Now this IS an interesting article regarding DNA!

Question

https://www.nytimes.com/2018/10/15/science/gedmatch-genealogy-cold-cases.html

This is an interesting article about the creators of GEDmatch.com.  A side article linked to this one points out that the case that set law enforcement on the path to using GEDmatch to solve old crimes involves a woman searching for her true parents who discovers she was probably kidnapped by the man who she knew as her father. This was truly important to her and answered some long-term mysteries for law enforcement. I think that is what we are about in genealogy: solving mysteries.

Comments

It seems that if DNA sampling can be used to find evidence of criminal wrongdoing, it can or should be used equally to exonerate those who are truly innocent. I think most of us on Wiki are intelligent enough to understand the many ways law enforcement can use, and hopefully not abuse this valuable tool. Certainly, I would draw the line at allowing it to become wrongly used to influence health officials to do what the majority of civilized people consider morally unacceptable practice.

Answer 1

Wow, great article.  It discusses the controversies surrounding use of DNA info from GEDmatch with its founders.  Great reporting.

Answer 2

I enjoyed the background about the founders, but those guys aren't going to last forever - I wonder if they have a plan for someone to take over some day?

I really just don't get how they can do the stuff they say they can do with a DNA sample and GEDmatch. They make it sound like they can figure out who you are, just from finding a match to your 3rd cousin.

That's nuts! You can't even tell the exact relationship from DNA - just a rough estimate - and I know I have thousands of 3rd cousins. Before I put my brother's data on there, my closest relative on there was the biological daughter of a second cousin. Before I helped her find her biological parents, my match with her would have been useless, if not outright misleading. How would you even know she was adopted, without asking her?

I'd need a better explanation for how they do what they say they do, to believe some of those statistics.

Comments

Of interest may be Cece Moore's interview/profile on CBS's 60 Minutes tonight (October 21). It will be a full-length piece for the news "magazine"; Steve Kroft is the interviewer: 7:30 p.m. Eastern, 6:30 Central, but 7:00 p.m. Pacific. I'm certain we won't see a dive into centromeres and centiMorgans, but Cece is always eloquent and I'll bet we do get a general sense of how she proceeds with Parabon and the forensic work.

---

If you read some of the associated articles , you will see that the process is not simple. They have a professional genealogist build a tree backward using other data and then forward to focus on a single individual. So, it's definitely not a one-step process. I will look forward to watching that interview with Cece tonight.

---

I'm guessing it's pretty much the same as any genealogist building a descendants tree from their own ancestors, with the added twist that they can't write any intro emails like "Hi, I may be your third cousin or so..." Another added wrinkle is that they have the DNA, but not the identity, so they don't have a starting place, but they can use "alibis" to rule out the potential DNA owners.

Not easy by any stretch, but possible. I'm sure there are other cases that hit brick walls just like we do--the only reason we're discussing the Golden State Killer is because it actually worked.

---

Thanks for the heads up on 60 minutes Edison!

I admire the dedication and commitment of the Gedmatch team  and love the overall concept of bringing the various testing companies together in a database that offers such comprehensive comparision of chromosomal values. I don't get the same start and end point at FT and so triangulations are not possible, (to my knowledge), without Gedmatch.com. My hat is off to them, and a HUGE THANK YOU! for offering this as a free service!

---

For those that didn't see the 60 minutes episode - it is available to watch online  https://www.cbsnews.com/video/genetic-genealogy-tracing-family-trees-to-catch-killers-60-minutes/

---

"Of interest may be Cece Moore's interview/profile on CBS's 60 Minutes tonight..."  I watched it, with interest.  I do pretty much the same thing as that professional genealogist does almost every day, usually without the cooperation of my so-called cousins (presumably non-felons, but you have to wonder given their witness-protection efforts), and no one is paying me anything.  I need to figure out how to monetize this hobby.  

Answer 3

I find this part of the article very worrying:

"But many observers disagree. When any one person’s DNA can lead investigators to hundreds of a suspect’s relatives, the standard model of individual consent does not hold up, said Rori Rohlfs, a professor at San Francisco State University who has studied familial searches. She finds it ironic that police in California must get approval from a judge to search criminal databases for a murder suspect’s brother, but can upload DNA to GEDmatch to identify cousins without any restrictions."

Many innocent people are involved in the investigation and can become a target of wrong accusations. Also, as we all know, this isn't a 100% error free process, on the contrary, it's easy to do an error.

Just watch this TED talk: https://youtu.be/xclg8ikPAvI

Comments

But Andreas, I don't believe who you're related to is private information.  All GEDmatch can do is eliminate most people and help identify potential candidates as possible suspects.  Normal detective work has to be done from that point on.  Nothing about the GEDmatch info convicts the person.  Even if they want to check the possible suspect's DNA, they have to use normal law enforcement procedures, not GEDmatch.

I see it more like hearing that a guy in a red shirt robbed a nearby store, and when you look, you see 3 people in red shirts running away.  They aren't guilty, just a good place to start the investigating.

---

For anything to be 100% error free is mythological. Errors are made with every single existing manner of investigation and incarceration. A vast majority of them are correct, but sadly a few are erroneous. The directive is to use science to minimize those errors, by making the odds of making an error astronomical.

In regards to genetic science I'm not convinced that it's easy to err with the process they appear to be using. In fact I should think it's harder than with older methods, such as fingerprinting, and especially eye witnesses. 

Imagine you have an unknown suspect who committed rape in locations A, B, C, D, E, and F. And you have the assailant's DNA from semen samples. By comparing the assailant's DNA to the genetic data of many people, a person is located. Now, without any other data, I would agree - this is not enough to be sure. The DNA in the database isn't complete, and you can't decisively say, "this is our man." There is no 100% comparison available. How do they know the target isn't a sibling or cousin of the assailant? But suppose the suspect they discovered also happened to live or work in locations A, B, C, D, E, and F, and had the means, motive and opportunity, and left other evidence that narrows down the odds? And suppose that he left descendants from some of those rape victims whose mother lived at those locations? With process of elimination, the odds against making an error become extremely high. Some math can be applied to calculate those odds. 

Now I'm not sure who decides when the "line is crossed." I think there should be more transparency with the process and methodology so that the people whose DNA is being tapped are fully informed of how this is done, and have the option to decline participation in any database that is allowing its use for such process. 

But it seems to me that the benefits (capturing numerous real murders and rapists and getting them off the street) outweighs the risk (the inconvenience of having my DNA compared to an assailant's with the very unlikely scenario that a close match might be found erroneously). 

Answer 4

"Some GEDmatch users are family tree completists who collect cousins the way some people collect baseball cards."

This is me, lol :)

Answer 5

Very interesting article.  Thanks for sharing it Edie!

Answer 6

I want to say that I have yet to read this article.  I am responding to the post and some of the responses made.  I uploaded the ancestry DNA profiles of my husband, daughter and myself to GEDmatch years ago.  I heard from a few people because of matches.  On Dec 23, 2017, I received an email with the following subject line, "possible DNA match on GEDmatch - help requested."  The body of the email went on to explain that my husband's and daughter's kits matched an unidentified woman who died years ago.  I was skeptical at first because, unfortunately, this world is full of scammers.  I asked for information and researched to find out this was a non-profit group that is working to use DNA and genealogy to solve John and Jane Doe cases.  The group did not give me a lot of information about the Jane Doe but I had enough to do a little research to find out it was a woman whose body was found in the early 80s and that she was a victim of an unsolved murder on top of being unidentified herself.  I have been researching our family trees for years and I do follow even distant branches in hopes that it will help in my direct line research.  At any rate, my husband and adult daughter gave me permission to share their DNA matches on ancestry with the charity, provide them with tree information I had, allow access to our ancestry tree and help build out the tree on WikiTree.  On April 23, 2019, I received an email that the identity had been solved and thanking us for our input.  I had to wait three long months to find out more when it was shared with the media and we were provided a link to watch the news conference live.  My husband did not grow up near his mother's family and the little contact he enjoyed with them was almost altogether severed when his mother died when he was 8.  The tree, in part, was a way for him to get to know his mom and where she came from.  The Jane Doe was not as close of a relative as thought because it seems his family had a few generations where one family's children married siblings from another family which made their offspring double cousins with more shared DNA.  Anyway, he did not know the family but we were glad to be Jane Doe's family while they searched for family who knew her.  We are thrilled for any part we played in her finally having a name and any peace it brought to any living family members.  I also know regarding GEDmatch that rules changed and for this group and law enforcement to see your DNA kit you would have to opt in.  If you put your kit on GEDmatch before the changes, you have to login and and opt in for sharing.  I am sorry to bang on about this but thought I might have a different view.

Comments

I like your point of view, Melanie. Putting one's DNA out there in order to solve a mystery is a bit scary, though, not knowing what skeletons may fall out of the closet.

---

LOL believe me skeletons will fall.  I had a 1st cousin match awhile ago that was an adoptee who based on shared matches was the child of one of my paternal aunts or uncles.  My dad was one of 16 so no small pool and there were 3 living at the time.  No one said they knew about it and I asked my cousins and no one knew anything.  Today, I found another 1st cousin match.  This match is also on my dad's side so again I don't know how to help.  All I can do is welcome them as a cousin and share family info and photos if they want.  It feels like it isn't enough.  I have cousins who I have tried to get tested before and they don't trust how DNA could be used so they won't so I don't know how to get them anything more specific.

---

Yes, what do you do?! I think we need to accept the relationship with good grace and move on. Not everyone is gracious, though. Hurt feelings...scandal, etc. Nothing to do with the adoptee, of course.