to thrive under Canadian conditions with minimal husbandry. The animals in the study are owned by Red Bow Ranching Ltd., Calgary. There are three other breeders of pure HC managed in conjunction with commercial beef production operations in Manitoba and Quebec. Genomics offers the potential to provide tools to increase the rate of selection in beef cattle as well as providing the opportunity to select for expensive or difficult to measure traits such as feed efficiency and carcass and meat quality. One of the challenges in making genomic selection work in different breeds relates to the different genetic architectures between breeds, so that it is not possible to apply results from one population to another. In the case of the HC’s under study the existing breeding population is now less than 120 cows and inbreeding is relatively high. This project seeks to develop a new improvement strategy utilizing genomics to improve rates of genetic improvement while controlling inbreeding. Results have been obtained for breed proportions and levels of inbreeding using SNP genotypes and these will be combined with phenotypic data being collected for yield and quality and expanded to feed efficiency.


20. Developing a small SNP panel to predict feed efficiency in Canadian beef cattle


Lansink, N.1,6, Crowley, J.1,2, Moore, C.1,6, Karisa, B.1, Abo-Ismail, M.3,4, Miller, S.5, Stothard, P.1, Bork, E.6, Basarab, J.1,7, Plastow, G.1


1Livestock Gentec at University of Alberta, 1400 College Plaza, 8215 112st Edmonton, AB; 2Canadian Beef Breeds Council, 165-6815 8 St NE, Calgary, AB T2E 7H7 ;3The Centre for Genetic Improvement of Livestock, University of Guelph, 50 Stone Road East Guelph, Ontario, N1G 2W1; 4Animal and Poultry Science Department, Damanhour University, Damanhour, Egypt; 5AgResearch, New Zealand; 6Rangeland Research Institute, 410 Agriculture/Forestry Center, University of Alberta, Edmonton, AB; 7Alberta Agriculture and Forestry, Lacombe Research Centre, Lacombe, AB


The objective of this project is to develop a new genomic tool for selecting beef cattle to improve feed efficiency. Previous work has identified significant associations between functional single nucleotide polymorphisms (SNPs) and feed efficiency in “discovery populations” of mainly crossbred or hybrid cattle [1,2]. However, these effects have not been validated in unrelated populations. We are combining these SNPs with those screened from sequences generated as part of the Canadian Cattle Genome Project and the 1,000 Bulls Genome Project [3]. Our focus is on predicting functional variants in candidate genes [4,5] that are segregating in Angus and Hereford breeds. These will be used to develop a small marker panel (<200 SNPs) that will then be used to genotype around 800 Angus, Hereford and Angus/Hereford crosses from the “phenomic gap” project, for which detailed phenotypic records for feed efficiency and carcass yield are available. Finally, this panel will be used to predict the feed efficiency of a cohort of heifers being tested at the Lacombe Research Centre and the Mattheis Ranch for feed intake and methane production. The ultimate aim is to produce a customized cost-effective SNP panel which will have the potential to significantly increase the accuracy of predicting feed efficiency in beef cattle.


[1] Abo-Ismail, M. K., Vander-Voort, G., Squires, J. J., Swanson, K. C., Mandell, I. B., Liao, X., Stothard, P., Moore, S. S., Plastow, G., Miller, S. P. 2014. Single nucleotide polymorphisms for feed efficiency and performance in crossbred beef cattle. BMC Genet; 15: 14.

[2] Karisa, B. K., Thomson, J., Wang, Z., Stothard, P., Moore, S. S., Plastow, G. S. 2013. Candidate genes and single nucleotide polymorphisms associated with variation in residual feed intake in beef cattle.

[3] Daetwyler, H. D., Capitan, A., Pausch, H., Stothard, P. et al.. 2014. Whole-genome sequencing of 234 bulls facilitates mapping of monogenic and complex traits in cattle, Nature Genetics, 46, 858–865

[4] Grant, J.R., Arantes, A.S., Liao, X, Stothard, P. 2011 In-depth annotation of SNPs arising from resequencing projects using NGS-SNP.  Bioinformatics 27:2300-2301.

[5] Huang D.W., Sherman B.T., Lempicki R.A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources, Nat. Protocols. 4 (2008) 44-57.


21. Diurnal pattern of methane emission from lactating dairy cows


Hailemariam, D.1, Manafiazar, G.1, Miglior, F.2, Basarab, J1,3., Plastow, G..1, Grelet, C. 4, Petreny, N5 and Wang, Z.1


1Department of Agricultural, Food and Nutritional Science, University of Alberta; 2Canadian Dairy Network, 660 Speedvale Avenue West, Suite 102, Guelph, Ontario N1K 1E5; 3Alberta Agriculture and Forestry, Lacombe Research Centre, 6000 C & E Trail, Lacombe, AB, Canada, T4L 1W1; 4Walloon Agricultural Research Centre, Rue de Liroux 9, B-5030 Gembloux, Belgium; 5CanWest DHI, 660 Speedvale Avenue West, Suite 101, Guelph, Ontario N1K 1E5



Enteric methane emission from dairy cows varies during the diurnal cycle. To characterize the diurnal pattern of methane emission from lactating dairy cows in the University of Alberta - Dairy Research and Technology Center (DRTC), we measured methane emission from 30 cows in two batches. The measurements were conducted using GreenFeed system (C-Lock Inc., Rapid City, SD). The unit was installed in an empty pen where cows from the tie-stall were moved to the pen for methane measurement. Cows were let to visit the unit voluntarily attracted by the protein pellet dropped from the unit upon visit. The experiment was conducted for 2 and 3 days for the first and second batch respectively. Feeding was conducted once per day at 0800 h. Enteric methane emission measurements were monitored at 8 time points in the diurnal cycle (0200, 0500, 0800, 1100, 1400, 1700, 2000 & 2300 h). The result indicated that the average peak methane emission was observed around 1200 h which is 3 - 4 hrs after feeding. The minimum methane emission was observed right before feeding. Both batch 1 and 2 animals showed similar pattern of diurnal methane emission with an abrupt increase after feeding. Taken together, enteric methane emission studies in dairy cattle need to consider the diurnal variability of methane flux.  



22. Genome-wide association study for temperament score in Nellore cattle


Valente, T.S.1, Baldi, F.2, Sant’Anna, A.C.2, Albuquerque, L.G.2,3, Paranhos da Costa, M.J.R.2,3


1Programa de Pós-Graduação em Genética e Melhoramento Animal, Faculdade de Ciências Agrárias e Veterinárias, Jaboticabal-SP, Brazil; 2Departamento de Zootecnia, Faculdade de Ciências Agrárias e Veterinárias, Jaboticabal-SP, Brazil; 3CNPq Researcher.


The aim of this study was to determine genomic regions associated with temperament score (TS), which is the most used method to evaluate temperament, in Nellore breeding programs. This method assesses the animal’s reaction in a pen after exiting the cattle crush, ranging from 1 (calm) to 5 (excitable). Phenotypic data were collected from 28,283 and full pedigree from 162,644 animals. A total of 1,405 animals were genotyped with BovineHD BeadChip. After quality control 455,374 SNPs and 1,384 genotyped animals remained. The association analysis was performed by single-step (ssGBLUP), a modification of BLUP with the numerator relationship matrix A-1 replaced by H-1 that uses the GEBV to estimate SNPs effects. The threshold animal model used to estimate (co)variance and genetic parameters included direct additive genetic and residual effects as random effects and contemporary groups as fixed effect. The SNP effects were calculated to segments of 10 sequential SNPs and results interpreted as the percentage of total genetic variance explained by each SNP window. Seventeen segments explained 1% or more of the total genetic variation and were considered as associated to TS on BTA1 (6Mb), BTA2 (90Mb), BTA3 (3Mb), BTA5 (85Mb), BTA6 (109Mb), BTA8 (106Mb), BTA9 (16Mb), BTA11 (76Mb), BTA13 (10 and 16Mb), BTA15 (3Mb), BTA18 (36Mb), BTA19 (13Mb), BTA22 (21Mb), BTA28 (30Mb) and BTA29 (23 and 25Mb). The biological function of genes within these regions may help to understand the individual variability of Nellore cattle temperament, contributing to improvement of genetic gain for this important trait.


23. Genome-wide linkage disequilibrium comparisons to detect selection signatures in pigs

Li, X.L.1,2, Tang, Z.L.3, Li, K.3, Yang, S.B.4, Wang, Z,Q.2, and Liu,B.1

1 Key Laboratory of Agricultural Animal Genetics, Breeding and Reproduction, Ministry of Education & College of Animal Science and Technology, Huazhong Agricultural University, Wuhan, China; 2Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB, Canada; 3The Key Laboratory for Domestic Animal Genetic Resources and Breeding of Ministry of Agriculture of China, Institute of Animal Science, Chinese Academy of Agricultural Sciences, Beijing, China; 4Department of Animal Science, College of Forestry and Biotechnology, Zhejiang Agricultural and Forestry University, Hangzhou, China.


The regions, suffered long-term natural selection and artificial directional selection in genome, showed low heterozygosity, long haplotype range and linkage disequilibrium (LD). Thus, comparing different patterns of LD among populations is one of the methods to identify candidate-selected regions. To detect selection regions in different pig populations, the varLD method was applied to compare the genome-wide LD variations in three pig populations, including 207 Chinese native pigs (nine breeds, CHN), 117 South American village pigs (SA) and 408 commercial Large White pigs (COM). These pigs were genotyped using the Illumina porcine SNP60K BeadChip. After quality control, 20,714, 27,278 and 32,323 SNPs were remained for selection signatures detections in CHN/COM, CHN/SA and COM/SA, respectively. The principal component (PC) and ADMIXTURE analyses indicated that these three populations almost have no gene flow among them. The LW showed much higher LD decay than the CHN and the SA pigs. Top 0.01 percentile of standardized varLD scores were characterized as the candidate selection regions, and 26 and 7 regions were detected in all comparisons for 0.01 and 0.001 percentile, respectively. Importantly, ADAMTSL3 gene (over top 0.001 percentile), which related to body measurement and skeletal muscle, were identified in CHN/COM and COM/SA comparisons. In the present study, new candidate selection regions have been identified among different pig populations using the fine mapped of LD variations method.



24. Developing mating allocation strategy including dominance effects in crossbred cattle

Li, X.L.1,2, Chen, L.H.2, Plastow, G.S.2, and Wang, Z.Q.2


1Key laboratory of Agricultural Animal Genetics, Breeding and Reproduction, Ministry of Education & College of Animal Science and Technology, Huazhong Agricultural University, Wuhan, China; 2 Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Canada.


Genomic prediction only considered the estimation of additive genetic effects in the prediction models while dominant genetic effects have been missed in most of the genomic selection program due to the difficulty of estimation. However, estimation of the dominance genetic effects in livestock breeding programs is also very important because it can not only increase the breading value prediction accuracy, but it can also be used in genomic mating program to capitalize the genomic heterosis in crossbred animals to maximum the performance of production. In this study, a genome-wide genomic mating program was developed to including both additive and dominant allelic effects to predict best possible pair of parents combination to produce crossbred individual for maximizing performance of production. Phenotypes of 9 (birth weight, wean weight, average daily gain, feed intake, average back-fat, carcass rib eye area, carcass weight, lean meat yield and carcass marbling) traits collected from 788 genotyped crossbred beef cattle using Bovine SNP50K BeadChip were used to demonstrate the genomic matting program.

[1] Presenter

[2] Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Posted in Uncategorised