Announcement

Collapse
No announcement yet.

Understanding Raw Data

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #46
    Note that David Reynolds has this to say about some of the Geno 2.0 results he's investigated:
    "Geno 2.0 results for L371, L705, L706, L720, and Z246 are anomalous and are being investigated. Note that the same alleles are also reported in U106+ and Hg Q results. Most likely erroneous! They are blacklisted and not displayed at this time."

    This is from his WTY/Geno 2.0 results table, available at:
    http://daver.info/WTY/R-L21.pdf

    So don't worry, you're not a total mutant...

    Comment


    • #47
      Originally posted by ScooterCat View Post
      Note that David Reynolds has this to say about some of the Geno 2.0 results he's investigated:
      "Geno 2.0 results for L371, L705, L706, L720, and Z246 are anomalous and are being investigated. Note that the same alleles are also reported in U106+ and Hg Q results. Most likely erroneous! They are blacklisted and not displayed at this time."

      This is from his WTY/Geno 2.0 results table, available at:
      http://daver.info/WTY/R-L21.pdf

      So don't worry, you're not a total mutant...
      Thanks ScooterCat! (Although I had been hoping that I might be making an appearance in the next X-men sequel).

      Comment


      • #48
        Originally posted by Bartot View Post
        no, I am saying that if you transfer from NG to FTDNA, then Ftdna should advise someone where a SNP is a negative because it was tested under NG.

        example, below my positive L446 is L447, if I did the NG test and it was transfered to Ftdna, then I would highly presume that L447 was tested. Since in 100% of all positive L446 tested are found negative L447 ( so far ) , then If I inquire ( or my project manager) to Ftdna to run a L447 test, Ftdna should advise ( with integrity ) it was already done and the result was ...whatever...without a fee.
        If you are interested in your result for single yDNA SNPs in Geno 2.0, there are two efficient ways to determine a negative result. Assuming that the SNP is on the Geno chip and that you've uploaded your Geno 2.0 results to FTDNA and you're in a project where the project SNP page is on, check your results on that page. If you don't see a result for that SNP, then you can assume that you're negative or perhaps it was a "no call." If you want to be sure, look in the raw data that you have from Geno 2.0 and see what the result is. Then you can cross-check that with ISOGG's ySNP index - at http://www.isogg.org/tree/ISOGG_YDNA_SNP_Index.html - to see which is the derived base and if you have it. If you got the result for the SNP you're interested in, then there would be no chance of ordering a SNP from FTDNA when you already have the result from Geno 2.0.

        Originally posted by Bartot View Post
        My understanding is that NG results are presented in rs numbers and not all SNPs have rs numbers ...at this time.
        No, that's incorrect. I have a spreadsheet of all 12,000+ yDNA SNPs on the Geno 2.0 chip and everyone one of them has an actual SNP name, with the usual L, DF, Z, etc. prefixes, plus the new CTS, F and PF prefix SNPs. If you'd like this spreadsheet, send me a private message with your e-mail address and I'll send it to you.

        Originally posted by Bartot View Post
        As per comment below made by experienced scholars of genetics
        Of the SNPs in the Geno 2.0 test, there are approximately four hundred SNPs in the y-chromosome that are derived (rather than ancestral) that have so far been seen in at least some haplogroup I1 people. Most of those SNPs will probably be phyloequivalent to the SNPs given in the I1 SNP tree pathway shown above. But there are still a few interesting Geno 2.0 SNP alleles that may be important to haplogroup I1 people. They are:

        F1583+, found in an P109+ person.
        F2642+, found in an Z140+ person.
        F3312+, found in an M253+, but Z58-, Z63-, Z131- person.
        CTS7362+, CTS9352+, CTS9477+, found in a single L573+ person.
        YSC0000261+, found in three separate Z140+ people.
        CTS8647+, found in the same three people as YSC0000261+, and also in another Z140+ person, a L573+ person, and a Z60+ person.
        It seems that there are genetic genealogists who are knowledgeable about I1 and the relevant SNPs on Geno 2.0. Anyone who is interested in these SNPs should obviously be in contact with them. They name 6 SNPs that are interesting for I1 people. Do you think that, in order to be told negative results for these 6 SNPs by FTDNA, thousands of negative SNP results, most totally irrelevant to I1 people, should be reported when you upload Geno 2.0 results to FTDNA? That's totally inefficient. Just look up the results for those 6 SNPs in your raw data.

        Originally posted by Bartot View Post
        It would seem , by your comments and the comments made by NG, that NG are far more interested in the ( negative , as you put it ) ancestral , so advising the tested of the negative would be paramount as that is the whole purpose of what NG 2.0 was set up to do. It was not a test to find family members from the past.
        I don't see why you think that anything I've posted or has been written by NG indicates that they're far more interested in the ancestral. When trying to place SNPs on a tree, both ancestral and derived are important to know. The derived results are given on FTDNA project SNP pages. You can assume that any SNP results not listed are ancestral. If you want to check specific SNPs, look in your raw data.

        Originally posted by Bartot View Post
        to me, NG is basically a test of one's auDNA signals as being more important and less so to yDNA and mtDNA
        For most of the general public who aren't involved in genetic genealogy already, you're right. The most attractive part of the test is the admixture analysis obtained from the autosomal DNA.

        However, for the genetic genealogy community, that's not the case. There are several free admixture tools available to analyze 23andMe and Family Finder results. The mtDNA results from Geno 2.0 are somewhere between HVR1/HVR2 and full sequence results, probably closer to full sequence. If someone wants comprehensive, they're better off getting the full sequence from FTDNA. The aspect of Geno 2.0 that really stands out is the 12,000+ yDNA SNPs and the potential to find new subclades and make major changes to the tree. In my view, the yDNA testing makes the $199 Geno 2.0 price worth it, all by itself. If Geno 2.0 didn't have all these yDNA SNPs, I wouldn't even consider spending $199 for Geno 2.0.

        Comment


        • #49
          Originally posted by MMaddi View Post
          If you are interested in your result for single yDNA SNPs in Geno 2.0, there are two efficient ways to determine a negative result. Assuming that the SNP is on the Geno chip and that you've uploaded your Geno 2.0 results to FTDNA and you're in a project where the project SNP page is on, check your results on that page. If you don't see a result for that SNP, then you can assume that you're negative or perhaps it was a "no call." If you want to be sure, look in the raw data that you have from Geno 2.0 and see what the result is. Then you can cross-check that with ISOGG's ySNP index - at http://www.isogg.org/tree/ISOGG_YDNA_SNP_Index.html - to see which is the derived base and if you have it. If you got the result for the SNP you're interested in, then there would be no chance of ordering a SNP from FTDNA when you already have the result from Geno 2.0.



          No, that's incorrect. I have a spreadsheet of all 12,000+ yDNA SNPs on the Geno 2.0 chip and everyone one of them has an actual SNP name, with the usual L, DF, Z, etc. prefixes, plus the new CTS, F and PF prefix SNPs. If you'd like this spreadsheet, send me a private message with your e-mail address and I'll send it to you.



          It seems that there are genetic genealogists who are knowledgeable about I1 and the relevant SNPs on Geno 2.0. Anyone who is interested in these SNPs should obviously be in contact with them. They name 6 SNPs that are interesting for I1 people. Do you think that, in order to be told negative results for these 6 SNPs by FTDNA, thousands of negative SNP results, most totally irrelevant to I1 people, should be reported when you upload Geno 2.0 results to FTDNA? That's totally inefficient. Just look up the results for those 6 SNPs in your raw data.



          I don't see why you think that anything I've posted or has been written by NG indicates that they're far more interested in the ancestral. When trying to place SNPs on a tree, both ancestral and derived are important to know. The derived results are given on FTDNA project SNP pages. You can assume that any SNP results not listed are ancestral. If you want to check specific SNPs, look in your raw data.



          For most of the general public who aren't involved in genetic genealogy already, you're right. The most attractive part of the test is the admixture analysis obtained from the autosomal DNA.

          However, for the genetic genealogy community, that's not the case. There are several free admixture tools available to analyze 23andMe and Family Finder results. The mtDNA results from Geno 2.0 are somewhere between HVR1/HVR2 and full sequence results, probably closer to full sequence. If someone wants comprehensive, they're better off getting the full sequence from FTDNA. The aspect of Geno 2.0 that really stands out is the 12,000+ yDNA SNPs and the potential to find new subclades and make major changes to the tree. In my view, the yDNA testing makes the $199 Geno 2.0 price worth it, all by itself. If Geno 2.0 didn't have all these yDNA SNPs, I wouldn't even consider spending $199 for Geno 2.0.
          There seems to be some confusion here.......basically, we know that NG tests for Ydna is up to Nov.2011 and the mtDna is up to April 2012 ( v14 ).

          So, if you test with NG 2.0 , ALL SNPs etc both Positive and Negative should be made available to the individual. Its only the SNP outside of these dates the someone needs to pay FTDNA to test to resolve for a fee.

          BTW, it also stupid for FTDNA to show me on my site a SNP test I could do when I already am shown positive for a different branch of that haplogroup
          tested - P327 T1b1a < wrong branch , I am proved to be negative
          P328 T1a2a < wrong branch , why ask me to test?
          P322 T1a2a < wrong branch , why ask me to test?

          Also, thank you for the link

          And, L447 as I said cannot go with L446 ( you cannot have both ) according to your link.
          Bartot
          FTDNA Customer
          Last edited by Bartot; 17 December 2012, 12:27 AM.

          Comment


          • #50
            Well the Download Data tab finally showed-up but when I click to download a box comes-up saying 'File Conversion' - it gives me 'Windows Default', 'MS-DOS', and 'Other.' None work it just comes out jibberish.

            Comment


            • #51
              Originally posted by JTR View Post
              Well the Download Data tab finally showed-up but when I click to download a box comes-up saying 'File Conversion' - it gives me 'Windows Default', 'MS-DOS', and 'Other.' None work it just comes out jibberish.
              Alright I figured it out - thanks! It was a .cvs.gz compressed so I had to un do the gz and then open it in Excel.

              One thing is that the file only gives you the base pair not whether you are + or -. What am I to do with this?

              Where do you find info on what genotype is + or -?

              Also, if Excel 2007 good enough? Or do I need to update my Excel program?
              JTR
              Registered User
              Last edited by JTR; 18 December 2012, 03:39 PM.

              Comment


              • #52
                Originally posted by JTR View Post
                Alright I figured it out - thanks! It was a .cvs.gz compressed so I had to un do the gz and then open it in Excel.

                One thing is that the file only gives you the base pair not whether you are + or -. What am I to do with this?

                Where do you find info on what genotype is + or -?
                For SNPs already available for testing by FTDNA (P312, U106, DF23 etc.), use the SNP index published by ISOGG - http://www.isogg.org/tree/ISOGG_YDNA_SNP_Index.html The last column gives the ancestral base, followed by the derived base.

                For new SNPs that are only on the Geno 2.0 chip, there is no guide to ancestral/derived values yet. That will take some work, comparing results from people in various haplogroups/subclades. Ultimately, the complete answers won't come until Spencer Wells publishes a new tree, based on Geno 2.0 results.

                Comment


                • #53
                  Originally posted by MMaddi View Post
                  For new SNPs that are only on the Geno 2.0 chip, there is no guide to ancestral/derived values yet. That will take some work, comparing results from people in various haplogroups/subclades. Ultimately, the complete answers won't come until Spencer Wells publishes a new tree, based on Geno 2.0 results.
                  Actually, the CTS and PF SNPs have been published to Ymap:

                  http://ymap.ftdna.com

                  F SNPs haven't been published yet.

                  The position on the Y tree hasn't been published yet, but the ancestral and derived alleles are there.

                  For those who haven't used Ymap before:

                  - Click "enter now" or the Y chromosome image on the homepage
                  - In the Landmark or Region box, delete the default text and then enter a SNP name, such as L21.
                  - Click Search
                  - After the page refreshes, scroll down to the large Details grid in the center of the page, and you'll see your SNP name on one of the lines.
                  - Click your SNP name, and you'll then get a details page for the SNP
                  - Look for the Mutation, allele_anc and allele_der entries.

                  Also, Here's a shortcut. Right-click and copy this direct link to L21's SNP details page:

                  http://ymap.ftdna.com/cgi-bin/gb2/gb...=ChrY;name=L21

                  Paste into your web browser's URL bar, change the L21 to any other SNP name and press Enter. That should bring you directly to that SNP's details page.

                  And finally, there's also a text file that you can view in your browser or download that contains all the SNP data from Ymap -- just use the standard Find function (usually Ctrl-F on Windows, perhaps Command-F on Macs?) to search the file:

                  http://ymap.ftdna.com/GRC/refSeq/snps.gff

                  Elise
                  efgen
                  Webinar Coordinator
                  Last edited by efgen; 18 December 2012, 06:57 PM.

                  Comment


                  • #54
                    Thanks! That is good for the Y but what about the MT, X, and Aut.?

                    Question regarding the NG data: Everyone seems to have every SNP, for example Z1 is listed with my base pair T T. Now I thought that if I was negative for this SNP that that meant I would not have a base pair to report - guess that was wrong - becuase these were AIMs - that is this marker (Z1) was only to be found in certain populations. But positive and negative, I guess mean something else?

                    Without FTDNA selecting my Y positives when I downloaded the data to them I would be in a hell of mess to go through evey one of these SNPs and figure out what ones I had as positve or negative. Or am I to just take my general Hg and check all SNPs downstream? I guess derived means positive and ancestral means negative.

                    What am I missing?
                    JTR
                    Registered User
                    Last edited by JTR; 18 December 2012, 08:34 PM.

                    Comment


                    • #55
                      Well after using the above chromosome browser I found some errors with the FTDNA transfer results. Since FTDNA is listing only postive results (I assume that means derived alleles) then there should not be any negative ones with the Geno transfer. But after looking at over 40 SNPs I found that 15 of them were actually negative and not positive as FTDNA lists.

                      Also on the Geno Data I am not sure what some of the symbols mean. For instance:

                      L566 is C>T but my alleles read T C and L781 is G>A and mine is a A G.

                      Also L366 reads 'ins' for the ancestral and 'del' for the derived.

                      And then L594 has T>C but my allele is A.

                      Comment


                      • #56
                        Originally posted by JTR View Post
                        ... Also on the Geno Data I am not sure what some of the symbols mean. For instance:

                        L566 is C>T but my alleles read T C and L781 is G>A and mine is a A G.

                        Also L366 reads 'ins' for the ancestral and 'del' for the derived.

                        And then L594 has T>C but my allele is A.
                        The split calls (ie A G) is really a positive result. For technical reasons some locations (SNPs) achieve only very faint results, which the probes are not designed to handle. NatGeo (with help from FTDNA) "gerry-rigged" the system to signal this double result (meaning the individual is positive/derived for this SNP).

                        Some SNPs are actually InDels (Insertions/Deletions). It is harder to design probes for the chips to accurately trace these, but some have been developed for the GenoChip.

                        Comment


                        • #57
                          Originally posted by Wing_Genealogis View Post
                          The split calls (ie A G) is really a positive result. For technical reasons some locations (SNPs) achieve only very faint results, which the probes are not designed to handle. NatGeo (with help from FTDNA) "gerry-rigged" the system to signal this double result (meaning the individual is positive/derived for this SNP).

                          Some SNPs are actually InDels (Insertions/Deletions). It is harder to design probes for the chips to accurately trace these, but some have been developed for the GenoChip.
                          Sounds good - so then if there is and 'I' or a 'D' for my result that would mean insertion and deletion?

                          As far as FTDNA I don't see how I can trust their transfer results when they marked so many of my listed markers wrong? How do I know if the ones listed are the exact number - I would have to check every marker from the Geno data unless there is some program that I can use that will tell me which ones are ancestral or derived. Is there such a program?

                          Comment


                          • #58
                            Originally posted by JTR View Post
                            Sounds good - so then if there is and 'I' or a 'D' for my result that would mean insertion and deletion?

                            ...
                            I cannot answer the second part of your query, but as to the first part: Yes, I[nsertion] or D[eletion].

                            For the SNPs I have observed (below R1b-U106) the mutation was a deletion, so the ancestral state was an "I" (which technically really isn't an insertion, but it would be way too wordy to put in a full explanation). It is likely there are SNPs on the GenoChip where their was an insertion and the ancestral state would be "D".

                            Comment


                            • #59
                              Originally posted by Wing_Genealogis View Post
                              I cannot answer the second part of your query, but as to the first part: Yes, I[nsertion] or D[eletion].

                              For the SNPs I have observed (below R1b-U106) the mutation was a deletion, so the ancestral state was an "I" (which technically really isn't an insertion, but it would be way too wordy to put in a full explanation). It is likely there are SNPs on the GenoChip where their was an insertion and the ancestral state would be "D".
                              Ok! I am still confused about one of my splits (since their positive) being 'approx. R-M417' seeing that I am I-Z138. I guess the same mutation is possible under different Hg's?

                              And I am still not sure what is up with L594 T>C yet I am A - I guess that's a mistake?

                              Thanks for the answers!

                              Comment


                              • #60
                                Originally posted by JTR View Post
                                Ok! I am still confused about one of my splits (since their positive) being 'approx. R-M417' seeing that I am I-Z138. I guess the same mutation is possible under different Hg's?
                                Yes, mutations at the same location on the y have been known to occur in different haplogroups. They're called parallel mutations. They may be private mutations, that have occurred recently within one paternal line, or they may define a new subclade of the haplogroup where the new mutation is found.

                                ISOGG's ySNP index has examples of these parallel mutations - http://www.isogg.org/tree/ISOGG_YDNA_SNP_Index.html - which are given the same name as the original SNP with ".x" added. One example is L159.1, defining I2a1a, and L159.2, defining R1b1a2a1a1b3a5a.

                                Comment

                                Working...
                                X