Announcement

Collapse
No announcement yet.

New Testing Service

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

  • #16
    Mention of Suracell was tongue-in-cheek not swab-in-cheek.

    Comment


    • #17
      I think it is FABULOUS that they can DO soooo many SNP's at once!!! deCODEme clearly has the advantage on numbers - they do a million - whereas 23andMe only does 580K.

      What does it mean or what can you do with the data?

      Well, deCODEme offers comparison to 300K other persons from around our little planet, although the majority will be Icelanders. And 23andMe offers comparisons to 1K persons, probably mostly investment bankers (SMIRK). Knome does the WHOLE ENCHILADA, but WE can't afford it and the utility is mostly bragging rights (PUKE).

      If you use 23andMe you lose the right to your SPIT. If you use deCODEme your buccal swab is destroyed.

      23andMe has Jonathan Pritchard on their Advisors and a lot of statistical heavyweights on staff. SO, if you want to buy a PROMISE of a great analytical tool to make sense of the data you would chose 23andMe.

      I wouldn't DO any of them. Obviously doing a lot of SNP's is CHEAP and the real GOLD is the analytical tool - and no one is being FORWARD about the analytical options.

      Of course if you are DYING to know whether your Type 2 Diabetes is the results of BAD GENES or your uncontrolled ADDICTION to Milk Chocolate you MUST test. (PUKING and SMIRKING at the same time).

      SAVE YOUR MONEY.

      Comment


      • #18
        Knome cost

        I found a blog entry that gave the cost of whole genome sequencing at Knome as $349,000.

        http://www.portfolio.com/news-market...nal-Gene-Tests

        Comment


        • #19
          I've asked 23andme specifically for the locations of the "more than 2000" sites they test of mtdna; haven't heard back yet (in my case, I want to cross check it against mitomap.org list of all locations reported associated with disease. Though i guess we can forget any reporting of heteroplasmy).

          I also haven't heard back yet from decodeme. I asked them if theyll be making the raw data available.

          One nice thing about 23andme is that all 1/2 million data points will be available- they have that nice table organized by chromosome. So far it's not clear that decodeme will be doing that. I'd rather have half a million raw data points than 1 million where only the results of their proprietary software can be viewed.

          It looks like from someone's post earlier on this thread they say 23andme owns your dna but decode me destroys it? I haven't seen that and will go and look-privacy differences might change what I think.

          Comment


          • #20
            Originally posted by penguin
            ...
            It looks like from someone's post earlier on this thread they say 23andme owns your dna but decode me destroys it? I haven't seen that and will go and look-privacy differences might change what I think.
            Lots of reading to be done!!! 23andMe claims ownership of your saliva sample while you retain all rights to genetic infomation contained in said sample. Odd.

            Comment


            • #21
              deCODEme and 23andMe

              You can download your genotypes from deCODEme for all 1,000,000+ SNPs as one dataset, in an ASCII table with rs (Reference SNP) ID numbers.

              I have also written to 23andMe without any response as yet. I think they went public before they were quite ready (the day after deCODEme), so they may not have sufficient support staff on hand for the initial flurry of queries. It's not at all clear to me that you can download your results as a dataset, or even whether they will publish the rs numbers for their hand-selected set of SNPs on their custom chip. We'll just have to wait and see how they respond.

              Right now the only Y-SNPs we know about would be the ones in Illumina's off-the-shelf 550K chip, and there are only 10. However, they list Peter Underhill (the Stanford geneticist who discovered many SNPs) as an "external editor" and their graphic of the Y chromosome coverage shows a lot of red, so I suspect they have a substantial number. The Human1m set used by deCODEme has 1418 Y-SNPs (plus 876 Copy Number Variation markers). Many of those will undoubtedly turn out to be redundant, though.

              I don't think either chip will replace the more complete data you can get with mtDNA or Y tests from the genealogical companies. The autosomal data is more intriguing.

              Comment


              • #22
                Originally posted by Ann Turner
                You can download your genotypes from deCODEme for all 1,000,000+ SNPs as one dataset, in an ASCII table with rs (Reference SNP) ID numbers.

                I have also written to 23andMe without any response as yet.
                Well, my message is out of date now. I had sent a query to 23andMe on the 19th and received a response this morning, so perhaps they are catching up with the initial flurry of interest. I asked if a customer could download the raw data (with rs numbers) and they answered that they do not currently provide it.

                Comment


                • #23
                  Hmmm. So, contra their proclamation that all persons are 'entitled' (by their Creator?) to actual, actionable FACTS, 23andMe is NOT going to PROVIDE actual, actionable facts. That reads like a question but it is not.

                  23andMe evidently aims to the vendor to those who wish to be medically frightened within the contextual embrace of an all-knowing and (potentially) all-caring consumer services company.

                  There are a lot of things I just don't get. And now I have another: id est, why a start-up genetic testing company would not address the concerns of the largest segment of the retail genetics testing community, genetic genealogists. Why?

                  Could it be that the genetic genealogy community is tapped-out? (That is a question). Or might it be that the genetic genealogy community does not provide a profit margin or market segment large enough to interest Silicone Valley? (No need to answer).

                  So, genetic genealogists, what has your genetic genealogy company done for you lately?

                  Or to put it in a way that all can feel good about ... you have paid for and built the database and THEY HAVE COME. What's next?

                  Comment


                  • #24
                    On 23andme's website it says:

                    "With raw access to all of your genetic data, you can home in on any gene or other feature you want. Or ..."

                    besides the fact that I assume they meant "hone" not "home", I inferred from this that customers would have access to the raw data. so does this mean it can be accessed but not dowloaded as a single file? is it downloadable maybe in seperate files, one for each chromosome and x, y and mito? or are they trying to split hairs and argue that "raw access to genetic data" is not the same as "acccess to raw data"? I think I might get a guarantee about raw data in writing before I go with any company-the advertising mnight not be inline with what one actually gets.

                    Comment


                    • #25
                      We'll need to have some first-hand reports from customers to see how things actually pan out, but my interpretation of the website is that you can view the raw data for any particular item you look at (see the Gene Journal PDFs, for example), but apparently you can't download all 550K SNPs as a dataset.

                      Maybe if 23andMe feels they need to offer that feature in order to be competitive with deCODEme, they'll change their website. But not that many people are going to be interested in raw data that looks something like this (made up example):

                      rs23157 CC
                      rs3982218 CT
                      rs75698002 GA

                      However, I'm interested, as I'm musing about the genealogical implications of studying autosomal "haplotype blocks", groups of SNPs located closely together on a chromosome so that they travel together as a unit for some number of generations. That may be too futuristic right now, though -- there are enormous computational problems in extracting haplotypes from genotypes and accounting for the recombination rate. In the above example, you could have inherited either allele from either parent.

                      Comment


                      • #26
                        Originally posted by Ann Turner
                        ...
                        I don't think either chip will replace the more complete data you can get with mtDNA or Y tests from the genealogical companies. The autosomal data is more intriguing.
                        The graphic depicting their mitochondrial coverage has gaps within the contol region that is depicted as solid on the depiction of 'the competitor' coverage. Does this mean that 23andMe control region mitochondrial results cannot be used on Mitosearch? And is there any justification for another, for their, non-conforming control region standard?

                        Comment


                        • #27
                          Duh ... the answer is in the question.

                          Obviously the goal of these genome-wide companies is to rapidly build large SNP databases for research purposes by offering broad scans at, comparatively, low rates.

                          The are a lot more research dollars in disease and pharmacology than in genealogy and a larger retail market share among medically-concerned consumers than hobby genealogists. The medically-concerned are a better target market than genealogists because being concerned about disease they are more likely to participate in a research study regarding their condition of concern.

                          As we all know, there are a lot of genealogical tests done under the 'do not share' option, a lot of results that do not get posted to the public db's, and that we are a group of fairly narrow interests; our family trees. So the current analytical offerings relevant to genetic genealogists are just a bone.

                          Inevitably, the companies will get into some variety of admixture-mapping in support of their research focus but that analysis may not immediately become a retail service. I think we can already see how that will grow, through the investigation of genetic constellations around particular medical conditions. It will quite some time before the database is sufficiently comprehensive to to support the 'definitive' admixture assessment.

                          Comment


                          • #28
                            replies from the 2 companies

                            I heard back from both companies:

                            "Hello,

                            Thank you for your interest in 23andMe's services. We do not share the complete
                            list of the SNPs that are included on our custom array.

                            Regards,
                            23andMe Team"

                            I had asked 23andme specifically about which mitochondrial locations were included - looks like they either don't want to give even that out, even though they showed it in a figure, and/or they didn't really read my messge very carefully. And here's an excerpt from decode me. Decode me sent a long message of several paragraphs in length.

                            "You will have access to the actual variant information on every single
                            one of the 1 million SNPs looked at. You will, however, not have access
                            to the temporary data files and calculation instruments used for
                            generation of our calculated results and images.

                            Your deCODEme results will not reside in a single static report, but in
                            an ever evolving information-rich and secure web account. The account
                            will include your actual genotypes for the one million SNPs analyzed. You..."

                            So my info concurs with the earlier posts of others on this list - it's the opposite of what I first thought based on websites and the new york times article about the availability of raw data.

                            Incidentally, I agree that looking at blocks of SNPs is all the rage now. This idea of linkage disequilibrium specifically for nearby SNPs is being pursued now. means among other things that we don't have to look at every single letter of the genome to have a very good idea of what we've inherited.

                            I want to be able to log onto accounts at both websites as a mock customer before knowing which, if any, company I want to pursue at this point. Since I can't do that, I'm inclined to wait until the summer perhaps when the dust settles.

                            Anyone taking the plunge now? any more info on either company?

                            Comment


                            • #29
                              Originally posted by penguin
                              ...
                              Incidentally, I agree that looking at blocks of SNPs is all the rage now. This idea of linkage disequilibrium specifically for nearby SNPs is being pursued now. means among other things that we don't have to look at every single letter of the genome to have a very good idea of what we've inherited.
                              ...
                              Ancestry studies based on linkage disequilibrium are not inconsistent with medically-focused studies of particular genes, as sequences flanking genes are also in LD - one might think of those flanking sequences as a frame that assures the preservation of the gene itself. However, those framing sequences are also involved, 'chemo-mechanically' in the gene expression. So, one could suppose, knowing the conditions at flanking sequences could allow an inference as to the character of the gene and thereby disclose medically-significant information.

                              Comment


                              • #30
                                Originally posted by penguin
                                ...
                                Anyone taking the plunge now?...
                                Not me. But deCODEme leads the pack in number of SNP's and access. And leaves open the possiblility for a third party ancestry analysis based on the data.

                                Comment

                                Working...
                                X