Big Data

By Nick Boddicker, Ph.D. Geneticist



Data collection and interpretation is a critical part to a swine genetics breeding program. Without data and its interpretation, genetic progress cannot be made. As technology evolves the ability to collect new data emerges, and a successful breeding company will capitalize on these new technologies. Furthermore, as technology evolves, the cost of the technology decreases. For example, the cost per gigabyte of storage in 1980 was ~$500,000. Today that cost is less than $0.03 (https://www.backblaze.com/blog/hard-drive-cost-per-gigabyte/). This is a huge cost savings as more and more data is collected.

In swine genetics, there have been waves of data collection. Basic data collection includes weight, total number born, number born alive, number weaned, wean to service interval, teat count, etc. These are traits that do not require sophisticated technology to collect. Ultrasonic backfat, loin depth, and intramuscular fat are now collected because of advancements in ultrasound technology.

Advancements in carcass and meat quality have emerged over the years, including equipment to objectively measure pH, marbling, color, and tenderness. All of these traits are important to eating quality. Genesus has over 820,000 carcass and meat quality measurements that have been collected since 1998, and this number continues to grow.

Another large data generator in the swine industry is individual feed intake and behavior. There are a number of systems available that will collect every feed intake event during the grow finish phase of production. To date, Genesus has over 43 million individual feed intake records during finishing. Furthermore, technology has been developed to measure feed intake during gestation or lactation. Genesus collects lactation feed intake as a component trait to lactation efficiency and currently has over 21 million feed intake records.

The most recent, and largest, data generator is genotype information that is used for genomic selection. Each animal genotyped receives ~55,000 data points. If 500 animals were genotyped every week, this would be over 27 million data points every week, or over 1.4 billion per year. Files with this much data cannot be opened or processed with typical software on computers, and therefore special programming must be developed by the breeding company to utilize this information.

Data has always been an integral part of a genetics program and the amount of data collected continues to grow exponentially. As technology evolves and becomes affordable, new measurements and data will be collected. There are a few areas that will be expanding in the future that will generate large amounts of data and these include epigenetics (study of heritable gene expression), proteomics (study of protein variants), and large-scale phenotyping. An example of a large-scale phenotyping program would be a small ear implant that collects body temperature data throughout the day of a pig. This data could be used to identify when a pig got sick, or information on energy expenditure, or heat detection for breeding. Another example is video recordings of pigs in a pen to monitor activity and behavior, which can be related to energy expenditure and efficiency. Most of us are aware of the amount of storage a 1-minute video clip consumes on our smart phones. Imagine the amount generated when filming 24/7 with multiple monitoring systems across several farms.

A successful breeding company will play an active role in keeping up with advancements in technology and utilizing new technology in their breeding programs. Genesus has always been aggressive with enhancing their breeding program through the use of new and enhanced technology to increase genetic improvement. These investments ultimately benefit the producers that use Genesus genetics.

Read more about our Genetic and Research Development Program

Share onShare on LinkedInShare on Facebook

Categorised in: ,

This post was written by Genesus