Expected Progeny Differences (EPDs): What are they?

Current genetic evaluation of livestock is based on the concept of Expected Progeny Differences (EPDs). An EPD is the predicted performance of future offspring of an individual, and is expressed as a deviation from the mean of the base group of animals. EPDs are based on direct measures of an animal's performance along with measures of the performance of an animal's relatives, including ancestors, siblings and progeny. These are adjusted for the differing amount of information available for each animal (number of records, number of progeny and the structure of the contemporary groups they are in, number of relative records, heritability of each trait, etc.). The evaluations in this report are based on a multiple-trait animal model, with separate models for terminal and maternal traits. Best Linear Unbiased Prediction (BLUP) genetic evaluation procedures are utilized to evaluate the information on each animal and the results are reported as EPDs. This allows for the direct comparison of all animals evaluated, and ranks the animals according to their genetic merit as parents, thus identifying genetically superior seedstock. Purebred swine breeders recognize that their product is breeding stock used in commercial swine production, and the genetic improvement made by breeders, is passed on to their customers resulting in more profitable pork production. NSR members, as well as their commercial customers, can make the fastest genetic improvement by utilizing the available EPDs in making their selection decisions.

Genetic Base Year

The genetic base of a population is the original group of animals at the initiation of testing and selection. The average EPD of this population will be zero, and this genetic base will be a point of reference for future generations. A genetic base year is therefore, the time period during which the genetic base animals are born, typically the span of 12 months. In the STAGES genetic evaluation program, a rolling or moving base year is currently in place, utilizing a four-year-prior base for maternal traits, and a three-year base for terminal traits. This base changes nightly, truly reflecting a dynamic population. For an example, we'll use Oct. 15, 2006, the date the information for this manual was collected as the date of genetic evaluation. The base year for this date for maternal traits will be Oct. 15, 2002, plus and minus six months, or include all animals born from April 15, 2002, to April 14, 2003. Terminal traits would then have a three-year-prior window of April 15, 2003, to April 14, 2004. The individuals born within these windows would have average EPDs of zero and average indexes of 100. The duration of the prior years accounts for a complete generation interval, allowing young animals to become parents with sufficient progeny for accurate evaluation. Figure 1 depicts these corresponding dates.

Figure 1.

So what does this rolling genetic base year mean? EPDs and indexes will remain in a relatively constant range for each breed. Breed means will hover close to zero for EPDs and 100 for indexes. Superior animals will be easy to identify because the numbers will be relative from year to year. A sire with a TSI of 130 is currently among the elite of that breed and will remain so in the future. This is based on comparisons to the existing population of animals within that breed.

With a fixed base, both individual EPDs and indexes were constantly moving in a positive fashion and because each successive year raised the breed average, it was necessary to know each breed's numbers to accurately assess the genetic merit of an animal. With a moving base, averages and ranges for each trait will be more consistent from year to year. This does not indicate that genetic progress has slowed, genetic trend lines in all breeds are constantly improving. Annual genetic progress is absorbed into the nightly base year adjustments, resulting in easier comparison of the numbers. This gives a more accurate view of an animal's relative breeding value within the breed, and allows breeders to implement a performance program with greater ease and accuracy. A producer who uses boars with a TSI of 130 can now continue to purchase animals or semen from animals with this value, and know that their genetic improvement is consistent with the progress of the breed. Previously, the breeder would have had to raise his minimum parameters on an annual basis, requiring new information with each purchase. This should make genetic progress easier for both breeders and commercial producers alike.

This Genetic Evaluation utilizes a rolling genetic base year and adjusts the base year each night, allowing more continuity in the numbers. Producers will no longer need to adjust for breed means in order to see where a particular animal ranks, and can more easily achieve a greater rate of genetic improvement.

Across Breed EPDs

This trait leader list includes animals from the Yorkshire, Duroc, Hampshire and Landrace herdbooks, the four breeds that comprise the National Swine Registry. The STAGES program is used for all breeds on our nightly in-house system, however EPDs cannot be compared across breeds. Many aspects of STAGES are the same for all breeds including the base year and general methodology, but data adjustments and variance components are breed specific. Because of the differences in genetic variation between breeds we cannot accurately compare EPDs from one breed to another.

Criteria for listing in NSR Sire Summary

You will notice that several of the lists do not completely fill the page. Previously, sire summaries had operated under less-restrictive guidelines regarding qualification to make these lists. Consequently, several of the previous lists contained sires that were deceased and no longer available to the industry, and had very few daughters in production. In order to better serve the purebred and commercial swine producers who utilize this publication, the Genetic Advisory Committee felt it would benefit all to include sires that could can still have a genetic impact, either directly or through multiple, active daughters. Thus the criteria for trait leader designation has become more stringent in recent issues of the Sire Summary. This tightening up of the specifications resulted in several of the lists finding fewer qualified sires.

In order for a boar to be recognized as a trait leader in the post weaning traits (Terminal Sire Index, Backfat, Days to 250 Pounds and Pounds of Lean), the requirements are:

1. Must have an "active" status, defined as siring a recorded litter in the past 12 months.
2. Must have 20 progeny with post-weaning performance data.
3. Must have a minimum of four pigs tested within the past six months.

For a sire to be recognized as a trait leader in the maternal traits and indexes (Sow Productivity Index, Maternal Line Index, Number Born Alive and Litter Weight at 21 Days), the requirements are:

1. Must have a minimum of 20 total daughter records (each litter produced by a direct daughter is a daughter record).
2. Must have 10 daughter records in the previous 12 months.

We also publish young sire lists for each of the breeds. The Duroc and Hampshire breeds publish Terminal Line Index listings, and the Yorkshire and Landrace breeds publish the promising sires in the Maternal Line Index. The purpose of these lists is to inform producers of potential future trait leaders in their respective breeds, with emphasis on the traits pertinent to each breed. Specifications for these lists are as follows:

Duroc and Hampshire Young Sires for TSI

1. Sires are less than 24 months of age (born on or after Oct. 15, 2004 in this summary).
2. Have an "active" status, defined as siring a recorded litter in the past 12 months.
3. Have five to 19 progeny performance tested.

Yorkshire and Landrace Young Sires for MLI

1. Sires are less than 36 months of age (Born on or after Oct. 15, 2003 in this summary).
2. Have an "active" status, defined as siring a recorded litter in the past 12 months.
3. Have five or more progeny performance tested.
4. Have one to 19 daughter records.

Genetic Trends

Evaluating average EPD genetic trends for several years can give us an indication of both breed direction and rate of improvement. As our genetic evaluation program improves methods of analysis and accuracy, the potential for genetic gain by a producer increases dramatically.

Breed genetic trend charts are presented for all traits and indexes in the preface for each breed. These graphs depict the improvement that each breed has made in recent years. Each year's data includes the average EBV for every animal born during that calendar year that has an individual record or is a parent or grandparent of an animal that has a performance record. Single trait selection can turn the slope of a graph upward rapidly, however often that selection method can create potential problems in other areas. The most profitable method of selection is to determine which traits are most important to your environment and market and concentrate on those areas, while keeping a balance in other economically important traits.

A Successful Performance Program

In order to best serve our commercial customers, purebred swine breeders must continue to make a priority out of producing animals with documented quality and predictability. The STAGES program gives us the methods and opportunity to do just that. While EPDs should not be the only criteria with which to make seedstock selection, they certainly represent a very valuable tool. Several items that breeders should address in order to operate a successful genetic evaluation program include:

1. Evaluate animals in proper contemporary groups. The basis of good genetic evaluation is having the ability to compare animals versus their contemporaries. A properly designed group includes animals that are: 1) the same breed or breed composition, 2) the same gender, 3) of similar age, and 4) have had similar care consisting of the same housing, nutrition and health status. Recommended parameters for post-weaning contemporary groups include animals born within a 45-day period, two or more litters sired by at least two sires and scan weight as close to 250 pounds as possible. Maternal contemporary groups include two or more litters sired by at least two sires and born within a 45 day window. Weaning age should be between 10 and 35 days.
2. Accurately collect and document performance information. Maternal information should be written down while still in the farrowing room, not when you retire for the evening. Post-weaning data is typically collected when animals are weighed and scanned coming off of a testing period.
3. Submit all data as soon as collected to the NSR office. Information that is collected and not evaluated benefits no one. NSR is equipped for rapid turn around of all information, but only if it is properly submitted.

Reading the Sire Summary and Explanation of Terms

1. Animal's registration number. The first six digits are the litter designation followed by the three digit earnotch number of the pig.

2. Official registered name of the animal followed by the name of the current owner. The animal's name starts with a herd prefix unique to each breeder, and ends with the pig's earnotch.

3. Sire and maternal grandsire of the animal.

4. The number of pigs and number of herds included in the growth data.

5. Backfat EPD. Backfat is measured in inches and adjusted to a 250 pound live weight by a NSIF equation which accounts for actual backfat, actual weight and the sex of the animal. Animals with negative (-) EPDs for backfat will produce offspring that have less backfat at market than offspring of parents with higher EPDs. In the example given, Big Dog should sire pigs with 0.10 inch less backfat than Sowmaker (-0.14 inches minus -0.04 inches equals 0.10 inch).

6. Days to 250 pounds EPD. Days to reach 250 pounds are adjusted by a NSIF equation that accounts for actual age, actual weight and the sex of the animal. Animals with a negative (-) EPD for Days will produce offspring that reach market weight faster than offspring of parents with higher EPDs. Big Dog's progeny should reach 250 pounds on the average of 4.80 days earlier than Sowmaker's offspring (-2.20 days minus 2.60 days equals -4.80 days).

7. Pounds of lean EPD. Pounds of fat-free lean adjusted to a 185-pound carcass or approximately a 250 pound live weight. The EPD of pounds of lean is calculated from the EPDs for backfat and loin eye area. A sire with a positive (+) EPD for pounds of lean will produce offspring that yield a higher percent of lean than offspring from a sire with an lower EPD for pounds of lean. Big Dog's progeny should have four more pounds of fat-free lean per 185-pound carcass when compared to Sowmaker's progeny (4.20 pounds minus .20 pounds equals 4.0 pounds of lean).

8. Terminal Sire Index. A bio-economic index that ranks individuals for use in a terminal crossbreeding system. TSI includes only EPDs for post-weaning traits. It weights the EPDs for backfat, days to 250 pounds, pounds of lean, and feed/pound of gain relative to their economic values. Each point of TSI represents $1 for every 10 pigs marketed or 10 cents per pig produced by a particular sire. Every 10 pigs sired by Big Dog should be worth $41 (or $4.10 per pig) more than a 10 pigs sired by Sowmaker when used in a terminal market hog production system [(136.5 TSI minus 95.5 TSI) times $1]. This is due to added growth and leanness.

9. Number of daughters and number of herds included in the maternal data.

10. Number born alive EPD. The number of live pigs farrowed in a litter adjusted for the parity of the sow. Daughters of sires with positive (+) EPDs for NBA will farrow larger litters than daughters of sires with lower EPDs for NBA. In the example given, Sowmaker's daughters should farrow one pig more per litter than Big Dog's daughters (0.90 pigs minus -0.10 pigs equals 1.0 pigs born alive).

11. Litter weight EPD. Litter weight adjusted to 21 days of age and adjusted for the parity of the sow and number after transfer. Daughters of sires with positive (+) EPD will wean heavier litters than daughters of sires with lower EPDs for litter weight. Sowmaker's daughters should wean litters that weigh 15 pounds more than the daughters of Big Dog (12.50 pounds minus -2.50 pounds equals 15 pounds).

12. Sow Productivity Index. A bio-economic index that ranks individuals for reproductive traits. SPI weights the EPDs for number born alive, number weaned, and litter weight relative to their economic values. Each point of SPI represents $1 per litter produced by every daughter of a sire. Each litter farrowed by a daughter of Sowmaker should be worth $25 more (or approximately $2.50 per pig) than a litter out of a daughter of Big Dog [(120.2 SPI minus 95.2 SPI) times $1]. This is due to more pigs at birth and heavier litter weights at weaning.

13. Maternal Line Index. An index for seedstock that is used to produce replacement gilts for crossbreeding programs. MLI weights EPD's for both terminal and maternal traits relative to their economic values, placing approximately twice as much emphasis on reproductive traits relative to post- weaning traits. Each point of MLI represents $1 per litter produces by every daughter of a sire. In this case, Big Dog and Sowmaker are nearly of equal genetic merit when used in a maternal line index, however each individual excels in different traits. It is important to analyze the needs and demands of your operation and select replacement animals accordingly.

STAGES Home Page