Talk:Haplogroup R1a1a

Underhill (2015) or snpedia.com?

@Iryna Harpy: which source is better for the tree: Underhill (2015) or snpedia.com? Joshua Jonathan -Let's talk! 15:50, 4 July 2016 (UTC)

Naming/coding

@Iryna Harpy: and is it correct that somewhat different codings/names can be found? For example R-Z93: R1a1a1b2 (ISOGG), R1a1a2* (Underhill 2015), R1a1a1h snpedia. Which system should be used? Joshua Jonathan -Let's talk! 15:59, 4 July 2016 (UTC)

And what about yfull.com? Joshua Jonathan -Let's talk! 18:18, 4 July 2016 (UTC)

@Joshua Jonathan: I'm not even going to pretend to be able to hold my hand to my heart on the details of the nomenclature, but yfull and snpedia are combinations of WP:PRIMARY and WP:USERGENERATED, therefore should be be treated as WP:NOR no-go zones.

Although there was no actual outcome for restricting the use of genetic studies for modern 'ethnic group' articles at the recent RfC you popped in on, there was a very strong theme of 'expert sources only' (along the lines of WP:MEDRS) for scientific publications surrounding genetic data. If we're repeating coding strings/names, we can't stretch beyond quality, peer reviewed research (i.e., publications like PLoS ONE, European Journal of Human Genetics, Nature (journal), etc.)

The POV pushing problems (and horrendous edit wars) arising over the last few years in Wikipedia revolve around ancient DNA, anthropology, linguistics, plus nationalist narratives being mashed together to create WP:SYNTH (with smoke and mirrors tactics thrown in in the form of dragging in blogs that are run by - and followed by - amateur enthusiasts) that looks convincing, but turns out to be WP:BOLLOCKS once it's carefully examined.

I do believe that the issue of sourcing, plus what content is significant for more contemporary 'ethnic group' articles will need to be addressed via an RfC properly sooner rather than later. Conflating 'empirical science' with good faith dabbling interest is not a good concoction for an encyclopaedic resource. At this stage, I'd prefer to defer to editors like Florian Blaschke and Mellsworthy who actually do have a working knowledge of the science behind a discipline still very much in its infancy. --Iryna Harpy (talk) 04:29, 5 July 2016 (UTC)

I've used ISOGG now as the main source, with additional info from yfull.com and snpedia to get Underhill, and some additional info from the Wiki-article, correctly into the ISOGG tree. Joshua Jonathan -Let's talk! 05:35, 5 July 2016 (UTC)

While I don't think scientists should hate ISOGG, and sometimes there is no better source, I think the specific issue of omissions in this page can be addressed with published literature from Nature, etc. I do not have the time to do this at the moment, so I'm just going to trust that Joshua Jonathan and others are doing the best they can for now. --Mellsworthy (talk) 08:56, 5 July 2016 (UTC)

@Joshua Jonathan:Just to be slightly helpful, I should say that, despite their popularity, it is generally better to use the more stable names like "R1a Z93" instead of "R1a1a1b2". This second style of naming is a compact representation of all the significant mutations/mutation sequences that differentiate the sub-haplotypes of R. But that means this style of name changes every time there is a new discovery of a mutation differentiating human groups that occurred before the last mutation included in the name, meaning that to correlate an old paper with newer info, you need to pay attention to which mutations are involved anyway (e.g. M17, Z93, etc.). As a compromise, many people include both. Hope this at least points towards less confusion. --Mellsworthy (talk) 09:08, 5 July 2016 (UTC)

@Mellsworthy: I've also used both, with references. The advantage of r1a1a etc, for laypersons, is that one can see the tree unfold. But the disadvantages are also clear, given the various codings. Anyway, the main point is the divergence between Z282 and Z93. Joshua Jonathan -Let's talk! 10:36, 5 July 2016 (UTC)

@Iryna Harpy: I guess you can say I have a working knowledge of the linguistic issues involved, and to a lesser extent (since it's not really part of my training, just amateurish dabbling) the archaeological side (although I have only very little idea of the raw data, all I really have is a really rough overview of European prehistory as reconstructed from them), but I don't nearly have a strong enough background in natural science and math to competently opine on the genetic aspect the way Mellsworthy can. @Joshua Jonathan: Iryna Harpy makes an important point. Blogs are just extremely dodgy and unnecessary as sources (unless verified to be run by an actual expert like John Hawks). I've actually received a private complaint about you using the Eurogenes blog as a ref. It's interesting, but I don't think the owner is a professional geneticist; he seems to be an amateur enthusiast as well. --Florian Blaschke (talk) 16:25, 5 July 2016 (UTC)

I'm just trying to figure out the tree and the coding; ISOGG seems to be okay, the rest is additonal. If I used Eurogenes, then it was for good reasons, I guess. yfull.com is commercial enterprise, as far as I can see. And well, private complaints: if people have complaints, they should share them here. Joshua Jonathan -Let's talk! 20:54, 5 July 2016 (UTC)

Ah yes, this note on Lazaridis et al. (2016): "See also eurogenes.blogspot, The genetic structure of the world's first farmers (Lazaridis et al. preprint) ." If I remember correctly, Lazaridis himself participated in the discussion there. Joshua Jonathan -Let's talk! 05:46, 6 July 2016 (UTC)

I second Florian Blaschke that we should not use blogs. I do think that Eurogenes usually has a perfectly reasonable viewpoint, but what is done there is not science--yet. I consider the sequence of Underhill articles on R1a to be perfectly good surveys of this narrow topic, and anything that comes after these should just be reported as additional information, in my opinion, and there's no call to violate WP:NOR or WP:Verifiability to go beyond that. I think the facts are clear, but they will come out in published form soon enough. We should remember that if there is no synthesis, we can summarize or transmit each reliable source, if they contradict each other, but not adjudicate them ourselves. I think that many of the paleogenomics papers themselves, where they include a synthesis of prior research and the implications of the current research on that literature, are sufficient for this kind of purpose in some cases. Unfortunately, for the purposes of this article, I don't think any of the paleogenomics papers with implications for R1a has a good survey of this topic. --Mellsworthy (talk) 06:51, 6 July 2016 (UTC)

Blogs are not being used as references here; only, in a few cases, as links within notes to provide access to discussions of the topic. Where there is confusion about the naming, I've mentioned various names in the tree, like R1a1a1b1a (Z282)[14] [R1a1a1a*] (Z282) (Eastern Europe)[19], putting ISOGG first, so readers can see what is what. In the article, I've tried to stick to the ISOGG naming. Anyway, the tree has been greatly improved, hasn't it? Joshua Jonathan -Let's talk! 07:32, 6 July 2016 (UTC)