UPDATE: Found at the bottom of the article!
There are different ways of measuring DNA and the most basic is by counting the bp, or base pairs, within a segment. This method counts each nucleotide – A, T, C, or G – within a segment. This is fine for defining where gene locations or finding SNPs, but if you’re trying to determine how closely related you are to someone else, it’s not the best measuring stick. For that, you want to use a cM table. Scientists created the cM table to take into account how likely it was that a segment would be sliced up during meiosis. [If you need a refresher on cM, please see “Ancesty DNA – Why Low Confidence Matches Matter” By accounting for the likelihood of crossing over, you get a better idea of how many generations separate you from the most recent common ancestor, MRCA, of you and your DNA cousins giving a much better estimate of whether they are a 4th or 6th cousin.
Indeed, using the cM table gives you a better idea if someone is even your cousin at all! Especially in a genealogical timeframe (the last 500-700 years.) A 5 Mb segment can be much smaller or larger on the cM scale. Do you really want to have someone who is a population level match (ie you’re both from Switzerland) listed in your DNA cousins? On the other hand, you could have a 4 Mb match that shares 10 cM with you. You don’t want to overlook them!
Ancestry DNA and the cM Debacle
So why am I bringing this up again?
Kenny Freestone gave a presentation on AncestryDNA at the National Genealogical Society Conference and stated that they were using both cM and Mb for matches depending on when you were tested. This information came from conference attendee Angie Bush via Your Genetic Genealogist.
Freestone showed a slide with the following information when explaining how they determine the cousinhood of your Ancestry DNA matches:
200 megabases for 2nd cousins
150 megabases for 3rd cousins
100 megabases for 4th cousins
30 megabases for 5th cousins
20 megabases for 6th cousins
10 megabases for those further out
This slide raised a question as to whether or not AncestryDNA is using centimorgans or megabases in their matching algorithms. Kenny clarified that they are using a combination. They switched to using centimorgans in November – December 2013. If you tested recently, then your matches are in centimorgans. If it was prior to that date, then your matches are in megabases.
Since cM is much more accurate for making matches and determining the matching segment length, why should I — or you, as an early adopter of Ancestry’s DNA service, be penalized by having my matches poorly sorted into categories and indeed poorly identified using the Mb system? I’d really like to know if there is a way to request my matches be rerun using cM. As an adoptee, I really depend on that kind of information to try to figure out who my people are. Must I buy a new kit to get the better cousin list? That hardly seems fair. I realize that there is probably a fear of backlash when folks get a ‘new’ list (and lose some old matches) but that’s no excuse for forcing Ancestry DNA’s most loyal, early customers to accept a very flawed product!
It was also left as unclear whether, as a pre-Dec 2013 customer, I continue to get matches under the Mb system or if my new matches at the least are in the more accurate cM system? There seems to be some indication that newer matches are in cM and in fact, the mixture of cM and Mb matches is causing a lot of confusion especially with those loyal customers who’ve bought more than one kit because parents are showing cousins as more distant than their children.
I put the following comment on Ancestry DNA’s Facebook page.
AncestryDNA said at the National Genealogical Society Conference that they were using both cM and Mb for matches depending on when you were tested. Since cM is much more accurate for making matches and determining the matching segment length, why should I, as an early adopter of your service, be penalized by having my matches poorly sorted into categories and indeed poorly identified using the Mb system? How do I request my matches be rerun using cM? If I’m forced to buy a new kit to get my matches done correctly, it is certain that I’ll be blogging about the differences. I realize that there is probably a fear of backlash when folks get a ‘new’ list (and lose some old matches) but that’s no excuse for forcing your most loyal, early customers to accept a very flawed product! Also, it was left as unclear whether, as a pre-Dec 2013 customer, I continue to get matches under the Mb system or if my new matches at the least are in the more accurate cM system? How do I find that out? Thanks!
So far, I’ve only gotten a reasoned reply from one fellow user who said,
relax, they’ve said on multiple forums that they were going to rerun all tests as they update their matching formulas. They reran everyone’s ethinicity breakdown when they updated those reference populations.
I’ve asked her for a link to the statement. I’ll update the post if I get one [it's certainly not her job to provide one...I'm just hoping! ]
If you have an opinion on the cM matches at Ancestry, or information that might be helpful on what Ancestry plans to do, please feel free to post on their facebook page too.
The Ancestry Twitter account responded on Monday asking me to contact their customer service for help on my request to get my matches rerun in the cM system. I had time to call today and got a very pleasant person who knew a bit about the DNA tests because she uses them herself. Unfortunately, she had not heard about the fact that everyone tested prior to December 2013 was on the Mb system and more recent customers were on the cM system. I asked if they were going to convert everyone, and all she could tell me is that such an update isn’t imminent or they would have been informed of it. She left the research team feedback and offered me a survey so I could leave additional feedback.
The only other factoid that came up is that Ancestry is working on a method to transfer kits from one account to the other.
There was no official response on the Facebook page.
So is that the end of the story? Not by a long shot. A few of us will be investigating further so watch this space!