Focusing on William A. W. Dempsey’s DNA Using Chromosomes Analysis and Segment Maps

I took a break from blogging to give myself time to work on a DNA problem. It was only supposed to be for a few weeks, a month tops. Except for my article on the flooding in our part of Europe, I haven’t posted any new content to my blog in two and a half months.

As many of my readers know, I’ve been doing genealogy for nearly three decades and blogging in my eighth year. Writing for my blog has taught me to be a better researcher and writer.

DNA is complicated

This may be one of the reasons people who have their DNA tested are more interested in their ethnicity than in looking into who they got their DNA from. Many are not into genealogy or have the time to spend hours analyzing match lists or creating quick bare-bones trees (also known as Q&D or quick-and-dirty trees) for matches. In writing this post, I hope to reach some of my many distant cousins who could help me with my search.

Understanding where the DNA comes from

I’ve been working with my brother’s autosomal DNA results for over five years, my own for nearly two years, and my mother’s for a year and a half. All three were done with AncestryDNA.

Maternal Matches

Mom’s test has helped sort the maternal matches but wasn’t really necessary. My brother and I have few matches who are descended from our maternal lines as our mother is Luxembourgish – with all known ancestors coming from Luxembourg or parts of France, Germany, and Belgium that were once part of a greater Luxembourg. Close cousins (4th cousins or closer) on AncestryDNA total 375 compared to the circa 3,000 that my brother and I have. Many of the 275 are descendants of Luxembourg emigrants who settled in America. Our mother is their link back to Luxembourg and helps anchor their DNA.

Paternal Matches

My brother’s and my autosomal DNA results have confirmed the paper trail we have for our known paternal ancestors for at least six generations. For some branches in the tree, we have confirmation for nine generations or more.

Color groups on AncestryDNA

To better understand where the DNA comes from, I worked out a color/group system on AncestryDNA that goes back to the 6th generation ancestors (my paternal 4th great-grandparents). This helps to sort new matches.

Screenshot courtesy of AncestryDNA.

As the parents of my 2nd great-grandfather, William A. W. DEMPSEY are unknown, the first group is for the 4th generation ancestors. This allowed me to split the HONAKER-WISEMAN matches into two sub-groups: HONEGGER-GOETZ (as HONAKER was previously written) and WISEMAN-DAVIS of the 7th generation. As can be seen by the numbers in parenthesis, these are large clusters of matches.

Abbreviations:
PGF – paternal grandfather (blue)
PGM – paternal grandmother (green)
MGF – maternal grandfather (pink)
MGM -maternal grandmother (yellow)

Using colors in the family tree

The colors I use on AncestryDNA for the groups match the colors used in genealogy software charts.

The pedigree chart courtesy of Ancestral Quest 16

Mapping the DNA segments with GDAT

The same color system has been used to map our known DNA segments using the Genealogical DNA Analysis Tool or GDAT.

Genealogical DNA Analysis Tool (GDAT)

Becky Mason Walker’s Genealogical DNA Analysis Tool (GDAT) is the repository I use to manage my DNA tests.

The database is stored locally on my computer and has no connection to the internet. I can import DNA matches from the different testing companies, do triangulation and in common with (ICW) comparisons, map the chromosomes of common ancestors, mark the most recent common ancestors (MRCA), add Ahnentafels (tress) of the matches, and do analysis work that helps with the family tree research. The tool provides easier-to-see patterns and clues to solve the genetic genealogy questions with all information in one place.

Segment Maps

I’ve mentioned the color groups, Shared Clustering, and GDAT in previous posts.

Look Who’s Finally Taken the Autosomal DNA Test

Unraveling the Mystery of George W. Dempsey, son of Seaton Y. Dempsey and Clementine Gowing (Part 3)

Mapping DNA segments is something I haven’t written about.

GDAT automatically maps DNA segments when the MRCA (parental/maternal side and group name) is identified. GDAT chooses the color for the segment but allows the user to change it using a color picker.

Autosomal DNA Segment Map courtesy of the Genealogical DNA Analysis Tool (GDAT). Group names on right for MRCAs for surnames B-J.
Autosomal DNA Segment Map courtesy of the Genealogical DNA Analysis Tool (GDAT). Group names on right for MRCAs for surnames K-W.

The DNA segment map shows the paternal (top) and maternal (bottom) sides of each chromosome. In the examples, the maternal side is mostly dark gray as we share WILDINGER-FOURNELLE (our grandparents/Mom’s parents) with our mother.

Although many of the maternal matches on AncestryDNA have been identified, very few segments can be added to the map as chromosome information is not available on Ancestry. Those seen are from FTDNA, MyHeritage, or GEDmatch.

This post is about my paternal matches and therefore only the top bar of each chromosome is of interest.

Comparing sibling DNA

The color groups on AncestryDNA as well as those in the family tree are used to map the DNA segments. For the example, below, the green, pink, and yellow groups have only two shades. I’ve kept these groups simple to show that siblings don’t share all of the same DNA. They share about 50% of the same DNA. Less color makes it easier to see the four groups of the grandparents.

My paternal grandfather’s paternal ancestry, the blue groups, include purple for first cousins who share all four color groups and red to highlight our DEMPSEY brick wall. A darker blue is used for second cousins and lighter blues for more distant cousins.

The maps show all segment matches that have been assigned a most recent common ancestor (MRCA).

Side by side comparison of siblings’ DNA segment maps for all generations.

On chromosome 1, my DNA segments are from my father’s paternal side: PGF (blue and red) and PGM (green). My brother received mostly DNA from our father’s maternal side: MGF (pink) and MGM (green). On chromosomes 5, 10, 17, and 19 we share more DNA from the same groups. Still, there are gaps – chromosomes segments that have not been identified (light gray, see chromosomes 6, 7, and 9). These are segments that could lead to several of the brick walls in our tree including the ancestry of William A. W. DEMPSEY.

The segment map in GDAT can be filtered by generation making it easy to see where segments are coming from.

Generation 2 (1st cousins)

Cathy’s segment map for 2 generations.

Purple segments are 1st cousins who share our paternal grandparents, Fred Rothwell DEMPSEY and Myrtle Hazel ROOP – the generation 2 ancestors. These include 1st cousins once removed (1C1R), matches from the younger generation. Seven of the 24 grandchildren of Fred and Myrtle are represented in this map. More would be ideal but I am happy to work with what I have.

Generation 3 (2nd cousins)

Cathy’s segment map for 3 generations.

The dark blue and pink segments cover the purple segments as they represent one generation further back.

Dark blue segments are 2nd cousins who share William Henderson DEMPSEY and Laura Belle INGRAM. Matches have been found for six of their eight children who had descendants.

Pink segments are 2nd cousins who share Walter Farmer ROOP and Rebecca Jane CLONCH. Three of their six children have tested descendants.

Generation 4 (3rd cousins)

Cathy’s segment map for 4 generations.

Red, more easily distinguishable from the rest of the blue groups, is for 3rd cousins who share MRCA William A. W. DEMPSEY (parents unknown) and Sarah Ann WOOD.

Green segments are the 3rd cousins who share Irvin Lewis INGRAM and Mary M. DEMPSEY (no known relationship to William A. W. DEMPSEY).

Pink segments are the 3rd cousins who share Gordon Washington ROOP and Milla Susan PETERS.

Yellow segments are the 3rd cousin matches back to Alexander CLONCH and Tabitha Ann COOLEY.

Chromosome Analysis

Adding another generation to the map further breaks down the larger segments shared with 1st and 2nd cousins and adds identification to some blank segments.

In the example for the 4th generation, the middle section of chromosome 1 now shows red where previously no color was seen. These are 3rd cousins who share the DEMPSEY-WOOD ancestors. This red section is not visible in the map showing all generations (see the first segment map earlier in this post) as it is a segment shared with matches who have more distant ancestors in common – ancestors of Sarah Ann WOOD, the wife of William A. W. DEMPSEY.

On this breakdown of the segments on Chr. 1, the red segment identified as generation 4 is also shared by matches who have HONAKER-GOETZ of generation 7 as MRCA. I received this DNA from Frederick HONAKER, father of Rachel HONAKER who married Elijah WOOD. This segment cannot be used to find more distant ancestors of my brick wall William A. W. DEMPSEY as the DNA is from his wife Sarah Ann WOOD, daughter of Rachel and Elijah.

Focusing on my father’s paternal grandfather’s side using the blue groups

What have I been doing these past two-plus months? I’ve been populating my DNA database with matches, trees, and notes. I’ve been focusing on my father’s paternal grandfather’s side using the blue groups. More specifically, I’ve been concentrating on the matches that, I hope, will lead to the parents of my 2nd great-grandfather William A. W. DEMPSEY (1820-1867) of Rockbridge County, Virginia, and Fayette County, West Virginia (then part of old Virginia).

The amount of DNA we receive from a particular ancestor decreases with each generation. There is a chance that very little or no DNA was inherited from a specific ancestor. An ancestor did not pass on the same DNA to each of his children. Those children, with their different combinations of their parent’s DNA, passed on different combinations to each of their children. The more descendants tested, the more DNA can be matched to the ancestor.

I need more RED! I need 3rd cousins who descend from William A. W. DEMPSEY to transfer their raw data from AncestryDNA to FTDNA, MyHeritage, or GEDmatch so that I can analyze the DNA using a chromosome browser.

By paying close attention to the MRCAs and the segments shared with cousins, I’ve been able to eliminate those who are related to me through Sarah Ann WOOD’s ancestors. Those are the lighter blue segments that overlap the red segments.

Sarah’s ancestors came from lines where many descendants have tested. The Wood, McGraw, Honaker, and Wiseman families were large and intermarried. All four lived in Monroe County, West Virginia (then still part of Virginia) at the time it was created from Greenbrier County in 1799.

While I have large clusters of matches for these four families, the mysterious clusters that are associated with William A. W. DEMPSEY are confusing. I hope that some of his descendants may share one or the other of the light gray segments (non-assigned DNA). This would help to identify the area that I need to research to open the door to this brick wall.

Light gray segments (non-assigned DNA)

  • The gaps on the chromosome map have plenty of matches but the common ancestors in my tree haven’t been identified.
  • Some of the matches have ancestors in common with each other but these aren’t names found in my tree.
  • Many matches have small or no trees to work with.
  • I need confirmed cousins on the segment to help figure out where the mystery ancestors may fit in my family tree.

I’ve identified 87 3rd cousin matches descended from William A. W. DEMPSEY through my great-grand aunts and great-grand uncles. Of these 87, only 17 have their tests on sites with a chromosome browser. Do any of the others share non-assigned DNA segments with my brother or me?

What further complicates my William A. W. DEMPSEY brick wall is the fact that his descendants have more than one connection to me due to marriages of grandchildren and great-grandchildren to spouses who descend from other common ancestors, i.e. Wood, McGraw, Honaker, Wiseman, Sims, Johnson, Kincaid, Ingram, and my other Dempsey line.

Why not try Y-DNA?

My connection to William A. W. DEMPSEY is through my father (Fred), his father (Fred), his father’s father (William H.), his father’s father’s father (William A.W.). This would make the males in our family good candidates for Y-DNA testing. I have a paternal uncle, three brothers, and nine male first cousins who are descendants of William A. W. DEMPSEY. My grandfather Fred Rothwell DEMPSEY had six brothers; his father William Henderson DEMPSEY had three brothers.

I don’t feel comfortable asking relatives to do DNA tests, either autosomal or Y-DNA. I don’t have the time or want to put the effort into a Y-DNA project. However, if a direct-male descendant of William A. W. DEMPSEY has done the Y-DNA test or is planning on taking it, I would be happy to work with them on the genealogy side. I have a feeling the Y-DNA surname is not going to be DEMPSEY. Maybe someone can prove me wrong!

Why I wrote this post

When I write my ancestors’ stories, weaving the facts into the story and checking off the sources used, I usually find unanswered questions. Writing actually helps me think through things. So this post was primarily for me, to see if I am on the right track with the system and procedure I use for analyzing the DNA. If I can explain it and it makes sense (to me), I hope it also makes sense to my readers.

I know this is beyond beginner DNA. This might give you an idea of how, maybe a bit further down the road, you can work with your results. You might also be more advanced and able to give me some feedback on how you would treat a similar brick wall. Comments are always appreciated.

Lastly, I’d like to thank the cousins who’ve given me guest access to their DNA. I hope this will help them see how very helpful their data has been to me.

© 2021, copyright Cathy Meder-Dempsey. All rights reserved.

Look Who’s Finally Taken the Autosomal DNA Test

It’s been several years since I wrote Look Who’s Using DNA for Genealogy Research. Thanks to my youngest brother I’ve been able to work with his results at AncestryDNA since the end of May 2016.

His test results have confirmed most of our known paternal lines back to the 4th and 5th great-grandparents. However, to date, I haven’t been able to open the door in My Most Frustrating Brick Wall, William A. W. DEMPSEY, our 2nd great-grandfather. My brother has matches with descendants of six of his seven children. We need their help to find the parents of William A. W. DEMPSEY.

I’ve Finally Had my DNA Tested

First of all, I want to thank my brother for sending me an AncestryDNA test.

I received it on August 21. I did the test, activated it, and sent it off the following day. I was a bit worried it had gotten lost until the notification arrived that the sample was received on September 10. Apparently, the time between mailing off and their acknowledging receipt can take up to five weeks. The sample was processed and DNA extracted on the 17th and analyzed on the 21st. The results were in the following day. This part took less time than I anticipated.

My DNA Results are Ready

I saw my results before being notified as I was doing my daily check of my brother’s most recent (above 20 cMs) matches. Often there are no new matches or only 4th cousins very close to the 20 cMs cutoff. This time he had a new match with 2,410 cMs across 68 segments!

I switched over to my profile to see if my match list was available. At the top of the list in the full sibling category was my brother. No surprise there. The matches that followed were the same two first cousins and dozen second cousin he also has as matches.

Setting Everything up for DNA Analysis

Since I’ve been working with my brother’s results for nearly three and a half years, I was ready to use all the tools necessary to gather and analyze my matches. The initial set up went as follows.

Jonathan Brecher’s Shared Clustering tool

First, I ran a complete download of the matches (6 cMs and greater) on AncestryDNA using Jonathan Brecher’s Shared Clustering tool. This can take up to several hours.

Gedmatch, FTDNA, and MyHeritage

While I was waiting for the Shared Clustering tool to gather the matches, I downloaded the raw DNA file from Ancestry for upload to Gedmatch, FTDNA, and MyHeritage. It would be a few days before these three sites processed the data and my profiles there would be ready to work with. As soon as the kit was tokenized on Gedmatch, I ran a one-to-one comparison to see which segments my brother and I share.

Colin Thomson’s Pedigree Thief

I used the Chrome extension Pedigree Thief to download all matches 20 cMs and greater (4th cousin or closer) on AncestryDNA. The more distant 5th to 8th cousins will be gathered later. The Pedigree Thief generates a CSV file that I can download and use with the next tool.

Becky Mason Walker’s Genome Mate Pro

I’d already started to set up my profile in Genome Mate Pro (GMP), an app to help manage the data collected from the different platforms for autosomal DNA research. My GEDCOM had been uploaded and linked to my profile and the next step was to add the Match Keys. This involved adding the key values associated with my profile in the files from the various sources (AncestryDNA, Gedmatch, FTDNA, and MyHeritage). The AncestryDNA and Gedmatch keys were immediately available while I had to wait for FTDNA and MyHeritage to process the uploads before I could enter the keys from these sites.

The CSV file generated by the Pedigree Thief on AncestryDNA after gathering the matches was imported into GMP. A second CSV file of the shared matches of matches (gathering these takes several hours) was also added to GMP.

When FTDNA was completed, I downloaded the CSV file of matches and imported it into GMP. After paying $19 to unlock the AncestryDNA upload to FTDNA was I able to download the chromosome data file and import it into GMP.

MyHeritage will send a CSV file for matches and another for chromosome data per email when requested. Both of these files were uploaded to GMP.

When the Gedmatch kit completed processing I was able to copy/paste the One-To-Many DNA Comparison Results into GMP (list of top 3,000 matches). One-to-one Autosomal Comparison for the highest matches was generated one by one and copy/pasted into GMP. The rest of the matches’ chromosome data will wait until I pay for Tier 1 membership.

I didn’t use the Tier 1 utilities for my brother’s test as all data was imported before the switch to Genesis and then back to the new Gedmatch version. As new matches have been few I was able to import them individually. Gathering the chromosome data using one-to-one autosomal comparison of my test against nearly 3,000 matches would be too time-consuming.

Genome Mate Pro is now set up with matches from four platforms. I will continue to update on a daily, weekly, or monthly basis depending on the site.

Back at AncestryDNA

Although AncestryDNA does not offer a chromosome browser, the new features they have implemented this year help sort through matches.

This is the system I’ve chosen for my profile. The maternal side of my tree is for the most part from Luxembourg. For my brother, I’ve found about 240 matches (of a total of 64,000) who are from the maternal side. The closest confirmed match is a 4C1R. Most of these maternal matches are descendants of Luxembourg emigrants in the USA. I’ve elected to use the star for sorting them on his profile as well as mine. All paternal matches will be grouped by colors.

I may be overthinking this but a similar system worked well for my brother’s matches. Ancestry does not offer enough groups (in my opinion) for this to work for everyone. Having mostly paternal matches allows me to disregard half of my tree. As can be seen in the pedigree chart the brick wall I mentioned earlier is in my father’s direct paternal line.

I’ve labeled a group for my 2nd great-grandparents as 4PGF Dempsey-Wood as they are four generations from me and on my paternal grandfather’s side. As I have only 7 sets of 3rd great-grandparents, I created groups for each of them indicating the generation, grandparent side, and number to keep them in order per the pedigree chart, i.e. 3 sets on the paternal grandfather and 4 sets on the paternal grandmother’s side. Then I created groups for 6 sets of 4th great-grandparents on my paternal grandfather’s side and 8 sets of 4th great-grandparents on my paternal grandmother’s side.

This left me with two free groups. One is a catch-all for matches that have not been figured out and is labeled !Needs to be worked out.

My goal is to have all 4th cousins or closer matches grouped so that when I view shared matches of a match I can more quickly evaluate where the connection may be. The groups beginning with 5 will become redundant and I can then use them for more distant generations.

Shared Clustering Report

The Shared Clustering tool gathered all matches 6 cMs or greater on AncestryDNA with at least three shared matches and generated a clustering report. I have a little over 56,000 matches on Ancestry. The Shared Clustering tool clustered 12,800 of these into 88 clusters.

As this download was done BEFORE I started to work with the matches the notes are blank, i.e. MRCA or other information is missing. Most of the clusters have known matches seen previously on my brother’s match list and his clusters. But there are several clusters of matches not seen on his test. This was my first sign of having inherited DNA from my father that my brother didn’t.

Each time the Shared Clustering tool is used to generate a cluster list the cluster numbers change. Therefore it’s important to keep notes on Ancestry which will help to determine the most distant common ancestor of a cluster.

One of my highest unknown matches is in Cluster 81 with 61 cMs across 2 segments. I’ve been working through all of the highest matches in this cluster adding their Ahnentafels to GMP with the help of the Pedigree Thief and color-coding them in the ![C81] temp 77 group – the last free group. When I figure out where in my tree this cluster is coming from I can change the color-code to the correct ancestral group and free up the group.

Time for a Call to Action

Now that I’ve set everything up, I can begin to work through my matches and find cousins who may help me open the doors in my brick walls. Are you seeing my name on your match list? I won’t be sending out messages for a while but will reply to any I receive!

© 2019, copyright Cathy Meder-Dempsey. All rights reserved.