Page 1 of 10 123 ... LastLast
Results 1 to 10 of 100

Thread: R-U106 Fireside Chat

  1. #1
    Registered Users
    Posts
    62
    Sex
    Location
    Amsterdam
    Y-DNA (P)
    R-Z18/Z378+
    mtDNA (M)
    H

    Netherlands

    R-U106 Fireside Chat

    This thread was created as a fork based on a slightly off-topic post in "Bell Beakers, Gimbutas and R1b"

    Quote Originally Posted by Wing Genealogist View Post
    Yes, Z18 & L48 should be considered separately. A third clade to keep in mind is Z156. According to the Age analysis of Dr. Iain McDonald, Z156 is older & larger than Z18. In addition the Royal House of Wettin is a Z156 clade.
    Z18 already IS considered separately. This was arranged some six years ago by a few seasoned people who had thoroughly considered the issues involved and decided, most progress would be made by starting a dedicated project, focussing on R-Z18 (the achievements of the project have proven them right over the years).

    The result was the R-Z18 project with its own web site and forum. Apart from extensively supporting Z18+ people and developing the R-Z18 Y-DNA tree, the other main activity of the project is the development of the tools that are used by the project. These tools (e.g. the R-Z18 Y-DNA SNP Database and the programs that generate our web site) make the R-Z18 project probably the technically most advanced Y-DNA haplogroup project (it's kind of hard to understand, nobody in ISOGG shows any interest in our Y-DNA Database ).

    The fact that the R-Z18 project is not uniformly recognised in the field have always prohibited us from making any of these tools available to others (one cannot reasonably expect to freely receive things (software and data) from an entity whose existence is not recognised ).
    Last edited by Peter M; 01-27-2017 at 10:17 AM.

  2. #2
    Registered Users
    Posts
    62
    Sex
    Location
    Amsterdam
    Y-DNA (P)
    R-Z18/Z378+
    mtDNA (M)
    H

    Netherlands
    Possibly, it's an idea to respond to the content of the original message as well, as a very significant improvement is possible here.

    Quote Originally Posted by Wing Genealogist View Post
    Yes, Z18 & L48 should be considered separately. A third clade to keep in mind is Z156.
    I personally am convinced it would be "much better" to apply the idea behind the R-P312 projects in R-U106 as well. This part would then consist of the following projects (four projects instead of the current four projects ):

    1. R-U06 (all U106* and all "smaller" SNPs under it; and concentrating on getting people to "test into" one of the other three);
    2. R-Z18 (as it is now);
    3. R-Z156 (including L1);
    4. R-Z301 (including L48 and L47).

    These projects could share things like a web site and possibly a common set of tools, but as such each project would focus on their own branch(es) of the tree, so that much more progress would be possible. The most relevant change would be for some people to move away from their concentration on "number of members" as the most important aspect of their project.

    I am aware, there are people who think (apparently they do ) that even the R-Z18 project should not exist and is ignored, because it's "part of" the R-U106 project. These people tend to forget, that if that rule is applied consistently, then the R-U106 project should not exist either, because it's part of the R1b project. If we apply this concept uniformly, then all projects should be stopped and only the haplogroup CT project will be continued (not considering Africans for the moment).

    Quote Originally Posted by Wing Genealogist View Post
    According to the Age analysis of Dr. Iain McDonald, Z156 is older & larger than Z18.
    At this moment I'm not sure which data and analysis tools the said good Dr has been using, I'm not aware of him getting ANY information from the R-Z18 project on these issues. It's absolutely clear he's had no access to software for geographic analysis of haplogroup or parts thereof. BTW, I don't see the geographic facilities on the FT-DNA web site as particularly useful.

    Z156 being larger than R-Z18 might well be true, in reality R-Z18 is only about 5% of R-U106, based on the information in the R-Z18 Profile Database (if one doesn't have reasonably reliable information, the group may look much bigger than it actually is).

    Quote Originally Posted by Wing Genealogist View Post
    In addition the Royal House of Wettin is a Z156 clade.
    Interesting.
    Last edited by Peter M; 01-27-2017 at 11:12 AM.

  3. The Following 3 Users Say Thank You to Peter M For This Useful Post:

     Clinton P (01-30-2017),  jdean (01-27-2017),  TigerMW (01-27-2017)

  4. #3
    Registered Users
    Posts
    791
    Sex
    Location
    Waterville, ME
    Ethnicity
    Great Migration Colonists
    Nationality
    American
    Y-DNA (P)
    R1b-U106 (S10415)
    mtDNA (M)
    J1c2g (FMS)
    Y-DNA (M)
    I1a-P109 (23andMe)
    mtDNA (P)
    T2b3 (23andMe)

    United States Gadsden England Scotland Ireland Wales
    There are benefits and drawbacks to dividing up the clades as well as combining the clades. We certainly need experts at the various clade levels but we also need to look at the Big Picture.

    Iain's work on the ages: http://www.jb.man.ac.uk/~mcdonald/genetics/tree.html shows what can be done by combining the clades. In fact, one of the major issues he is now trying to address is that much of the uncertainty around the earliest branches of the tree is due to the fact his work stops at U106. He needs to incorporate the results of the clades above U106 to tighten up his age estimates for U106 as well as its top branches (including Z18). He is actively working to bring in the P310 level and hopes to go back to the M269 level some day.

    Iain fully recognizes his information is going to be "wrong" at many levels, but his willingness to put it out to the public does advance our knowledge.

    In the end it is all about the research, and the folks ordering the various tests. I strongly believe the best thing FTDNA has done is to create the volunteer Project Administrators who have turned into the experts in the field. Many of them are light years ahead of the so-called professionals and their willingness to share their expertise with the customers (and the general public) is what has moved the field at the incredible pace it has achieved.

  5. The Following 10 Users Say Thank You to Wing Genealogist For This Useful Post:

     dp (01-27-2017),  gotten (01-27-2017),  Inigo Montoya (05-22-2017),  jdean (01-27-2017),  Kelso (01-28-2017),  lgmayka (01-27-2017),  Michał (01-27-2017),  Pribislav (01-31-2017),  TigerMW (01-27-2017),  wlk (02-19-2017)

  6. #4
    Registered Users
    Posts
    62
    Sex
    Location
    Amsterdam
    Y-DNA (P)
    R-Z18/Z378+
    mtDNA (M)
    H

    Netherlands
    Quote Originally Posted by Peter M View Post
    These projects could share things like a web site and possibly a common set of tools, but as such each project would focus on their own branch(es) of the tree, so that much more progress would be possible. The most relevant change would be for some people to move away from their concentration on "number of members" as the most important aspect of their project.
    I'm well aware, I'm using the expression "much more" here, which is normally routinely used by fans of hackers, who (the fans) in reality don't have the faintest idea what they're talking about, so that "much more" is always a pretty suspect claim (at least to me).

    In this context I mean that a group of people working together, e.g. on a Y-DNA tree, will be making much more progress if they concentrate on a smaller more coherent branch of the tree. An example of this are the sponsored R-Z18 Panel tests that have been organised by the project in the past. People who would normally not consider ordering such a test, were enabled to do so as a group at a significant discount. This resulted in over 100 tests ordered (in a project with about 600 members), and allowed the project to get the results available of a much larger group of people testing the new SNPs discovered (by the project: the ZPn SNPs) in the Big-Y results.

    Currently, we (most people in the field) are still working with the test results from people who are interested enough and have the funds available to order a modern (Full-Y) test. These are not always the most interesting samples. If the member base of a project is more coherent, it might well be possible to organise "focussed testing" of the most interesting profiles, the tests being paid by the project using the donations of the members. This necessitates a lot of organisational work (for all the said panel tests in R-Z18, hundreds of emails have been sent), but as a result "much more" progress will be made. This will only be doable in a smaller, more focussed, project. The ideal case being a project in which groups of people know each other.
    Last edited by Peter M; 01-27-2017 at 12:45 PM.

  7. The Following User Says Thank You to Peter M For This Useful Post:

     Michał (01-27-2017)

  8. #5
    Registered Users
    Posts
    62
    Sex
    Location
    Amsterdam
    Y-DNA (P)
    R-Z18/Z378+
    mtDNA (M)
    H

    Netherlands
    Quote Originally Posted by Wing Genealogist View Post
    There are benefits and drawbacks to dividing up the clades as well as combining the clades. We certainly need experts at the various clade levels but we also need to look at the Big Picture.
    Personally, I think projects should be focussed on branches that share a clear common theme and not so much on their actual size or the ambition of their administrators. This is especially important if we see a project as supporting their members (advising on tests and explaining test results) as this only works if there's something of a binding factor. Given the amount of time it takes to handle all requests in R-Z18 (about 600 members), then a project ofthousands of members seems a wee bit impractical.

    Quote Originally Posted by Wing Genealogist View Post
    Iain's work on the ages: http://www.jb.man.ac.uk/~mcdonald/genetics/tree.html shows what can be done by combining the clades. In fact, one of the major issues he is now trying to address is that much of the uncertainty around the earliest branches of the tree is due to the fact his work stops at U106. He needs to incorporate the results of the clades above U106 to tighten up his age estimates for U106 as well as its top branches (including Z18). He is actively working to bring in the P310 level and hopes to go back to the M269 level some day.
    Of course, there's nothing wrong with (project-) independent researchers who do their own investigation in this case on the age of SNPs of a number of branches, but it could be on any subject the researcher deems relevant. There is no necessary implication to the scope of projects in this, he just creates his own blog somewhere. Whether one sees the good Dr. as an independent researcher is a matter of opinion, I guess (I'm aware, I'm not responding to his approach, BTW).

    Quote Originally Posted by Wing Genealogist View Post
    Iain fully recognizes his information is going to be "wrong" at many levels, but his willingness to put it out to the public does advance our knowledge.
    The only problem I see, is that lots of people see his results as facts and start to act based on them, instead of as an opinion of a single person who works individually (at the very least, he hasn't ever consulted the R-Z18 project) and is a fan of age calculation. Especially in the case of newcomers in the field, this is an important issue.

    Quote Originally Posted by Wing Genealogist View Post
    In the end it is all about the research, and the folks ordering the various tests. I strongly believe the best thing FTDNA has done is to create the volunteer Project Administrators who have turned into the experts in the field. Many of them are light years ahead of the so-called professionals and their willingness to share their expertise with the customers (and the general public) is what has moved the field at the incredible pace it has achieved.
    I agree, although it would be nice if the projects decided to co-operate more, as I think the field currently is not getting any stronger.

    To summarise, I think there are two issues here: first are the projects (a) that support their members (advising on tests and helping explaining results, in effect a knowledgeable help desk as an alternative/enhancement to FT-DNA's) and (b) evaluate all results available and use them to develop a Y-DNA tree for their branch, and there's individual and independent researchers who refrain from advising members of any project, except of course in the area of their research (e.g. SNP ages in this case). These researchers should make very clear that what they're saying is primarily a result of their research and as such a personal opinion.
    Last edited by Peter M; 01-27-2017 at 02:42 PM.

  9. The Following User Says Thank You to Peter M For This Useful Post:

     jdean (01-27-2017)

  10. #6
    Registered Users
    Posts
    897
    Sex
    Location
    California
    Y-DNA (P)
    R1b-Z2103>Y14416

    Quote Originally Posted by Wing Genealogist View Post
    There are benefits and drawbacks to dividing up the clades as well as combining the clades. We certainly need experts at the various clade levels but we also need to look at the Big Picture.

    Iain's work on the ages: http://www.jb.man.ac.uk/~mcdonald/genetics/tree.html shows what can be done by combining the clades. In fact, one of the major issues he is now trying to address is that much of the uncertainty around the earliest branches of the tree is due to the fact his work stops at U106. He needs to incorporate the results of the clades above U106 to tighten up his age estimates for U106 as well as its top branches (including Z18). He is actively working to bring in the P310 level and hopes to go back to the M269 level some day.

    Iain fully recognizes his information is going to be "wrong" at many levels, but his willingness to put it out to the public does advance our knowledge.

    In the end it is all about the research, and the folks ordering the various tests. I strongly believe the best thing FTDNA has done is to create the volunteer Project Administrators who have turned into the experts in the field. Many of them are light years ahead of the so-called professionals and their willingness to share their expertise with the customers (and the general public) is what has moved the field at the incredible pace it has achieved.
    Nobody has made enquiries to the administrators of the R1b Basal Subclades project where the P310+, M269- folks reside. Frankly, we wouldn't let them go without a consensus of the members. Most of our members are committed to going with our project trees and YFull.
    Last edited by Joe B; 01-27-2017 at 04:32 PM.
    YFull R1b-M269>L23>Z2103>Z2106>Z2108>Y14512>Y20971>Y22199, ISOGG R1b1a1a2a2c1b Y14416, FTDNA R-M64

  11. The Following 2 Users Say Thank You to Joe B For This Useful Post:

     jdean (01-27-2017),  lgmayka (01-28-2017)

  12. #7
    Registered Users
    Posts
    791
    Sex
    Location
    Waterville, ME
    Ethnicity
    Great Migration Colonists
    Nationality
    American
    Y-DNA (P)
    R1b-U106 (S10415)
    mtDNA (M)
    J1c2g (FMS)
    Y-DNA (M)
    I1a-P109 (23andMe)
    mtDNA (P)
    T2b3 (23andMe)

    United States Gadsden England Scotland Ireland Wales
    Quote Originally Posted by Joe B View Post
    Nobody has made enquiries to the administrators of the R1b Basal Subclades project where the P310+, M269- folks reside. Frankly, we wouldn't let them go without a consensus of the members. Most of our members are committed to going with our project trees and YFull.
    Iain is currently working on the various subclades of P312 and hopes to examine the M269+, P312- U106- clades at some later date. His work does not call for anyone to be moved from any projects. I cannot speak for him in regards to who he is contacting, but I do know he is working with Alex Williamson and would imagine he is working with the various project administrators as well.

    His work is similar to the work YFull is doing on the age analysis, but he does add in some additional steps to further refine the estimated ages.

  13. The Following 3 Users Say Thank You to Wing Genealogist For This Useful Post:

     gotten (01-27-2017),  lgmayka (01-28-2017),  Michał (01-27-2017)

  14. #8
    Registered Users
    Posts
    704
    Sex
    Location
    Texas
    Ethnicity
    English, Irish, German
    Nationality
    US
    Y-DNA (P)
    L21>L226>FGC5639

    England Germany Netherlands France Ireland Switzerland
    Quote Originally Posted by Peter M View Post
    The other main activity of the project is the development of the tools that are used by the project. These tools (e.g. the R-Z18 Y-DNA SNP Database and the programs that generate our web site) make the R-Z18 project probably the technically most advanced Y-DNA haplogroup project (it's kind of hard to understand, nobody in ISOGG shows any interest in our Y-DNA Database ).
    Could you summarize the tools and their functions ? I know about the TMRCA charting tool which is really nice. I am more interested in the tree building aspect than the age estimates. Also, I saw several posts where initial work was being done on creating a 400 - 500 YSTR database based on NGS tests. I am very interested in that.

    I am assume that you also have some form of database for U106 research as well. There are a lot of those databases out there - but they are mostly available in spreadsheet form or web sites which are less useful. Making a robust database available directly to the public (vs. personal research) may be an issue with FTDNA. The FTDNA terms and conditions could legally shut down such a database tool. There is a lot of useful YSEQ data that is tracked very manually due to the ID translation issue. The Big Tree is a database of sorts, but it is really an output summary vs. a true database. Glad to see that U106 assisted Alex with expanding the scope to include U106. I do have tools to collect FTDNA data which includes all of U106 (it covers over 4,000 projects) which is in a MySQL database.

    We also have the issue of competing haplotrees. It used to be ISOGG leading the way (I rarely even go to their haplotree these days). When Big Tree hit the scene that was the best source for P312 YSNP haplotrees and he is now collecting YSTR data as well (but not from NGS tests). YFULL has pretty good coverage for certain haplogroups but really lacks P312 data. Shame on Full Genomes for not having any usable database. FTDNA has not hardly even shown up on the real map for a viable NGS database that includes private YSNPs. Even their YSNP reports are bloated with worthless NATGEO data but miss 80 % of the relevant data from their Big Y tests. Here is a spreadsheet version of YSTR reports (last full pull was July, 2016 and 20 % was pulled in November, 2016 - caution this is a 29 MB file):


    http://www.rcasey.net/DNA/Temp/HG_R_...20161117A.xlsx

    Lately, I have become very interested in automating (coding) a charting tool. Now that R-L226 has 500 67 marker tests with around 20 % coverage with YSNPs and 42 branches, I am able to chart over 75 % of L226 with fairly high reliability:

    http://www.rcasey.net/DNA/R_L226/Hap.../L226_Home.pdf
    Last edited by RobertCasey; 01-27-2017 at 06:45 PM.

  15. The Following User Says Thank You to RobertCasey For This Useful Post:

     MitchellSince1893 (01-27-2017)

  16. #9
    Registered Users
    Posts
    62
    Sex
    Location
    Amsterdam
    Y-DNA (P)
    R-Z18/Z378+
    mtDNA (M)
    H

    Netherlands
    You've covered a lot of ground in a few lines and to reasonably respond to all issues touched will be hard, as I've worked on the vast majority of these issues for years are there's an awful lot to say about them. I will give things a fair try.

    Quote Originally Posted by RobertCasey View Post
    Could you summarize the tools and their functions ? I know about the TMRCA charting tool which is really nice. I am more interested in the tree building aspect than the age estimates. Also, I saw several posts where initial work was being done on creating a 400 - 500 YSTR database based on NGS tests. I am very interested in that.
    I have the best part of a hard drive filled with software and data. I could write a book on this software. Some of it was written in the past and not used recently (e.g. a program that generated Network Diagrams in full colour (to indicate different sub clades)). Some are minor utilities that prepare input data for other programs. But essentialy it is a set of tools that automate all steps in a project's workflow, as I tend to see it.

    I'm not concerned with age estimates, for the simple reason that I have strong doubts about them. A few years ago, I wrote a program that automated the Generation5 spreadsheet by Ken Nordvedt (it's still used in the STR Results display on the R-Z18 web site). Unfortunately, it is not 100% reliable (e.g. it doesn't handle nulls correctly); I have to look at it again, but that's not high on my priority list. So I share your liking for the tree building process.

    I'm not too enthousiastic about the idea of working with 400-500 STRs, I don't think this will be worthwhile, although it might be an idea to have a look at the results to date, trying to find a few extra STRs that might be useful and to include those in databases. A have a currently hibernating program that can be used to automatically call STRs (and SNPs) from NGS results in a .bam file (called gyca: Genealogical Y-Chromosome Analyser).

    Quote Originally Posted by RobertCasey View Post
    I am assume that you also have some form of database for U106 research as well. There are a lot of those databases out there - but they are mostly available in spreadsheet form or web sites which are less useful. Making a robust database available directly to the public (vs. personal research) may be an issue with FTDNA. The FTDNA terms and conditions could legally shut down such a database tool. There is a lot of useful YSEQ data that is tracked very manually due to the ID translation issue. The Big Tree is a database of sorts, but it is really an output summary vs. a true database. Glad to see that U106 assisted Alex with expanding the scope to include U106. I do have tools to collect FTDNA data which includes all of U106 (it covers over 4,000 projects) which is in a MySQL database.
    That's a save assumption, but currently it is only used for R-Z18 research (no other groups have been interested; I have used things occasionally to help people sort out issues). The word "database" is used for a lot of different things. Personally, I tend not to see a spreadsheet as a database, for the simple reason that it is very difficult or next to impossible to actually DO something with the data that was not designed in, in the structure of the sheet.

    Making a data set (neutral term) available via the internet might annoy FT-DNA depending on the organisation behind it; it's not very likely FT-DNA will take legal steps to fight a party that represents the vast majority of their customer base. A useful database would add value to the things FT-DNA is already displaying, otherwise it would be useless and this added value is a good way of countering any copyright claims.

    Quote Originally Posted by RobertCasey View Post
    We also have the issue of competing haplotrees. It used to be ISOGG leading the way (I rarely even go to their haplotree these days). When Big Tree hit the scene that was the best source for P312 YSNP haplotrees and he is now collecting YSTR data as well (but not from NGS tests). YFULL has pretty good coverage for certain haplogroups but really lacks P312 data. Shame on Full Genomes for not having any usable database. FTDNA has not hardly even shown up on the real map for a viable NGS database that includes private YSNPs. Even their YSNP reports are bloated with worthless NATGEO data but miss 80 % of the relevant data from their Big Y tests. Here is a spreadsheet version of YSTR reports (last full pull was July, 2016 and 20 % was pulled in November, 2016 - caution this is a 29 MB file):
    Competing trees? These trees all share roughly the same issues: (a) they all assume there's a single party (person or very small team) who is responsible for defining a "covering" Y-Tree (a tree including all branches); (b) these trees are all defined offline by said party; and (c) are all using ancient and static technology for display of the tree.

    as to (a): I don't believe a single entity will be able to follow all new developments in all branches of the Y-Tree in order to be able to define a complete Y-Tree. I am a strong believer in focussed groups that each concentrate on a single area of the tree and follow all test results from all labs.
    as to (b): requires much more explanation, so I will leave this for the moment.
    as to (c): none of these trees use state of the art graphics (as that would complicate things dramatically) and e.g. allow the viewer to drill down into the underlying SNP database and/or other data sets.

    Apart from these shared issues, there are things that are particular to a single tree names, e.g. the ISOGG tree is based on manual labour using only an html editor, which are outdated tools (few people code web pages by hand nowadays, most use CMSs). ISOGG insist on not involving knowledgeable people in their tree maintenance decision-making and the results are revealing. FT-DNA are selling tests to individuals and do remarkably little, and/or with comparatively low quality, for groups of customers. FGC cannot be expected to offer analysis support as they are selling FULL-Y tests at the lowest prices possible to knowledgeable people. Y-Full AFAIA are not known for their co-operation with other parties and that's a real requirement if you want to build a tree and do not use all test results available.

    Personally, I think a somewhat new and revolutionary approach is needed.

    Quote Originally Posted by RobertCasey View Post
    Lately, I have become very interested in automating (coding) a charting tool. Now that R-L226 has 500 67 marker tests with around 20 % coverage with YSNPs and 42 branches, I am able to chart over 75 % of L226 with fairly high reliability:

    http://www.rcasey.net/DNA/R_L226/Hap.../L226_Home.pdf
    This looks nice in comparison to most of the other tree displaying pages I've seen around. For static display of a tree as part of a document, this will work fine, but for presentation online on the web, there are lots of things, I would expect to be available, that will be very hard, or impossible, to add using a document-oriented thing like .pdf. But it is a very nice example of how a Y-Tree could look like.

  17. #10
    Registered Users
    Posts
    4,397
    Sex
    Location
    Tierra de Ayllon
    Y-DNA (P)
    U152>L2>Z49>Z142>
    mtDNA (M)
    H1bt
    Y-DNA (M)
    I2a2a1b2a1b1>Y4925
    mtDNA (P)
    H37

    England Scotland Wales Germany Ireland Sweden Finns
    You two should start a dna database business together.
    Last edited by MitchellSince1893; 01-27-2017 at 10:27 PM.
    Y DNA line continued: Z142>Z12222>FGC12378>FGC12401>FGC12384
    35% English, 26% Scot/Ulster Scot, 14% Welsh, 14% German, 5% Ireland, 3% Nordic, 2% French/Dutch, 1% India
    Hidden Content

Page 1 of 10 123 ... LastLast

Similar Threads

  1. U106* (U106+ Z381- Z18-) & L199.1+ & L217.1+
    By TigerMW in forum R1b-U106
    Replies: 63
    Last Post: 08-26-2020, 07:50 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •