PDA

View Full Version : Information from bloggers VS. Information Published in Scientific Journals



motzart
22-11-14, 19:59
I notice that a lot of you on here seem to think that information on ancient genomes from various bloggers using GEDMatch and other open source genetic analysis tools trumps whatever information was actually published in the papers from the researchers who performed the initial analysis. The open source tools are not vetted in the same way that academic research is, nor is it of the same quality. GEDMatch is unreliable from a serious scientific perspective as data in GEDMatch results can be easily faked and manipulated, when I upload my own 23andme results to GEDMatch I could easily go and edit the text file to give myself a bunch of unique SNPs, or copy other openly available data from an ancient genome to make myself appear a 50% match to whatever ancient genome I wanted. The same can be said for all of the "homebrew" admixture calculators floating around.

When there is a discrepancy between results obtained from open source or hobbyist dna tools and official research it is evidence that the open source tools have flaws, not vice versa.

Greying Wanderer
22-11-14, 21:51
Although that is likely true I think there is another aspect as well. The academic papers have to double check everything before publication whereas amateurs can take the current state of data from one set of papers and make guesses for fun which may or may not be correct and need to be confirmed by later academic papers anyhow.

Angela
23-11-14, 16:36
I agree. People have no idea how easy it is to fake this stuff. Relying on the statistical analysis of a blogger with no academic credentials to lose is very perilous. As a general matter, as you say, it's best to stick to the results in the papers. At least you know that there's no out and out fraud being perpetrated. They can't, because they have to post their methodology, and so they'd get caught. That's why I ignore anything where a detailed methodology isn't provided, or at least I treat it with great caution.

For anyone who would blithely accept statistical data from some blogger where the exact methodology and reference populations aren't provided, I can offer you the Brooklyn Bridge at a really good price. :) Or, as was once famously said, "There's a ****** born every minute, and two to fleece him.":grin:

Even so called "official" 23andme results can be faked. For one thing there's no way of authenticating that a certain result does stem from "X" place. You have to take the person's word. If you share with enough people on there you have results in your file from all over the world and can label them in any way you choose. (For example, I share with "colonial" Americans, Iberians, French people, Italians from all over, Greeks, Russians, people from the Levant, etc. )

Plus, I personally saw results for certain people which totally changed from one day to the next. They had one profile in my list, where I could see the name, and they posted other profiles on a discussion thread. Obviously, they took a screenshot of the profile of someone with whom they shared and posted it as their own. When I pointed out in general terms that I had seen that, I suddenly got dropped by a few people. http://cdn.eupedia.com/forum/images/smilies/main/laughing.gif

That's why I stopped sharing with people on 23andme years ago. The number of crazies who are interested in this topic is truly staggering. (That's also the reason why I almost never PM people. You never know what kind of loon you might have on the other end. I've been burned a couple of times...so, I'm much more cautious now.)

ED. It's amazing how perfectly innocent words, quotes from another time, what have you, get censored by internet gods with a very dirty mind! :grin: Last time it happened to me because I was talking about the fichi d'India, the *****ly pear. (I repeat it because I want to see if it censors me again. I find it amusing. :smile:)

Aberdeen
23-11-14, 18:03
..............

For anyone who would blithely accept statistical data from some blogger where the exact methodology and reference populations aren't provided, I can offer you the Brooklyn Bridge at a really good price. :) Or, as was once famously said, "There's a ****** born every minute, and two to fleece him.":grin:

...................


ED. It's amazing how perfectly innocent words, quotes from another time, what have you, get censored by internet gods with a very dirty mind! :grin: Last time it happened to me because I was talking about the fichi d'India, the *****ly pear. (I repeat it because I want to see if it censors me again. I find it amusing. :smile:)

That's funny - they did censor "thorny pear" again. At least they're consistent. And I guess if you want to quote P.T. Barnum, you'll have to say something like "there's a lollipop born every second", which doesn't have quite the same ring as the original.

Edit: What annoys me more than censorship is incorrect autocorrect, but I think that's coming from my computer, even though I turned it off. It changed "they're" to "their" - good thing I noticed.

Angela
24-11-14, 16:01
That's funny - they did censor "thorny pear" again. At least they're consistent. And I guess if you want to quote P.T. Barnum, you'll have to say something like "there's a lollipop born every second", which doesn't have quite the same ring as the original.

Edit: What annoys me more than censorship is incorrect autocorrect, but I think that's coming from my computer, even though I turned it off. It changed "they're" to "their" - good thing I noticed.

What a world we live in....you can be treated to the most vulgar filth in the world (I watched part of the AMA show last night, until Iggy won one too many awards :startled: ), but you can't say your wool sweater is *****ly, your children used to get *****ly heat in the summer, and your boss has a *****ly personality. We also now have to say my female Wheaton terrier is going to have puppies, and on and on. :grin:

I realize it's trivial, but language is being played with in much more serious ways as well. Propagandists know that if you change the meaning of a word you change perceptions and attitudes. It's mind control of the stupid.

OK. Enough off topic and enough preaching. :smile:

sparkey
24-11-14, 17:44
Although I don't put too much stock into hobbyist autosomal DNA analyses, it's very important to be skeptical of academic papers as well. We could easily devote a thread to mistakes and bad interpretations of data in academic papers. Keep in mind that it has taken academia much longer to come around to viewpoints that hobbyists have held for much longer. The best example of that effect is the consensus on the origin of haplogroup R1b in Europe. If academia was so slow to correct their Y-DNA analyses and hobbyists got things right much earlier, why should we think that the opposite will happen with how we understand autosomal DNA?