About us
Goal
The objective of this project is to match up individuals or families who share a common male ancestor with the Roberts surname and variant spellings including variant Robert, Robarts, Robartes, Robart, Robarts, Ropartz, Robberts, Ropert, and Ruppert.
Who should join
The two types of people who should join are those with DNA of males with a Roberts surname, or those who think their surname should genetically be Roberts (e.g. an adopted male, or a male whose father’s surname is Roberts, even if his name is not). Women with a connection to the Roberts name or who manage a Kit for a Roberts male are welcome to join.
Summary of the project
There are four main pages which contain the findings of the project:
The objective of this project is to match up individuals or families who share a common male ancestor with the Roberts surname and variant spellings including variant Robert, Robarts, Robartes, Robart, Robarts, Ropartz, Robberts, Ropert, and Ruppert.
Who should join
The two types of people who should join are those with DNA of males with a Roberts surname, or those who think their surname should genetically be Roberts (e.g. an adopted male, or a male whose father’s surname is Roberts, even if his name is not). Women with a connection to the Roberts name or who manage a Kit for a Roberts male are welcome to join.
Summary of the project
There are four main pages which contain the findings of the project:
- The Y-DNA Results Overview chart, introduced in late 2024, is arranged into sub-groups according to likely genetic relationships
- The Project Activity Feed page where results may be discussed and members can post questions and comments
- A web-site which contains a discussion of the results and records traditional family history connections between members
- A repository of family trees organised by kit-number
A quick guide to interpretation of the results
The best way to view the results is through the Y-DNA Results Overview chart. A few weeks after your DNA results are added to the project you should check this page, and search for your kit number (using your browser search key). Note that only 200 names are displayed so you should check page 2 (etc.), or increase the number of names on each page. You should have been moved into a group.
At present there are four types of groups.
1. Family matched.
These are groupings that may have an established paper trail linking their families based on traditional genealogical studies. We encourage members to test Big Y-700 to discover their SNP branching for their family lines.
4. Unmatched non Roberts observers.
The best way to view the results is through the Y-DNA Results Overview chart. A few weeks after your DNA results are added to the project you should check this page, and search for your kit number (using your browser search key). Note that only 200 names are displayed so you should check page 2 (etc.), or increase the number of names on each page. You should have been moved into a group.
At present there are four types of groups.
1. Family matched.
These are groupings that may have an established paper trail linking their families based on traditional genealogical studies. We encourage members to test Big Y-700 to discover their SNP branching for their family lines.
2. Probably matched.
These are groupings which have a 50% chance of being related in the past 10 generations (250 years) based on simple rule (a 60/67, 33/37, 24/25 or a 12/12 match). In many cases there will be an extremely close match, in other cases no link will have been established. The group administrators do not necessarily have any more information than is provided on the Y-DNA Results chart. These may be men that form a cluster or clan based on their Big Y results (they share a common Y-DNA SNP branch node represent sub-branches of that SNP branch node).
3. Unmatched Roberts. These are groupings which have a 50% chance of being related in the past 10 generations (250 years) based on simple rule (a 60/67, 33/37, 24/25 or a 12/12 match). In many cases there will be an extremely close match, in other cases no link will have been established. The group administrators do not necessarily have any more information than is provided on the Y-DNA Results chart. These may be men that form a cluster or clan based on their Big Y results (they share a common Y-DNA SNP branch node represent sub-branches of that SNP branch node).
4. Unmatched non Roberts observers.
You don’t have any “probable” relatives in the surname project. It is possible for the administrators to miss a connection among members which may exist, so if you see your Y-DNA group, and think it is “closely linked” to another sequence or set on the Roberts page, then email robertsdna@hotmail.com, or the other administrators.
Defining “closely linked” is important: Below I describe a rationale for using a match based on 33/37, 60/67, 100/111 as a definition of “closely linked”. Other definitions are possible such as their SNP branch from a common Y-DNA ancestor. Note that you may match someone with the Roberts name who is not in the project (note that administrators can’t see these matches easily), you could invite them to join the Roberts project to form a new group.
5. Ungrouped
The administrators have not allocated you to a group yet.
Submitting family trees or interpretations
The Y-DNA grouped results page is the main way of establishing or studying the relationships between families. A site was set up in November 2011 to provide a repository of family trees associated with the ancestors of individuals in the project, as well as a location for the findings of the project (what the gene sequences tell us about family interrelationships). Submissions of family trees in text or html formats, or interpretations/paper connections can be made to roberts.dna.project@gmail.com.
Project History and Discussion
Pre 2011 The project was initiated by Peter Roberts, and his work collecting and matching a number of genealogies is available on the webpage at: http://home.comcast.net/~libpjr1/robertsdna.htm . Peter’s page contains an alternate version of the Y-DNA results page located on this site with an extensive range of additional family tree information and discussion about the relationships between different family groups.
Nov 2011 As the number of members has grown it has been hard to maintain the project at a very detailed level for all participants. At present I have suggested a method of grouping participants (see below) and presenting information on the colorized Y-DNA page located on the FTDNA site. I have implemented the method beside Peter’s groupings as they are quite similar. (Please note: These methods were written over ten years ago (eg in 2011) and since then we have Y-DNA sequencing that discovers the exact Y-DNA SNP branching for various families. FT DNA are now also extracting Y-DNA SNPs from Family Finder raw data. Y-DNA results from Family Finder may give us an idea of some of the Y-DNA branching but members with only Family Finder should add a Y-DNA test for matching and better placement on the charts. We now encourage members to upgrade to Big Y-700 to discover their Y-DNA branching.)
A second stage of the project will be to provide a web location which is both for storing family tree information (for those who wish to provide it) and discussion about the relationships among genetic groupings. This would be similar to information in the form provided on Peter’s page, but probably at a more basic (and hopefully easier to maintain) level.
Methodology of Grouping 2011 - Tony Roberts
Group administrators can choose how they group the results of a surname project. I describe two approaches, both of which are used on the Roberts page.
1. Proven or extremely close matching
People are linked into groups according to a proven family tree connection, or very tight genetic matching (e.g. 35-37, 63-67, 105-111 or Big Y SNP results). This is clean and allows tight well defined family trees/histories to be associated with each set.
2. Probable genetic grouping
In this project I have been grouping people based on the likelihood of a probable relationship in the last 10 generations. For example a 50% chance of being related in the past 10 generations (250 years) (*). Roughly this means a 60/67, 33/37 match, 24/25 match or a 12/12 match. Note that some of the marker mutate faster than others so it is hard to make strict rules about grouping (7 mismatches in high mutation rate markers could signify a closer relationship than 3 mismatches in very slowly mutating markers). These mutation rate differences are said to be incorporated in TiP reports which administrators can run between members of their family projects. These reports provide the definitive "probability" of a genetic link N generations ago. Some paternal lines may also exhibit a recLOH event that throws out the matching. It affects the multi-allele markers.
I favour this grouping philosophy because genetic genealogy seems to me to be about going beyond the paper trail and making connections via DNA. Even if the purported connections are not able to be proved, it is interesting to know of, and perhaps contact, other people who share ancestors in some time over the past 100-500 years or so. This may help you determine that your ancestors lived in Wales 400 years ago (for example), even though your paper trails stops in London, Kentucky or South Africa. Also early naming patterns in different regions of the United Kingdom and Europe may bring up matches with different surnames now (for example fixed surnames started in Wales much later than in parts of England).
FTDNA Y-DNA colourized charts are especially useful for examining clusters (families), a number of rows with very little or no colour variation show strong family ties. If we separate out these group by strict proven relationships we lose the ability to see how close a related (but unproven) cluster is. Thus there is some advantage to not breaking off families into strict subgroups.
There are some simple naming tools that group administrators can use to arrange things on the chart to better see connections as well. For example, if we had two R1b1a2 groups (the first being Group 2) that might have a bridge between them, we could change the name of the second group to R1b1a2 Group 2a. This would appear right below Group 2. The group names have to be precisely written to ensure that the alphabetical order gives the correct order. (ie. Group 2a should appear just below Group 2).
A disadvantage of approach 2 is that is harder to write family based genetic histories if distinct families are not broken off into well defined groups. I propose to deal with this by collecting family information on a rootsweb site thus leaving the FT-DNA Y-DNA Results chart to tell the DNA story, not the paper based story.
(*) My generation spans, year ranges and probabilities may be updated over time. They are based on the discussion which can be found on the FTDNA frequently asked questions (FAQ) pages.
Reference: Understanding Results: Y-DNA Short Tandem Repeat (STR); Questions. 23, 24, 25 and 26 on this FAQ page.
Defining “closely linked” is important: Below I describe a rationale for using a match based on 33/37, 60/67, 100/111 as a definition of “closely linked”. Other definitions are possible such as their SNP branch from a common Y-DNA ancestor. Note that you may match someone with the Roberts name who is not in the project (note that administrators can’t see these matches easily), you could invite them to join the Roberts project to form a new group.
5. Ungrouped
The administrators have not allocated you to a group yet.
Submitting family trees or interpretations
The Y-DNA grouped results page is the main way of establishing or studying the relationships between families. A site was set up in November 2011 to provide a repository of family trees associated with the ancestors of individuals in the project, as well as a location for the findings of the project (what the gene sequences tell us about family interrelationships). Submissions of family trees in text or html formats, or interpretations/paper connections can be made to roberts.dna.project@gmail.com.
Project History and Discussion
Pre 2011 The project was initiated by Peter Roberts, and his work collecting and matching a number of genealogies is available on the webpage at: http://home.comcast.net/~libpjr1/robertsdna.htm . Peter’s page contains an alternate version of the Y-DNA results page located on this site with an extensive range of additional family tree information and discussion about the relationships between different family groups.
Nov 2011 As the number of members has grown it has been hard to maintain the project at a very detailed level for all participants. At present I have suggested a method of grouping participants (see below) and presenting information on the colorized Y-DNA page located on the FTDNA site. I have implemented the method beside Peter’s groupings as they are quite similar. (Please note: These methods were written over ten years ago (eg in 2011) and since then we have Y-DNA sequencing that discovers the exact Y-DNA SNP branching for various families. FT DNA are now also extracting Y-DNA SNPs from Family Finder raw data. Y-DNA results from Family Finder may give us an idea of some of the Y-DNA branching but members with only Family Finder should add a Y-DNA test for matching and better placement on the charts. We now encourage members to upgrade to Big Y-700 to discover their Y-DNA branching.)
A second stage of the project will be to provide a web location which is both for storing family tree information (for those who wish to provide it) and discussion about the relationships among genetic groupings. This would be similar to information in the form provided on Peter’s page, but probably at a more basic (and hopefully easier to maintain) level.
Methodology of Grouping 2011 - Tony Roberts
Group administrators can choose how they group the results of a surname project. I describe two approaches, both of which are used on the Roberts page.
1. Proven or extremely close matching
People are linked into groups according to a proven family tree connection, or very tight genetic matching (e.g. 35-37, 63-67, 105-111 or Big Y SNP results). This is clean and allows tight well defined family trees/histories to be associated with each set.
2. Probable genetic grouping
In this project I have been grouping people based on the likelihood of a probable relationship in the last 10 generations. For example a 50% chance of being related in the past 10 generations (250 years) (*). Roughly this means a 60/67, 33/37 match, 24/25 match or a 12/12 match. Note that some of the marker mutate faster than others so it is hard to make strict rules about grouping (7 mismatches in high mutation rate markers could signify a closer relationship than 3 mismatches in very slowly mutating markers). These mutation rate differences are said to be incorporated in TiP reports which administrators can run between members of their family projects. These reports provide the definitive "probability" of a genetic link N generations ago. Some paternal lines may also exhibit a recLOH event that throws out the matching. It affects the multi-allele markers.
I favour this grouping philosophy because genetic genealogy seems to me to be about going beyond the paper trail and making connections via DNA. Even if the purported connections are not able to be proved, it is interesting to know of, and perhaps contact, other people who share ancestors in some time over the past 100-500 years or so. This may help you determine that your ancestors lived in Wales 400 years ago (for example), even though your paper trails stops in London, Kentucky or South Africa. Also early naming patterns in different regions of the United Kingdom and Europe may bring up matches with different surnames now (for example fixed surnames started in Wales much later than in parts of England).
FTDNA Y-DNA colourized charts are especially useful for examining clusters (families), a number of rows with very little or no colour variation show strong family ties. If we separate out these group by strict proven relationships we lose the ability to see how close a related (but unproven) cluster is. Thus there is some advantage to not breaking off families into strict subgroups.
There are some simple naming tools that group administrators can use to arrange things on the chart to better see connections as well. For example, if we had two R1b1a2 groups (the first being Group 2) that might have a bridge between them, we could change the name of the second group to R1b1a2 Group 2a. This would appear right below Group 2. The group names have to be precisely written to ensure that the alphabetical order gives the correct order. (ie. Group 2a should appear just below Group 2).
A disadvantage of approach 2 is that is harder to write family based genetic histories if distinct families are not broken off into well defined groups. I propose to deal with this by collecting family information on a rootsweb site thus leaving the FT-DNA Y-DNA Results chart to tell the DNA story, not the paper based story.
(*) My generation spans, year ranges and probabilities may be updated over time. They are based on the discussion which can be found on the FTDNA frequently asked questions (FAQ) pages.
Reference: Understanding Results: Y-DNA Short Tandem Repeat (STR); Questions. 23, 24, 25 and 26 on this FAQ page.