Getting our Science On

I apologize that I haven’t had a new blog in quite some time. I have been busy completing the Nashville office move, visiting the Pacific Northwest Alpaca Association, hosting a face-to-face Board meeting here in Lincoln, and preparing for the Camelid Community Fiber Conference. At the same time, I have struggled with my blog this time around. This is the third blog I have written. I didn’t like the first two drafts I wrote this time around, so at least for now, they will go unpublished. Perhaps I will revisit those topics at a later time.

Over the past several weeks, AOA’s DNA testing and parentage validation process have been called into question by a few people. As such, I would like to take a moment to give the facts about AOA’s process.

First, let’s be clear about something; there is no shortage of people who have had some issue with a registration over the years. After more than 230,000 alpaca registrations and dealing with tens of thousands of different alpaca owners who are taking blood samples and sending them in, there of course have been alpacas whose registrations have not just sailed through without a problem. Considering the fact that around 15% of all first sires submitted don’t end up being the actual sire whose parentage is validated, it is easy to go out and say “Who has had a problem with an ARI/AOA registration?” That type of a question is vague enough that I would guess many farms out there could answer yes. That doesn’t mean that the problem was generated by the organization.

A better question might be “How many out there have ever had a registration issue that was ultimately never resolved, or in which ARI/AOA has totally given up on?” The answer to that question, at least during my 9 years here, is near zero. I can’t think of a time when we have given up trying to register an alpaca. There are times when we have to say that we need more information before proceeding, or times that the owner has decided they don’t want to spend any more time working on it, but not times in which we have refused to keep working on a registration. We will keep working a registration as long as a member would like, even though it is likely costing the organization resources that exceed the cost we charged to register the alpaca.

One of the things that may be difficult to understand when you are frustrated because a parent isn’t matching is that the AOA staff wants to register your animal. It is not to our benefit to prolong a registration, because it only costs the organization money. One of the criticisms that the staff gets sometimes is that we require retesting or DNA comparisons, just because we want to make more money. This comment is really disheartening for the staff who are truly your advocates and are working hard to register your alpaca as quickly as possible. Additionally, the price of retesting is $25, which is right at cost and in most cases there is so much additional staff time on these issues, that we likely are losing money on these.

Speaking of the testing itself, I have seen some combine DNA testing and Parentage Validation into ‘DNA validation’ and then say that we are doing this in-house by people without degrees. This gives the impression that our DNA testing is being done in-house, which is not and never has been the case. Saying ‘DNA validation’ is actually combining two terms that don’t go together. There is DNA testing, which is when the markers are identified, and then there is Parentage Validation, which is when the markers received from the DNA testing on the cria are actually compared with the markers from the dam and sire to determine if the parentage validates.

Our DNA testing is conducted by DDC Veterinary, a division of DNA Diagnostics center, which is one of the world’s largest human parentage providers in the United States. They conduct hundreds of thousands of human tests per year in the same lab and with the same lab procedures as our alpaca testing. I encourage you to read the article about DDC on our website. It can be found here: In addition, our testing is blind testing, which means that DDC doesn’t know who the alpaca is they are testing, or the farm that owns that alpaca. It is all done through the bar codes that you all enter into the system from the cards themselves.

Once the DNA testing is complete, DDC’s online system sends the DNA markers to AOA’s online system. Our system does a variety of checks to ensure the data is correct before it allows the information into our system. Once that information is in, the AOA online system immediately conducts all the Parentage Validation within the system itself. Parentage validation is black and white and the rules are programmed into the system to ensure there is no human error and to eliminate any potential for any sort of fraud. Essentially, not even I have the ability to register an alpaca if it doesn’t actually validate with its parents in the system. This is very important to us because it helps create even more confidence in our registration process by our members.

AOA (previously ARI) has always taken scientific testing very seriously, and has consistently tried to plan for the future by periodically increasing the number of markers tested for and utilized to validate parentage. Each time we have done so, we have evaluated how such a change would affect the membership, and the Board has tried to make changes that do not require retesting of all previously tested alpacas. Here is a timeline for changes in validation over the years:

  • Until 1998 — Blood Typing
  • 1998–2005 — 10-marker panel
  • 2005–2010 — 14-marker primary panel, 10-marker additional secondary panel if needed
  • 2010–present — 18-marker primary panel, six-marker secondary panel if needed
  • Coming Soon — 18-marker primary panel, plus another 17 markers available if needed in additional panels

Throughout this history, each Board has understood that in order to continue to take advantage of new technology and provide the most accurate parentage validation possible, some retesting would be inevitable. The expansion of markers over the years has become important due to the fact that the Alpaca Registry is a closed registry and many alpacas are closely related. We could have continued to test for only 10 markers since 1998, but eventually it would have led to an increase in false validations because there would not be enough data to distinguish between closely related alpacas. The challenge has been to time the decision to expand the number of markers used in the panels well before we begin to have any issues. This strategy ensures that more alpacas will already have data from the larger panels prior to those issues arising, thereby reducing the volume of retesting required, lowering the potential for added costs to members.

The history of our DNA testing includes several testing labs in the more than 15 years that we have been conducting DNA testing. While the agreement with UC Davis did not specify the testing requirements of the validations during that time, we do know that many alpacas tested on the original 10 marker panel had fewer than 10 markers reported. The testing requirements with subsequent labs have been:

  • 2004–2005 — Ten markers were tested and it was required that a minimum of eight markers were able to be reported or the lab would request a retest.
  • 2005–2010 — Fourteen markers tested and it was required that a minimum of 12 markers were able to be reported or the lab would request a retest.
  • 2010–present — Eighteen markers are tested and it is required that a minimum of 17 markers are able to be reported or the lab will request a retest.

It is important to understand that in cases where fewer markers were reported than were tested for (e.g., eight of 10 or 12 of 14), these missing markers are not “lab errors.” In any rapid throughput DNA testing method there is the occasion that a marker does not produce enough “signal” to be measured by the instrumentation, and is therefore reported as “no data.” This is especially true in animal DNA testing, which doesn’t use as many repeats as something like forensic human testing. The missing markers simply mean that these don’t amplify in a particular test, and therefore are not able to be reported. This absence of markers can also be affected by the sample itself, which, in the early years, was more problematic until members learned how to use the DNA blood cards. These cards are highly effective and have been proven to be just as accurate as whole blood and follicular (root ball) testing. The blood cards utilized for ARI testing provide a larger amount of DNA than follicles, are much more convenient than whole blood, and are far easier to store for archival purposes. In fact, similar cards are utilized by crime labs around the world and by the United State military for human testing.

We currently test for up to a total of 24 DNA markers (primary and secondary) in order to complete parentage validation, and ARI has by far the largest camelid DNA database in the world, with approximately 230,000 animals with DNA profiles. As a result, our DNA testing laboratories have been able to utilize our vast amount of data to make adjustments to improve both the accuracy and the robustness of our panel, which provides a much better and more consistent result for our members. In 2005, based on analysis, it was determined that one marker (YWLL44) was not a very informative marker. This means that the marker had only a small number of alleles occurring in very high frequency in the population, therefore rendering it of little value to differentiate between animals.

With this information, it was decided to drop that marker and replace it with one that would be more informative. This change was not made casually, for the Board of Directors at that time knew that removing YWLL44 meant a large set of alpacas would now have that marker removed from consideration in parentage validation. However, because that marker was less informative, it was determined that it would be better to switch it out immediately, rather than having the issue compound itself over subsequent years.

After the DNA test, the Parentage Validation process must occur as discussed at the top of this blog. While all alpacas at the time YWLL44 was dropped were being tested on a 14-marker panel, this change meant that alpacas which had previously only been tested on the original 10-marker DNA panel could have as few as seven markers for comparison in Parentage Validation. When the 10-marker panel was utilized, the rule up to that point for parentage validation was that at least eight markers must be held in common (dam, cria, sire) in order to confirm parentage. With eight markers in common, the Registry allowed one marker to mismatch and still validated the parentage.

With this concern in mind, the lab was consulted, analysis was done, and it was determined that seven markers could be allowed, and still provide an accuracy exceeding 99 percent reliability, even when choosing the worst case scenario of the 7 least informative markers. However, to guarantee that reliability, no mismatches would be accepted when there were only seven markers in common. Keep in mind that this only occurs if the dam, cria, and sire only have seven markers in common. This included only 5 alpacas in the past 2 years out of almost 25,000 registrations. In all other cases, the regular rules apply. An alternative approach would have required the retesting of thousands of alpacas at great expense to breeders. While some sires and dams still have to be retested from time to time because they don’t have a sufficient number of markers in common with the cria, that number has been greatly reduced by using this approach. Since 2005, when this analysis was originally done on our parentage validation rules, it has been recalculated several times using the updated information in the database and the result has not changed. We still remain at about 99.87% accurate, even in the worst case scenario.

All testing and parentage validation rules that we use are periodically revisited. The parentage validation rules (accepting a seven-marker match) were revisited in 2009, 2013, and 2014. During each of those discussions, using updated statistics and after having reviewed the number of alpacas that could be affected, the Board determined that since the current policy still provides a sufficient level of accuracy, we would continue to use the current rules.

Lastly, over the years, every ARI (and now AOA) Board has to re-examine the parameters surrounding ARI’s tradition of rigorous scientific testing. There are three primary considerations that play a role for any livestock registry when looking at testing. They are cost, quality, and turnaround time. Each one affects the other and livestock registries around the world do their best to choose a balanced approach that allows for good quality, good turnaround time, and a reasonable cost. If we try to adjust one of these three, the others are also impacted. For example, increasing accuracy even closer to 100 percent could be achieved, but that would significantly increase the cost of a test, and turnaround time may increase by many weeks, perhaps even more than a month. Neither of those would be acceptable to most AOA members. As a result, we have selected a level of cost, quality, and turnaround time that is appropriate and reasonable to our members.

To conclude this WAY too long blog, I just want to reiterate that we are here to help you. AOA is a non-profit that is essentially controlled by all of you as members. We have nearly 10,000 active members at any given time and each of you has a different set of priorities, so we work hard to find middle ground to be sure we can address as many as possible. This is not easily accomplished, but is a great goal to have. However, the really cool thing about AOA is that the members determine the direction and if at any time you aren’t happy with the direction, you can ask us to change it or you can vote new Board Members in to run the organization, or even run for the Board of Directors yourself. The Board Members are alpaca owners with a desire to be able to utilize their end product, just like you. The staff is dedicated to the success of you as members, because when you succeed, the organization succeeds, which helps ensure stability in our chosen profession. We are all trying to do things and make decisions that positively affect your experience in the industry and with AOA. There is absolutely no incentive for us to make things more difficult for members. At the same time, ensuring the integrity of the organization and its Alpaca Registry is paramount and is directly linked to the success of our members and the industry as a whole.

The AOA Board and staff are committed to operating the organization in the most responsible way possible. It is not always easy making decisions that will affect things for years to come. Every Board member takes their responsibility to the organization and the industry very seriously. They, too, are alpaca breeders and are affected by the decisions they make. Over the years, the Registry has taken advantage of developments in science to ensure accurate validation of parentage. No doubt new technologies will become available for AOA’s consideration to perpetuate the robust pedigree database achieved thus far. There will always be a balancing act to weigh the benefits of those new technologies against the accuracies they provide and the cost to breeders, but the Registry will always be focused on ensuring its process is the most robust in the world.

Additional Resources: