Friday, February 26, 2016

Comparing Match cMs At Different Sites

After a discussion at ISOGG Facebook I decided compare the data from matches who have results in multiple places including AncestryDNA, Family Tree DNA, 23andMe, and GEDmatch. I copied all my mother's match names from these site. I then sorted the names alphabetically. I found it was impossible to compare with AncestryDNA testers because most do not use their first and last names. Because so few testers use first and last names I was not able to use this method to find testers who were also in the other databases. It would be too time consuming to pick out those using their own first and last names. So I decided to do a more scaled down comparison using known cousins who have results in multiple places.

My results demonstrate that segment cM's are generally close to the same when comparing at Family Tree DNA, 23andMe, and GEDmatch. I did find a case where a segment cM's were 10 cM's apart between Family Tree DNA and GEDmatch. SNP totals at GEDmatch are often lower. Now I know to turn down the SNP totals when comparing at GEDmatch. I'll use 500 SNPs now.

Since AncestryDNA doesn't share their segment information I couldn't compare using segment totals. Instead I compared with cM totals. I didn't use segments under 7 cM's in the Family Tree DNA totals. It looks like GEDmatch always has the highest total cMs.  Ancestry always has the lowest. The average difference between Ancestry DNA and the other sites is 17 cM's. AncestryDNA phases and filters matches raw results, which is the reason for the differences in total cMs

Most of these matches are predicted in about the same cousin range at Ancestry and the other sites. The problem can be seen in my first chart. 23andMe, Family Tree DNA, and GEDmatch all show the person in chart one line 1 as a match. This person did test at Ancestry isn't a match with my mother there, even though she is a confirmed 4th cousin. I hadn't noticed until putting this together. I'm noticing more matches at the other sites who don't match at Ancestry. I have at least 5 confirmed cousins who did match at Ancestry, but don't now. Likely because of Timber. I'm not seeing this when looking at matches elsewhere. I'm sure some don't match at Family Tree DNA, but match elsewhere because of the 20 cM requirement. I have not encountered that because 1 cM segments are included.

Someone said if the results are different between sites what difference does it make? Ok, if each company has slightly different ranges but come up with the same matches then there isn't any problem. If confirmed cousin matches are being lost than I believe the companies should be rethinking their testing and matching procedures. Third cousins, and more distant cousins, are the ones affected by unreliable matching techniques. If a match shares only once segment they are more likely to be disappear as a match with additional processing.

Putting this together I have found more difficulties working with AncestryDNA than the others sites.
  1. Ancestry doesn't allow you to download matches or their cM numbers (I used the chrome extension. Doesn't include cMs). 23andMe and Family Tree DNA allow you to download spreadsheets.
  2. Ancestry should encourage testers to provide full names if they want to participate in sharing with other testers. I understand why some may not want to use their real names. They should use a consistent pseudonym, and use it everywhere, if they want to collaborate.
  3. It would be nice if we could filter matches by total cMs.  
  4. It would be nice if we could search by username.

Tuesday, February 23, 2016

Triangulation Example

Some ISOGG group members at Facebook have been wanting to see examples of triangulation at the 6th cousin level. My Melvin family segment triangulations would be closest to this cousin range. One match is a 5th cousin 1x removed, which is pretty much equivalent. This triangulation is with descendants of  John Melvin b. abt. 1776, Maryland and Mary Redden b. abt. 1777 Maryland. The Melvin segment matches are as follows (see chart above, which includes my Aunt, myself, and two other distant cousins):
  1. The light blue segment, on chromosome 1, represents my Aunt on my Paternal side. She shares this 22.1 cM segment with a 3rd cousin. This match is a descendant of our common ancestors John Melvin and Mary Redden.
  2. The light pink segment, on Chromosome 1, of the same size is my segment match with the same person as my Aunt. This is a 3rd cousin 1 x removed to me.
  3. The smaller dark pink segment sandwiched between the ones described above belongs to another John Melvin and Mary Redden match. This 14.2 segment is also shared by my Aunt and myself. This match is a 5th cousin to my Aunt, and a 5th cousin 1x removed to me.
  4. The green segment is where the 3rd cousin range match, to my aunt and myself, matches our 5th cousin range match. These 5th cousins share a slightly larger segment which is 18.6 cMs. You'll notice it extends passed the segments my aunt and I share.
Elijah Hicks and John Melvin sign
marriage bond
Both of our Melvin matches have good trees. Our 5th cousin range match has all lines going out at least 6 generations. Looking at other possible lines which may also be the source of these segments I don't see any other matching ancestors.

My proof of relationship to this Melvin family is through  US Census research, a bible record, and the Elijah Hicks and Nancy Melvin marriage record.

Examining whether these segments are likely IBD it would seem that they are in that cM range. Checking to see if my mother shares the same segment on chromosome 1 with all of these matches. No she doesn't match. You can see here my comparison between my paternal aunt with my mother. All of our Melvin matches matched between 165,698,481 to 180,598,459 on chromo 1,:

My Mom and my Paternal Aunt's shares in the same place as a 3rd cousin and 5th  Melvin cousin match

Looking a little more for possible places where our ancestors may have crossed paths I made this chart. Are we all from the same ethnic background? Could these be population segments? My paternal aunt and I have a fairly unique ethnic makeup. My 3x great-grandparents William Owens b. 1820 and Nancy Hick b. 27 Oct 1831 and their ancestors make up our only Colonial American line.

Here you see our lines are Austro-Hungarian, French Canadian, Colonial American, and Irish Catholic.  When looking at the places of origin for the Melvin matches of my aunt and I, we find both have quite a bit of Colonial Ancestry. I don't see any other shared ancestors between either of the other two testers. We all have Colonial Ancestry, but no other shared ancestors. My Colonial line on the paternal side is very small. Neither tester has French Canadian roots like my Aunt and I. Neither has Burgenland, Austria ancestry, as my Aunt and I do. They don't have Irish Catholic roots either. Our 3rd cousin match has a large Scandinavian line, which none of the rest of us have.

I think it's more likely than not that this Melvin Family triangulation is a good triangulation.

Some of those who would discredit triangulation would say, well it could just be a coincidence that we all match in the same place on chromosome 1. They would also say it's nearly impossible to share segments with cousins in that range. Chances of matching at all at that range are minuscule. It would be like being struck by lightening to triangulate at that cousin distance, so they would say. What are the chances we would all match in the same place and share the same ancestors? Wouldn't that be as unusual if you are sticking with statistical probability? I have a feeling we have a long way to go before we even really know what the statistical probabilities are? We aren't able to do enough comparisons, or look at enough possible triangulations to get an idea of how likely or unlikely they are to occur. A company is holding a huge amount our genome information, but they aren't sharing it with customers. They will sell genomes for medical research though.

PS This company now has no chromosome browser in 29 countries!

Wednesday, February 17, 2016

DNA On Fire AncestryDNA 4th Quarter 2015 Report

The Fourth Quarter, and 2015 full year report, at AncestryDNA emphasized the importance of the DNA product. This product has resulted in an increase in Ancestry subscriptions, which is Ancestry's core product. The 1 million new testers in 2015 helped increase subscriptions from 2,115,000  in the year ending December 2014, to the 2,264,000 in the year ending Dec 2015. An increase of  149,000. The subscribers who came to Ancestry through the DNA product are more engaged, and tend to subscribe to more expensive packages. They also tend to renew their subscriptions, according to Ancestry.

AncestryDNA now has 1.5 million testers in their database. The reason we are not seeing more tools like, a chromosome browser, is because sales are "on fire" according to one Ancestry Official. Black Friday 2015 sales were up 200% over last year. AncestryDNA has a lower profit margin than subscriptions, so as long as sales are brisk we won't be seeing new tools which would cost money to add.

The Ancestry Executives were also asked if the new medical focus has resulted in more hacking attempts? One Executive said he didn't want to divulge that information. Interesting at that point in the conference call the line suddenly went dead. I thought, were they hacked lol?

One Executive said a show Ancestry is sponsoring will likely increase DNA sales. Long Lost Family which will premiere its second season on TLC  March 6, 2016 will be sponsored by Ancestry. Sounds like it's based on a British show.

The sentiment regarding the DNA product's 2015 sales, and the current 2016 sales, has led these CEO's to forecast continued fiery sales of the DNA kits in 2016.

Monday, February 15, 2016

DNA: Not The Endogamy Of Cousins

We've been discussing endogamy in the ISOGG group. I took a poll at ISOGG a few weeks ago and learned that first cousin marriage was not uncommon in earlier times. This is true of certain groups. People living in isolated places with few prospective mates, for instance. I have not found that kind of cousin marriage in my family. Most of my lines go back 6-8 generations. My ancestors sometimes married outside of their ethnic group, and also married mates who came in different waves of immigration from the old world.

When I say not the endogamy of cousins I mean for most of us we aren't seeing any close cousin marriage out to 6 or 8 generations. We do see the affect of the smaller early American population. When my American ancestors came to America they initially settled in Pennsylvania and Maryland. They migrated from there to either the Midwest or South. Small populations in these areas, and common migration patterns,  could mean my matches' ancestors may have crossed paths with my ancestors more than once. I have found in a couple of instances that I could be related to a match through two different couples. This is a potential pitfall if you haven't carefully compared your tree with a match. This isn't the endogamy of cousin marriage, because it doesn't represent close cousins marrying, it's just that your ancestors crossed paths more than once when the population was smaller.

This is where mapping out chromosomes helps. Using the segments of matches, and your immediate family, going out to the 3rd cousin range you can begin naming your segments for family lines. I will never be able to do this for certain ethnic lines due to the lack of surviving records in the home country, so in that case I just name the segments according to ethnicity. Filling in the chromosome chart with named segments helps to identify matches who's segments overlap with confirmed family.

We are able to collect these segments at Family Tree DNA and 23andMe.  Family Tree DNA makes it easy by allowing us to see the segment information for all matches. At 23andMe you generally have to ask to share. 23andMe now has opt in sharing, which is working out for me better than expected.

At AncestryDNA there is no way to see the chromosome information. This creates a problem considering our ancestors may have crossed paths more than once. Without the possibility of mapping how do we know which or two, or more, couples we may have gotten our shared DNA from? AncestryDNA also has more of the segments we need to create such a map. I have closer matches there than at the other two companies. Their data could help me a great deal, and all of us.

Ancestry believes in DNA mapping because they recreated the genome of David Speegle using this technique. Some say we have segments going back to endogamy; if not more recently from America then going back to the old countries. These suppositions didn't seem to affect Ancestry's genome recreation?

Many will tell us compare at GEDmatch. Few of my matches have agreed to compare there. The process is confusing for those who aren't computer savvy. Others worry about the privacy of the site. The best solution, which would produce the most compliance, would be an opt in segment sharing system, like 23andMe.

Another problem with AncestryDNA is the problem plagued messaging system. If we don't hear from our matches it could be they aren't interested in sharing, or they didn't get the message at all?

Here is what Mapping can do for you:
  1. A well filled in map can help identify the ancestral route of a segment. This helps even when a match isn't cooperative.
  2. It can help to identify which of two or more couples a segment came from.
  3. It can help you eliminate IBS segments. You might find segments your parents don't match on.
  4. If you're using smaller segments as proof mapping can help confirm them.
  5. Ethnicity Chromosome Chart
  6. Chromosome matching segment maps can be compared to ethnicity chromosome maps to confirm ancestry. If you are 100% European the ethnicity chromosome chart won't help.
Without mapping we are hamstrung in certain situations.

This couple (below) now has 59 members in their circles. This could greatly help with chromosome mapping. Maybe Ancestry will sell us their genomes at some point?

Tuesday, February 2, 2016

Wife Of 3RD Cousin 5X Removed In An Ancestry Circle?

I misunderstood the Circles at AncestryDNA. I had thought they were reserved for direct line ancestors. Apparently they can include aunts, uncles, cousins and their spouses; if they are in your tree. I just found a Circle for a several times great-aunt. When I click on her Circle I'm listed as a potential descendant. This can cause confusion if you don't read all the descriptions carefully. I'm not included in the Circle though.

To me it would make more sense to include non direct line ancestors in NAD's. The Circles should form for the strongest links. If they are extended beyond that to the wife of a 3rd Cousin 5x removed, for example, then we are getting into some very weak associations. Couldn't ancestry just exclude certain relationships from Circles?