Tuesday, November 23, 2010

News from 23andMe - Consolidated offerings, Personal Genome Service, upgraded chip and possible sale

The DNA genealogy mailing lists and Twitter are all abuzz about 23andMe's new consolidated offerings and price structure, as well as the possibility of another $99 sale tomorrow.  Back in September, I wrote about 23andMe's new subscription plan - the Personal Genome Service here and here.  Not surprisingly, they appear to have changed their model to require this subscription for all new orders starting today. (It seems that 23andMe has taken a lesson from direct response marketing - where we used to say that product is King, but now we all know that continuity is King!) They have done away with the separate Health and Ancestry Editions and now will only offer what was formerly the Complete Edition for $499 plus a minimum one year subscription to their PGS ($5 per month). All customers who formerly had the separate editions have been automatically upgraded and will immediately have access to their raw data as well as all tools currently offered on the site.

23andMe has also announced that they have upgraded their testing chip to the Illumina OmniExpress Plus Genotyping Beadchip, which was originally released in January of this year and enhanced in March. Information on this chip (before enhancements) from the press release:
  • Premiere Genomic Coverage - Greater than 700,000 strategically selected tagSNPs provide genomic coverage up to 90% for Caucasian and Asian populations as assessed by the International HapMap Project.
  • Proven Data Quality - Industry-standard Infinium HD Assay affords greater than 99% average call rates and greater than 99.9% reproducibility.
  • Industry Best Throughput - With the iScan System, researchers can process in excess of thousands of samples per week.
Specifications for the original are here. 23andMe may have a custom designed version of this. The original press release described a chip with coverage of 733,202 markers, while the enhanced "Plus" version appears to cover greater than 900,000 markers. Either way, it is a significant upgrade from the previous 580,000 SNPs. (I am not an expert on this, so please refer to the original sources yourself for details and clarification; more here.)

[Update - From 23andMe's updated FAQs :  
The DNA chip that we use genotypes hundreds of thousands of SNPs at one time. It actually reads 1,000,000 SNPs that are spread across your entire genome. Although this is still only a fraction of the 10 million SNPs that are estimated to be in the human genome, these 1,000,000 SNPs are specially selected "tag SNPs." Because many SNPs are linked to one another, we can often learn about the genotype at many SNPs at a time just by looking at one SNP that "tags" its group. This maximizes the information we can get from every SNP we analyze, while keeping the cost low.
In addition, we have hand-picked tens of thousands of additional SNPs of particular interest from the scientific literature and added their corresponding probes to the DNA chip. As a result, we can provide you personal genetic information available only through 23andMe.
There is a list of the 733,202 non-custom markers here.]

Existing customers have the option to upgrade their results to this new testing platform for $89, but it also requires subscription to the Personal Genome Service. Existing customers will not be required to subscribe or upgrade and will still receive their updates as before. Not surprisingly, without the upgrade, some new information based on specific markers will not be available.

All week 23andMe has been tweeting about upcoming sales and one tweeter who may or may not have inside info tweeted this morning, "@23andMe $99 discount returns; code B84YAG to be live 10 AM Wednesday for the new v3 chip."  I imagine this sale price will require a subscription to the Personal Genome Service, but at only $5/per month, this is still a great deal.

I will post updates as I get them, but keep your eye on 23andMe!

**UPDATE - 23andMe is determined to keep the details secret until tomorrow. We will just have to be patient! Check back then for more...

Sunday, November 7, 2010

Genetic Genealogy in the News: Another misleading, inaccurate story

ABC4.com out of Salt Lake City recently ran an article about genealogy and DNA testing. I am always happy to see my hobby in the news, however, I have real concerns about some of the claims that are being made in this story. As a long time genealogist and experienced genetic genealogist, I find this story very misleading and, potentially, detrimental to the industry.

First of all, the article states that "Amanda Gilbert is the only person in the world who can claim accused witch, Rebecca Nurse, and persecutor, Reverend John Hale as grandparents."  Wow, really? These people lived in 1692. They could each potentially have thousands of descendants. A quick search of the Internet or Ancestry.com turns up many who claim descendancy from each Reverend Hale and Rebecca Nurse. I guess it is possible that Ms. Gilbert is the ONLY one descended from both Hale and Nurse, but that is quite a claim. Doesn't she have any siblings or children? Did the reporter do the extensive genealogical research necessary to substantiate this claim? Before I made such a far-reaching statement, I would be sure and do a thorough and extensive genealogical study of each of the subjects, following all descendants down to the present day. I could chalk it up to naivete and a lack of understanding of genealogy on the part of the reporter, but the problem appears to go deeper than that.

The article goes on to "quote" Scott Woodward, president of the DNA testing company GeneTree and principal investigator at the Sorenson Molecular Genealogy Foundation, “We can identify the pieces of DNA that belong to each one of those individuals in the past, and then we compare that to other people’s DNA and identify the great, great, great grandparents you have in common” and referring to Gilbert's DNA, “We put that into the data base and started looking around. She’s also related to the Reverend John Hale.”

I am simply dumbfounded by these statements. It certainly sounds as if he is referring to autosomal DNA testing. As a Beta tester for the only two companies who have introduced commercial relative matching through autosomal DNA testing so far (23andMe and FTDNA) and an active member of the genetic genealogy community and ISOGG (International Society of Genetic Genealogy), I find these claims incredulous. There are many of us who have been working long hours studying our autosomal DNA results, attempting to determine which pieces of DNA came down to us from which ancestors, but it is difficult and slow work. This type of DNA research is new and, to my knowledge, none of us yet claim to be able to definitively identify the blocks of autosomal DNA from specific ancestors more than a couple of generations back. There may be a few on the forefront of autosomal DNA family studies who, in limited examples, appear to have strong evidence that links blocks of DNA to a specific ancestor, but they stop short of making absolute claims at this early date. I have confidence that, in the future, we will be able to accomplish this, but not until MANY more people are tested with solid genealogies. To make matters worse, the way it is written, Woodward seems to be claiming to be able to identify DNA from an ancestor who lived more than three hundred years ago! The companies and the scientists leading the way in this new science (autosomal DNA testing), are tentative about predicting shared ancestry more than a couple of hundred years in the past at MOST. If GeneTree has evidence that they can do what Woodward seems to be claiming, there are many of us who would love to see it and will happily and enthusiastically congratulate them on their accomplishment.

Since, according to their website, GeneTree doesn't even do commercial autosomal testing, perhaps Mr. Woodward is referring to Y-DNA testing. In that case, his claim that DNA showed that Ms. Gilbert is descended from Rev. Hale may be possible, but it couldn't have been her DNA. Maybe Ms. Gilbert has a male relative who was tested by GeneTree and whose Y-DNA matched those Hales known to descend from the Reverend. It is certainly possible that the reporter may have misunderstood and taken Mr. Woodward's statements out of context. If this is what he meant, then more explanation was certainly needed. Perhaps GeneTree can clarify this. I know from personal experience with the press that what you say is not always what they write, so we should give Mr. Woodward and GeneTree the benefit of the doubt in regard to these seemingly outrageous claims.

Whether intentional or not, these types of public statements could spell big trouble for the future of DNA Ancestry testing.  One always runs the risk of being misquoted by the press, therefore DTC genetic testing companies would be prudent to err on the side of caution in their claims and interviews. As we all know, regulatory agencies have their eye on these companies. If those of us involved in the industry make what appear to be unsubstantiated or misleading statements in regard to what can and cannot be accomplished with DNA testing, it gives these agencies an excuse to interfere, even in ancestry testing. While we cannot always prevent how we are portrayed in the press, we can be very clear about the current limitations of DNA Ancestry testing. Any time we speak to the press, we need to keep in mind that the reporters writing about our complex industry may not have the necessary expertise to fully understand what we are saying, so we should make a concerted effort to educate them. When and if an article is published that contains inaccuracies or misstatements, it must be our responsibility to publicly refute it.  In order to minimize the risk of misinformation and misleading claims, we need cooperation and industry-wide standards among DNA Ancestry testing companies. It is my fear that if we do not take proactive steps toward some form of self-regulation, we are inviting outside interference.

** Please read the update in the comment section below from Dr. Ann Turner.

[Disclosure - My company StudioINTV has an existing production agreement with FTDNA that has no bearing on the opinions I express. I receive no other compensation in relation to any of the companies or products referenced in my blog.]  

Saturday, October 9, 2010

Ancestry.com's New Feature "View Relationship To Me"- a time saver for genetic genealogists

Have any of you noticed the new feature on Ancestry.com's Member Trees called "View Relationship To Me"? I just happened to come across it today while investigating a match from Jim McMillan's spreadsheet. The new match and I share the surname Eastman, so I was looking at the immigrant Roger Eastman on my Ancestry tree to determine our connection and I noticed a box on his profile page that said, "View Relationship To Me". I clicked on it and it showed me the entire line from Roger down to me, telling me that Roger is my 10th great grandfather (and, it turns out, our connection)! I used to have to do this manually by counting down the generations. (Sometimes I would even get lost if I had listed the siblings in each generation.)  Now, with one click I can see how anyone in my tree is related to me. This is a really great addition and definitely a time saver for those of us working on finding our connections to our DNA matches. There is also an option to add the relationship on your relative's profile just below their name, so anyone looking at your tree can see how you are related to a specific ancestor. If widely utilized, this may cut down on the "How are you related?" messages on Ancestry that so many of us receive (and send).
There is a post about this new feature on the Ancestry Blog. Some readers have noted that they could do the same thing with their genealogy software, but I had never seen this before. Have you?

Sunday, September 26, 2010

Known Relative Studies with 23andMe: Great Grandchild DNA Inheritance

Great Grandchild Inheritance Pattern

I am very fortunate to have the opportunity to see the inheritance pattern of a great grandchild in my known relative studies at 23andMe, so I decided to share it. For those of you who are not familiar with the 23andMe user interface, the above chart is an illustration of the 23 pairs of chromosomes that we all possess. For simplicity, only one chromosome is displayed to represent each pair, thus the 23 bars. Using the Family Inheritance Advanced tool one can choose to compare selected individuals and a chart is generated to illustrate the shared DNA between them.

In this chart, the light blue is the shared DNA with the mother, the light green is the shared DNA with the grandmother and the dark blue is the shared DNA with the great grandmother. Of course, the parent and child (light blue) share DNA across all 23 of the chromosomes, as would be expected. This represents the 50% of shared DNA between the child and parent. The grandparent and child (light green) usually would share approximately half as much DNA with the stretches being broken up into smaller blocks. In this case, the child inherited a larger than expected amount of shared DNA with the grandparent at 31.54% (expected ~25%). The great grandparent and child (dark blue) should share approximately half as much DNA as the child and the grandparent, with the blocks broken up into even smaller fragments of shared DNA. In reality, the percentage is less than half at 14.75% of shared DNA, which is still rather high compared to the norm (expected ~12.5%).

As you can see, the percentages will vary from the expected values. Notably, 31.54% is the highest sharing I have seen between a grandparent and child in my research so far. Since the child inherited 50% from the maternal side, there is only ~18.46% that could have been inherited from the other maternal grandparent, which is well below the expected 25%.

Notice that as the relationship gets more distant, in general, the shared blocks of DNA get smaller, disappearing completely on some of the chromosomes. On Chromosome 13, there are no stretches of shared DNA (that meet the threshold for this tool) between the grandparent and child. This means that on Chromosome 13, this child has inherited significant DNA from the other grandparent's ancestors (in this case, the maternal grandfather). Following this same pattern, on Chromosomes 6, 13 and 14, the great grandparent and child do not have any significant blocks of DNA. Also notice on Chromosome 10, the grandchild inherited the entire maternal grandmother's chromosome (light green across the entire bar). That means that Chromosome 10 has no (significant) DNA from the maternal grandfather.

Please remember that, in this case, we are only looking at half of the child's chromosomes. The child has another set of 23 chromosomes inherited from the father. Since we are not comparing the child with any paternal relatives, none of those chromosomes are represented in this chart or analysis.

**For more posts on my family studies, please see here. **

Known Relative Studies with 23andMe: Expected Percentages

I have had quite a few known relatives test with 23andMe recently. In the coming months, I will be writing a number of posts based on these results. As an introduction, I wanted to review some of the information that 23andMe and FTDNA provide in their FAQs regarding their autosomal DNA tests, Relative Finder and Family Finder.  All of these percentages are estimates and will vary due to the random nature of genetic inheritance. The following statistics apply to both 23andMe's Relative Finder (RF) and FTDNA's Family Finder (FF).

Expected percentage of shared DNA between:
Parent/child/siblings = 50%
Grandparent/grandchild/aunt/uncle/nephew/niece/half-siblings = 25%
1st Cousins/great-grandparent/great aunt or uncle/grandnephew or niece = 12.5%
1st Cousins once removed = 6.25%
2nd Cousins = 3.125%
2nd Cousins once removed = 1.563%
3rd Cousins = .781%
4th Cousins = .195%
5th Cousins = .049%
6th Cousins = .012%
7th Cousins = .003%
8th Cousins = .001%

The likelihood of detecting shared DNA with a known relative significant enough to show up in RF or FF:
1st Cousins and closer: 100%
2nd Cousins:  > 99%
3rd Cousins:  ~ 90%
4th Cousins: ~ 45% (FTDNA says > 50%)
5th Cousins: ~ 15% (FTDNA says > 10%)
6th Cousins and more distant: < 5% (FTDNA says < 2%)

In some families, there will be situations that will complicate the above predictions, such as cousin marriages and half-siblingship, but, in general, they are the guidelines that should be used in analysis and comparisons.

Further reading from this series -
Known Relative Studies: Great Grandchild Inheritance
A Success Story and the Randomness of Autosomal DNA Inheritance (Fourth and Fifth Cousins)
Known Relative Studies: Second Cousins 
Known Relative Studies: More Second Cousin Comparisons 
Known Relative Studies: Second Cousin Comparisons, Allen Great Grandparents
Known Relative Studies: Second Cousins or Half Second Cousins 
Known Relative Studies: Identifying DNA from Great Grandparents Using Second Cousin Comparisons
Known Relative Studies: I Found My Third Cousin Today! 
Known Relative Studies: A Third Cousin Comparison and More Random Autosomal DNA Inheritance
Known Relative Studies: Purdy Fourth and Fifth Cousins
Autosomal DNA Matching and the Importance of Testing Multiple Family Members (Ninth Cousins)
Ratekin Seventh Cousin
A Second Cousin Adds to My Chromosome Map and Answers a Nagging Genealogical Question

Sunday, September 12, 2010

"Ashkenazi and Me": Making a case for the Euro DNA Calculator application using ancestry analysis tools from 23andMe

Some time ago, I ran my raw 23andMe data through the Euro DNA Calculator Application. (Euro-DNA-Calc, compiled by Dienekes Pontikos, computes an admixture estimate with Northwestern European, Southeastern European, or Ashkenazi Jewish parental populations. The calc uses 300 markers from a scientific study.)  I was surprised to find that it estimated 86% NW European and 14% Ashkenazi for me.

My Euro-DNA-Calc Chart

Interestingly, my mother showed no Ashkenazi at all in her results with a 100% NW European prediction. Since I have no known Jewish heritage, after some research into the application, I disregarded it.

When 23andMe's Ancestry Finder was launched and began noting Ashkenazi segments, I noticed a spot on my Chromosome 7 that appeared to have a clump of self-declared Ashkenazi matches. When 23andMe started providing the download of the matches associated with Ancestry Finder, I saw that I had no less than 21 "Ashkenazi" matches on Chromosome 7 between 53.3m and 94.3m. My mother had none. These matches all appeared to be rather small and only on that one spot, but it did get me thinking about the possibility of distant Jewish ancestors again.

My Ancestry Finder chart set at 1 Ashkenazi grandparent

Since my father is deceased, I asked my paternal uncle to test in his place. Yesterday, I received his results. There, clear as day, on his Chromosome 7 appears that same Ashkenazi clump that I have (plus a little more: he has 37 self-described Ashkenazi matches on Chromosome 7 on the Ancestry Finder download).

Paternal Uncle's AF chart set at 1 Ashkenazi grandparent
Paternal Uncle's AF chart set at 4 Ashkenaki grandparents

Equally as significant, I could immediately see that he has a number of Public Matches in Relative Finder who list their ancestry as Jewish. Most of them are predicted as "Distant Cousin", which probably explains why they didn't show up in my Relative Finder. In this case, testing just one more generation back revealed very useful information.

While this Ashkenazi ancestry is clearly quite distant, it certainly appears to be authentic after all. Although the Euro DNA Calc greatly overestimated the likely percentage, it does seem to have picked up on legitimate Ashkenazi markers in my DNA.

Bravo, Dienekes!

Friday, September 10, 2010

23andMe's Relative Finder - Matching a cousin on both sides of my family

I had long suspected that Cousin E and I were related on both sides of my family, but didn't have proof until today. On Relative Finder, I match Cousin E on two segments, but she only matches my mother on one. The other segment was a fairly small one (7cMs), so I thought it could possibly have just been missed.  I was curious and wanted to find out.
My father is deceased, so I recently asked my paternal uncle to test at 23andMe in the place of my father. Today, I received his results. Sure enough, he too matches Cousin E!


The dark blue on the chart above signifies the areas that my DNA matches Cousin E. The green is where my mother matches E and the light blue is where my paternal uncle matches E. As one can see, my uncle and myself share the match with E on Chromosome 5 (represented by the bar with the number 5), and my mother and I share the match on Chromosome 7 (represented by the bar with the number 7). Therefore, the matching segment on Chromosome 5 was inherited through my father's ancestry and the match on Chromosome 7 was inherited through my mother's ancestry.
Cousin E and I have not determined our exact connections yet, but we do share Finnish ancestry from the same area (my mom's side) and New England Colonial ancestry (my father's side), so the matches do seem to make sense. We are predicted to be 4th Cousins (3rd to 5th) based on our two matching segments and .37% shared DNA. Since the two segments are obviously inherited from two different ancestors, our likely connection is further back. My mother is predicted to be a 4th Cousin (3rd to 7th) to Cousin E with .28% shared DNA and my uncle is predicted to be a 7th Cousin (4th - 10th) to Cousin E with .09% shared DNA.