Announcement

Collapse
No announcement yet.

Data Mining and Screen Scraping – Right or Wrong?

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

  • #16
    Originally posted by jbarry6899 View Post
    We have some project members who have BigY results and YFull analysis. Does FGC add anything to those findings? I'm willing to make the investment if it seems worthwhile.

    Jim
    I don't know the answer to that question. I don't recall a case where someone has had BAM analysis from both YFull and FGC.

    The only difference I could think of off the top of my head is that one of the two companies may have a larger database to make comparisons for matching shared novel variants. I don't know if there's that much of a difference in the size of their databases and, if so, which one is the larger.

    If you ask FGC to comment, perhaps they can give you good reasons to have them re-analyze a BAM file already analyzed by YFull.

    Comment


    • #17
      Is FGC developing a tree with estimated dates for the various subclades? If so, how do they compare with yfull?

      Timothy Peterman

      Comment


      • #18
        Originally posted by T E Peterman View Post
        Is FGC developing a tree with estimated dates for the various subclades? If so, how do they compare with yfull?

        Timothy Peterman
        Not that I'm aware of. You can go to their website and send them an e-mail about that.

        Comment


        • #19
          Originally posted by MMaddi View Post
          I don't know the answer to that question. I don't recall a case where someone has had BAM analysis from both YFull and FGC.


          If you ask FGC to comment, perhaps they can give you good reasons to have them re-analyze a BAM file already analyzed by YFull.
          I'll do that. Thanks!

          Comment


          • #20
            For many reasons the Z58 + Z63 project group encourage members to use YFull for the analysis of the BAM file including so results can be seen and included in the project spreadsheets and YTree.

            Comment


            • #21
              I get the impression that yfull is building a tree based on the comparison of BAM files. I think that they are trying to be as unbiased as possible, so they don't even report the existence of a subclade, unless a BAM file suggests that it exists.

              Timothy Peterman

              Comment


              • #22
                Originally posted by T E Peterman View Post
                I get the impression that yfull is building a tree based on the comparison of BAM files. I think that they are trying to be as unbiased as possible, so they don't even report the existence of a subclade, unless a BAM file suggests that it exists.
                If the subclade is well-documented elsewhere (e.g., ISOGG), I think that YFull may respond to an explicit request for tree placement--especially if at least one BAM file has the SNP. But clearly, their primary focus is on SNPs common to two or more BAM files.

                Comment


                • #23
                  Information provided by isogg on yfull.com

                  http://www.isogg.org/wiki/YFull

                  Comment


                  • #24
                    I uploaded a BAM file yesterday. It is #3,836. They are beginning to get enough data to offer good results, but I think they still have a ways to go to get it properly refined.

                    Timothy Peterman

                    Comment


                    • #25
                      At least there are 4 people in Haplogroup O who get BAM analysis from both YFull and FGC.

                      Comment


                      • #26
                        Originally posted by Parameswara View Post
                        At least there are 4 people in Haplogroup O who get BAM analysis from both YFull and FGC.
                        Do you know of any feedback from any of them as to what additional or differing information they received when comparing the two companies' analysis?

                        Comment


                        • #27
                          YFull shares on Internet at https://docs.google.com/spreadsheets...OQvODrhSDuXATo, a list of SNPs they discovered/named.

                          Is there anywhere a list that has positions at which FGC discovered/named SNPs ? SNP names would be nice, but are not required .

                          I am familiar with http://ybrowse.org/. So I can enter in there, let's say 22,367,457 or FGC35522, however multiple searches are time consuming.

                          Thank you - W. (Mr.)

                          P.S.
                          To me that is one of visible differences between those two companies.

                          Comment


                          • #28
                            Originally posted by dna View Post
                            YFull shares on Internet at https://docs.google.com/spreadsheets...OQvODrhSDuXATo, a list of SNPs they discovered/named.

                            Is there anywhere a list that has positions at which FGC discovered/named SNPs ? SNP names would be nice, but are not required .

                            I am familiar with http://ybrowse.org/. So I can enter in there, let's say 22,367,457 or FGC35522, however multiple searches are time consuming.

                            Thank you - W. (Mr.)

                            P.S.
                            To me that is one of visible differences between those two companies.
                            As I posted earlier in this thread, FTDNA offers a few hundred FGC SNPs on their Advanced Orders menu. Go to the menu, type in "FGC" as the search term and you'll see these SNPs. Since the FGC SNP numbers they assigned my singletons go up to FGC13492, I think it's safe to say that they've named well over 10,000 SNPs, although many of them at this point are singletons. I think it's also safe to say that if FTDNA is offering an FGC SNP for testing, it's not a singleton, but has been established as a clade-defining SNP.
                            Last edited by MMaddi; 14 July 2015, 02:37 PM.

                            Comment


                            • #29
                              Originally posted by MMaddi View Post
                              As I posted earlier in this thread, FTDNA offers a few hundred FGC SNPs on their Advanced Orders menu. If you go to the menu, type in "FGC" as the search term and you'll see these SNPs. Since the FGC SNP numbers they assigned my singletons go up to FGC13492, I think it's safe to say that they've named well over 10,000 SNPs, although many of them at this point are singletons.
                              http://ybrowse.org/ has entries from FGC1 to at least FGC35522, however, I am looking for an easier look up of multiple positions.

                              Mr. W.

                              Comment


                              • #30
                                Originally posted by MMaddi View Post
                                Do you know of any feedback from any of them as to what additional or differing information they received when comparing the two companies' analysis?
                                Those 4 people (including me) sent their BAM files to both company last year. So, I am sure there are a lot more improvement since then.

                                From the last year report, YFull is very good at Y-STR. FGC use Lobstr to read the STR, which is not very reliable. YFull use their own software to read STR, and I think the result is very good.

                                About mtDNA, FGC use Yoruba sample as reference, while YFull use rCRS and RSRS. They both create FASTA file for mtDNA.

                                We can read YFull report at their website and download it, but we only can download FGC report. So YFull report is user friendly. FGC report InDel. I think the FGC report is very good but complicated.

                                FGC named every SNP they found in Y-Chromosome, although FGC's new SNP standard is more strict and Private SNP generally less than YFull give (I am talking about interpretation from Big Y BAM Files, not from FGC NGS Test), but FGC give name to all SNP they found. YFull always update their report.

                                I was very pleased with their analysis. Last year I paid YFull $49 and FGC $25. I think if someone can afford to order Big Y, then why not just pay more $100 for the interpretation from FGC and YFull ?

                                Comment

                                Working...
                                X