Incorporating Genomic Information in Routine Genetic Evaluation

Dinesh Thekkoot PhD


For many years, pig producers depended on physical characteristics and phenotypic evaluations to select pigs to produce their next generation. Later on, breeding and cross-breeding methods, using quantitative genetics were employed to estimate an animals genetic merit more accurately.

By late 1980’s the genetic progress increased its pace with the introduction of a new statistical procedure called best linear unbiased prediction (BLUP), which allowed breeders to utilize performance information from many animals and to combine it with pedigree information (genetic relationship between animals). With the advent of marker assisted selection (MAS), breeders started to combine information from molecular markers with the traditional quantitative procedures. 

The swine industry also benefitted from individual DNA tests, to screen animals for various markers like RYR1 (PSE – pale soft exudative meat), MC4R (growth rate), ESR (litter size), PRKAG3 (meat quality), etc. Other than a few specific examples, most of these markers explained only a very small percentage of genetic variation of these traits, and suffered from practical difficulties in integrating with regular genetic evaluation procedures. Most of these shortfalls were addressed in a method called Genomic Selection (GS) first proposed by Meuwissen et al. (2001). But the technologies and information required for the practical implementation of GS were not available at that time.

Three main subsequent developments made the practical implementation of GS possible.

  1.  Sequencing of animal genome and identification of millions of single nucleotide polymorphisms (SNP), 
  2. development of high throughput genotyping technologies for cost effective genotyping of thousands of SNPs, and 
  3. development of new statistical methods to estimate and incorporate SNP information in routine genetic evaluations (Samore and Fontanesi, 2016).

These advancements resulted in the first practical implementation GS in 2009 in dairy cattle in the USA.

The incorporation of genomic information into the routine genetic evaluation in pigs was made possible with two further technological breakthroughs;

  • development of first commercial SNP panel for high throughput genotyping in pigs, and
  • sequencing of the pig genome in 2010.

Following this, various genotyping companies released SNP panels ranging from 10k to 650k for genotyping pigs. Genesus currently uses a custom-made SNP panel from Affymetrix that contains more than 55,000 SNPs for routine genotyping. Various statistical methods were also developed during this period for incorporating genomic information into the evaluation. Legarra et al., (2009) and Christensen and Lund (2010) developed a method called Single step genomic BLUP (SSBLUP) that utilizes genomic along with pedigree and performance information to estimate the genetic merit of selection candidates. Currently this method provides the highest prediction accuracies for estimated breeding values (EBV) of young animals, and Genesus has implemented this in the routine genetic evaluation. An increase in accuracy of prediction results in an increased genetic gain per year. The accuracy of EBVs and its impact on genetic gain was discussed in detail in a previous article and it can be read here.

Genesus has been conducting research in the field of GS from 2011 onwards. We studied each trait in detail, by simulating various GS selection scenarios, mimicking real population structures and data. These studies helped us to predict and analyse the advantages and potential problems that we may face during the implementation of GS. Our studies to validate the increase in accuracies using SSBLUP has shown that by using SSBLUP, the average prediction accuracies for growth traits (age to 120 kg, back fat and loin depth etc.) increased by 71%, 83% and 76% for Yorkshire, Landrace and Durocs respectively. Similarly, in Durocs, the average accuracies increased by 44% for meat quality traits (color, marbling, pH etc.) and by 88% for carcass traits (carcass back fat, carcass loin and hot carcass weight).  These increases in accuracies will result in Genesus Durocs continuing to excel in growth, efficiency, carcass and eating quality traits. For maternal breeds (Yorkshire and Landrace), by using SSBLUP the average litter size at birth accuracy increased by a factor of 67%. All these increases in accuracies will be reflected in our genetic response and contribution to maximising profitability for Genesus customers.



Christensen O. F., and Lund, M. S. (2010). Genomic prediction when some animals are not genotyped. Genetics Selection Evolution. 42, 2

Legarra A., Aguilar I., and Misztal I. (2009). A relationship matrix including full pedigree and genomic information. J of Dairy Sci. 92: 4656 – 4663

Meuwissen T. H. E., Hayes B. J., and Goddard M. E. (2001). Prediction of total genetic value using genome-wide dense marker maps. Genetics. 157:1819–1929

Samore A. B., and Fontanesi L. (2016) Genomic selection in pigs: state of the art and perspectives, Italian Journal of Animal Science, 15:2, 211-


Categorised in: ,

This post was written by Genesus