User talk:Austronesier/sandbox3

Outline
I really have no broad picture about how to best organize the material. I think the Background section is a nice start, but we should let it culminate with the MA-1 find and then close with something like "Subsequently, several other individuals from the Upper Paleolithic were assigned to the same population group, which is now broadly referred to Ancient North Siberian".

Maybe we can rename "Origins" to "Ancient North Siberians" and elaborate on all core specimens from Yana RHS to AG3. I'm not sure how organize what follows. We could talk about the diverse ancient descendant groups like Kolyma/APS and Baikal EBA in the east and the ANE-EHG cline in the west, and how they contributed to the gene pool of European, (modern) Paleosiberian and Indigenous American populations. We should also not fail to mention the Tarim BA mummies which have about 75% ANE ancestry. Finally, we might also mention the ANE admixture into the (pre-)Jomons although this is probably a bit premature. Austronesier (talk) 20:03, 4 October 2022 (UTC)


 * I really have no broad picture about how to best organize the material - me neither!
 * let it culminate with the MA-1 find 👍
 * rename "Origins" to "Ancient North Siberians" hmm... might get confusing if we use that for everyone from Yana RHS to AG3, since that's also the name specifically used for just those YanaRHS fellows?
 * I'm not sure how organize what follows. well... we could either periodize it (Mesolithic, Neolithic, EBA...), or just have one big 'derived populations' section, or divide it by populations (EHGs, Indigenous Americans, Tarim...) I'm not sure myself. My own articles seem to be self-organizing, in the sense that if I rearrange them enough, at some point it looks organized just by pure random coincidence. I'll have a think about this anyway. Thanks for starting the article, it should turn out quite good I expect.  Tewdar   20:14, 4 October 2022 (UTC)
 * I haven't written review papers that often when compared to papers about my own research. This is also why I'm not really good at producing WP articles beyond stubs and start-class articles. Let's try the self-organizing approach which also better suits my own chaotic working-style 😁
 * I will have to check the current nomenclature about ANS and ANE. Some sources use "ANS" for the whole complex, but maybe the labeling ANS=Yana vs. ANE=MA1/AG3 is still more established. –Austronesier (talk) 20:53, 4 October 2022 (UTC)


 * You might be right that more sources use ANS for all of them now. Perhaps Origins > "Upper Palaeolithic" or something like that, if it turns out we're wrong? Just one possibility of course. Self organizing works great, you should see my wardrobe! 😂  Tewdar   21:03, 4 October 2022 (UTC)

I have added a scribbleboard, so we don't have to use the draft to quickly put down notes about things that might (or not) be relevant for the article. And btw, cheers to all who are here not for a dogma, but for faithfully reflecting good scholarship. The steppe hypothesis has become such a fetish; it's an important piece (the biggest chuck, actually) in the puzzle of IE expansion, but not a TOE. It seems that for many people complexity sucks. 😂 –Austronesier (talk) 10:29, 8 October 2022 (UTC)


 * Excellent idea, I won't move the 'Ancient Paleo-Siberians' scribbles to it, since I'll do that next, but I'll use it for all subsequent scribbling.
 * Article seems to be going quite nicely, I'd say. 😁  Tewdar   10:26, 8 October 2022 (UTC)
 * Yes, I must confess that my 'fetish' is for accurately reporting what scholarship actually says, not twisting it to reflect my own personal preferences. Sadly, Proto-Celtic probably doesn't originate from Harlyn Bay... 😂  Tewdar   10:36, 8 October 2022 (UTC)
 * Sorry, mate, but I'm afraid I'm potentially closer to it 😁. Goolge Maps says I can walk in 7 hrs to Glauberg (nah, not really the Celtic homeland, either). –Austronesier (talk) 19:44, 8 October 2022 (UTC)

I wonder how we can best talk about admixture in quantitative terms. E.g. Sikora et al. mention 22% EA ancestry based on admixture graphs, but a few sentences later talk about a 2:1 ratio. In Vallini et al. (2022), the EA ancestry goes up to 50%. The proposed ratios of ANE ancestry for EHG are even wilder, ranging from 8% to 80%. –Austronesier (talk) 20:55, 12 October 2022 (UTC)
 * Well, we could either do a 'depending on the model used, ancestry ranges from x% to y%' or just attribute (some of) the suggested proportions given in the various papers. If we put "ranges from 8% to 80%" we'd probably sound a bit silly 🤪, but I suppose it wouldn't really be our fault... We should probably also mention somewhere that EHG may turn out to be a sister lineage to ANE, not an admixed descendent.
 * So, as you can see, my answer is basically "errrr..." 😁 Hopefully I'll have a better answer for you tomorrow.  Tewdar   21:26(ish, manually added wtf?), 12 October 2022 (UTC)
 * On the subject of EHGs, have you seen this preprint? It's quite interesting.  Tewdar   10:36, 13 October 2022 (UTC)
 * The supplement is well worth reading too.  Tewdar   10:47, 13 October 2022 (UTC)
 * So... I actually think the descriptive paragraph about the ANE-WHG cline is not bad, and we should remove the "...75% of their ancestry from Ancient North Eurasians" bit, and maybe replace it with some percentages from actual samples eg Sidelkino, attributed to the study we're basing the information on. Does that sound like a reasonable approach?  Tewdar   11:45, 13 October 2022 (UTC)
 * Nice preprint! It will be interesting to see with subsequent samples from Mesolithic Russia if EHG remains a well-defined cluster pointing to some kind of temporary isolation, or if it just becomes a reference point in a continuous cline.
 * I still have to make up my mind about these admixture figures. We should keep in mind that many readers might take these figures at face value if presented in a prominent way (as e.g. in tables). That's why I have been experimenting with putting such details into notes. –Austronesier (talk) 18:17, 13 October 2022 (UTC)


 * I mean... we don't need to give any figures for EHG (SHG, UHG, LHG, WRHG...) at all in the main text, do we? We just say those groups are (probably) on a cline, more ANE-related in the east, less-to-none in the west, shove some percentages into the notes, and booyaaka! EHG isn't the main topic anyway...   Tewdar   18:25, 13 October 2022 (UTC)
 * Hmmm...I don't think we can fully avoid figures/proportion/ratios since it still makes a difference to talk about e.g. EHG and Central Steppe populations like Botai as predominantly ANE, and OTOH about e.g. Yamnaya-derived populations and APS as carrying significant, but not predominant ANE ancestry. In Xinjiang and the Central Steppe it was less a cline but a rapid "dilution" due to the increase of mobility in the BA. –Austronesier (talk) 12:58, 17 October 2022 (UTC)
 * Perhaps, for EHG, we should just pick a few figures from the range, with attribution, since (even if we exclude the Wang outlier) we still have 40-80% ANE depending on which model we use? That might be the best approach for all these groups..?
 * I started plotting ancient individuals on a map, but the built in Wikitools don't seem particularly suited to this task... I might knock something up in ArcGIS if I get a chance. Be nice to have a map with some arrows, etc.  Tewdar   13:12, 17 October 2022 (UTC)

Have you seen anything more recent about the Mansi than Wong et al. (2017). Or more generally about ANE ancestry proportions? It's just 'dumb' TreeMix, so these figures are a bit more problematic than the ones from models with manually checked topologies. Tambets et al. (2018) have a mechanical evaluation of ancestry proportions for Uralic populations with a fixed qpGraph topology, but ANE ancestry is hidden in the two EHG portions and (at a smaller rate) in the "Siberian" (= Nganasan) portion (Fig 6). –Austronesier (talk) 18:07, 22 October 2022 (UTC)


 * Not yet... I was actually just going through the sources I have in order, making notes about whatever they say about modern populations, then see whether we have anything passably admissible at the end. Direct proportions from e.g. MA1, AG3 would be nice, but... I'll keep looking, hopefully I'll have more time tomorrow, or next week.  Tewdar   19:09, 22 October 2022 (UTC)
 * Many recent studies usually derive modern gene pools from ANe-derived sources rather than directly from ANE. Dai et al. (2022) is a nice case study for tracing Tarim mummy-ancestry in modern Central Asian populations. You will like Fig. S13. It's a pity they don't have included samples from Nganasan, Mansi etc. I'd expect that they share more drift with Tarim_EMBA1 that the Pamir Tajiks. –Austronesier (talk) 19:57, 24 October 2022 (UTC)
 * I think the model fits are usually much better with less distant sources than MA-1, AG3, etc. You're right, I did like fig. S13, except we probably can't do much with it. And can't these people hit the 'export pdf' button? docx, indeed! What next, LaTeX source files?!  Tewdar   20:37, 24 October 2022 (UTC)


 * What are we going to call this article? We can't really call it 'The genetic history of East Siberia' because a lot of the article is about South Siberia etc. 'Genetic history of Siberia' is probably too broad. I'd like to call it 'Genetic history of peoples who split off from West Eurasians c. 40kya and then admixed with Tianyuan-related populations" but that's not very catchy. Got any good ideas?  Tewdar   19:08, 26 October 2022 (UTC)
 * "Genetic history of ancient northeast Eurasia", maybe? NE Eurasia is more vague than Siberia, E Siberia etc. With such a title we could keep our focus on ANS and descendants, Ancient Americans (at least their initial formation period) and APS (and descendants). In any case, we should have "ancient" in the title, otherwise we will also have to cover the Tungusic-Turkic-Mongolian expansion, which IMO is a completely different topic. –Austronesier (talk) 19:58, 26 October 2022 (UTC)


 * "Genetic history of ancient northeast Eurasia" sounds pretty good. I can't think of anything better anyway. 😁👍  Tewdar   20:27, 26 October 2022 (UTC)

Unfortunately, I have no institutional access to Current Opinion in Genetics & Development. Do you have an OA link to Zhang & Fu, or can you kindly share a copy with me? –Austronesier (talk) 21:20, 27 October 2022 (UTC)
 * My institution has access! 😁 Did you get my email?  Tewdar   21:59, 27 October 2022 (UTC)
 * Thanks, it works! –Austronesier (talk) 05:36, 28 October 2022 (UTC)
 * Great, let me know if you need anything else.  Tewdar   08:40, 28 October 2022 (UTC)
 * The presentation of data in Zhang & Fu is good, but the writing is terribly technical. This is exactly the kind of style we should try not to emulate. Yang's review article (which is promoted by WCF in so many articles as the ultimate wisdom about Asian ancestry) is a great example of clarity and accessibility for non-specialist. There's some bits in it about ANS/ANE too. –Austronesier (talk) 18:01, 28 October 2022 (UTC)
 * Yes, Yang's review is very nice, I already added it to my to-do pile for adding to the article, if I ever get an evening where I don't have to spend 95% of it pretending to be a pirate or something... (aaar, where's me bottle o' rum? 🍾😁)  Tewdar   19:27, 28 October 2022 (UTC)
 * Off-topic here, but can you give me an exact IPA of "aaar" in your ideolect? 😁 –Austronesier (talk) 22:08, 29 October 2022 (UTC)
 * It's [ˈaːːːːːːːːːːːɹ]. Does IPA have a symbol for 'very drunk'? 🤣  Tewdar   22:32, 29 October 2022 (UTC)
 * We could suggest one for Extended IPA 😁 –Austronesier (talk) 17:45, 30 October 2022 (UTC)

Whenever I'm fiddling with the ANE-article after other(s) have fiddled with it, I woefully remember this work-in-progress. Short question: are you aware of a study that explicitly tracks the East Eurasian ancestry of ANS/ANE all the way into EHG? Many studies talk about the East/West admixture ratio in ANS/ANE, many also discuss the relation of ANE to EHG and other West Eurasian populations, but I don't know of any published one that links these two topics. So far I have only seen this in the Allentoft preprint. –Austronesier (talk) 17:52, 16 January 2023 (UTC)


 * Not that I can remember (standing in a field in the dark at this moment so a bit hard to do a search, but I don't think there is...)


 * Guess the peer review for Allentoft et al. is taking a while. It is absolutely huge after all.


 * I think our wip is not too bad. Hopefully we'll get it finished one day. I keep getting distracted though, I guess you do too! 😁  Tewdar   18:49, 16 January 2023 (UTC)
 * Guess what. While doing some reading to give the Lazridis (2018)-fan a piece of guidance in Talk:Natufians, I have come across Fig S19.A in this paper. Seek not and you shall find! 😁 –Austronesier (talk) 17:02, 18 January 2023 (UTC)


 * Great news! I actually don't have that paper, I'll have to get it from the, uh, "library" later.😉 I'll paste up the Bell Beaker genetics in a sandbox later for you to take a look at.  Tewdar   17:07, 18 January 2023 (UTC)


 * That's an interesting model. EHG = East Asian + WHG related, leaving out the middlemen...  Tewdar   18:26, 18 January 2023 (UTC)
 * It works and even gets a Z < |3|!
 * Here's another (but lesser) objet trouvé that will certainly be of great interest to my esteemed Celtic wikiconfrère ( will love this too):
 * Waddafuq! It's in the lede of Tarim mummies to support the statement: "Victor H. Mair's team concluded that the mummies are possibly speakers of Indo-European languages such as the Tocharians". It's not entirely clear if Coonan made this up like the Celts stuff, or if he actually paraphrases something he heard from Mair without attributing to him. Whatever, my fingers are starting to feel a TNT-itch! 😁 –Austronesier (talk) 22:21, 22 January 2023 (UTC)
 * Ah yes, "The Independent", that scholarly journal in which scientists diligently and carefully report on the latest findings in anthropology, genetics, and, indeed, climate science (where I seem to recall them suggesting all life on earth would be totally wiped out by a 3 degree rise in global temperatures, which is surely a totally accurate description of what the climate models are suggesting with not a bit of embellishment).
 * I used to read teh Independent sometimes when it was a Newspaper. Even then it was a bit nutty, but since it went wholly online, I treat it with much the same level of distrust as the Daily Mail. And as for people using newspapers as sources in scientific articles... the emoji I am reaching for should be angry cheeked, facepalming, and puking all at the same time, but it doesn't exist! :) Sirfurboy🏄 (talk) 08:00, 23 January 2023 (UTC)
 * Try putting in a request here. Don't let Michael Everson find out about it, he will probably veto it... ☹️💩👎  Tewdar  13:24, 23 January 2023 (UTC)
 * We should write the geophysical characteristics of planet Earth section in a similar fashion to the Tarim archaeogenetics: A 24th century BC study by an anonymous Egyptian author concluded that the Earth is flat. This was supported in later 1st millennium BC studies by Homer and Hesiod. However, a 6th century BC study by Anaximander concluded that, in fact, the Earth was a short cylinder with a circular top. This was disputed in yet another 6th century study by Pythagoras, who argued that the Earth was actually a sphere. Yet another 3rd century BC study by Eratosthenes suppprts this view. However, a 17th century study by Pope Urban VIII concluded...  Tewdar   09:30, 23 January 2023 (UTC)
 * Yeah, it really looks like a upper midbrow version of the Daily Mail. And let's do an RfC in Talk:Earth, Genetics sections could be a trendsetter for a better WP. –Austronesier (talk) 20:03, 23 January 2023 (UTC)
 * Mandatory Scribbleboard sections in all articles...  Tewdar   20:41, 23 January 2023 (UTC)
 * I don't know if I ever will take this draft up again (for obvious reasons, I'm still waiting for the Seima-Turbino paper to come out), but I feel a bit dumb in retrospect for having overlooked Steppe Maykop in Wang (2019). I had only noticed this western extension of ANE ancestry (apart from EHG) with Mereke from the Central_Steppe_EMBA cluster in Narasimhan et al. (Mereke is really close to the Volga basin). I only have come across Steppe Maykop now in the discussion of excess ANE-ancestry in the Eneolithic European steppe in the new Reich-laboratory preprint. Have you already had the time to read it all and fully digest it? –Austronesier (talk) 11:54, 9 May 2024 (UTC)
 * How did we forget that, they literally drew us a picture! I've read the new preprint and supplement, dunno about 'fully digested' yet...they talk a bit about high 'Siberian' ancestry and 'unusual' formation of Steppe Maikop population as far as I remember. I completely forgot about this draft to be honest. It's actually pretty good, I think, maybe we'll get it in article space some day.
 * I suspect the next article about Indo-European origins will put the Urheimat literally on the Caucasus Mountains, instead of one side or the other...  Tewdar   18:03, 9 May 2024 (UTC)
 * I suspect the next article about Indo-European origins will put the Urheimat literally on the Caucasus Mountains, instead of one side or the other...  Tewdar   18:03, 9 May 2024 (UTC)

Ancient populations

 * "North Eurasian Early Holocene cline"
 * Narasimhan (2019): WHG-EHG-WSHG-ESHG


 * EHG-EA cline
 * Damgaard (2018): Botai_CA, CentralSteppe_EMBA, Okunevo_EMBA, Baikal_EN and Baikal_EBA genetic cline from EHG to N/BA Baikal HGs in Central Asia (who are genetically close to DG)


 * EHG
 * Haak et al. (2015): 38–40% ANE (MA-1), 60–62% WHG (Fig S8.6). (Alternative topologies where EHG and ANE are unadmixed sister lineages, with WHG being admixed, are not rejected)
 * Lazaridis et al. (2016): 80% ANE (MA-1), 20% WHG (Fig S4.11).
 * Wang et al. (2019): 9% ANE + 91% WHG-related (qpGraph)
 * Sikora et al. (2019): 44.5±11% ANE (MA-1), 21.1±4% CHG, 34.4±11% WHG (qpAdm). Modeled with qpGraph as two successive admixture events: ghost "EHG1" (78% MA-1-related + 22% CHG-related), then 76% ghost EHG1 + 24% WHG-related
 * Lazaridis 2018 The evolutionary history of human populations in Europe. "Eastern European hunter-gatherers (EHG), a population of mixed WHG and Upper Paleolithic Siberian ancestry (related to the Mal’ta and AfontovaGora specimens from Lake Baikal"... "What drove ancient Siberian populations westward to produce the European/Siberian intermediate EHG?"
 * WSHG ('West_Siberia_N')
 * Narasimhan (2019, "While no model fits according to our acceptance criteria, one model works marginally with p-value 0.003"): 73% ANE, 20% EHG, 6% ESHG (no detectable ancestry from WHG and Anatolian farmers)
 * Zhang (2021): "High-ANE" groups can successfully modeled with Tarim_EMBA1 as ANE source: 68.3±7.4% AG3, 19.7±6.1% EHG, 12.1±2.3% Devil's Gate (p=0.903) or 67.1±3.0% Tarim EMBA_1, 32.9±3.0% EHG (p=0.392)


 * ESHG ('Baikal HGs', 'Baikal_EN')
 * Damgaard (2018): Baikal_LNBA = "around ~5–11%" ANE (= MA1), the rest Baikal_EN (source of AEA) (table s12(2): 92% Baikal_EN, 8% MA1). Part of a cline from EHG to Baikal_EN. In their model, Baikal_EN is an unadmixed East Asian sister lineage of Han.
 * Narasimhan (2019): "Represented by sixth millennium BCE hunter-gatherers from the Lake Baikal region with ancestry deeply related to East Asians"
 * Jeong (2020): "Baikal_EN has been modeled to have ~10% Ancient North Eurasian (ANE) ancestry"
 * Yu (2020): Predominantly "Northeast Asian" (Devil's Gate), with 14.3% ANE (AG3) ancestry


 * Botai
 * Damgaard (2018): 88% ANE + 12% East Asian in a very simple qpGraph model (Fig S28) with unadmixed ANE, EHG and CHG.
 * Narasimhan (2019): 90% WSHG (= HG from Tyumen), 10% ESHG from Lake Baikal; but - "We do not have a model for Botai.SG that works given the thresholds for our modeling procedure"; Botai individuals lie between WSHGs and ESHGs on PCA, with similar components to WSHG but with higher ESHG ancestry
 * Jeong (2019): Botai closely related to MA-1 & AG3, later substantially contributed to Okunevo, f4 stats suggest EA gene flow into Botai, ANE-related ancestry in Botai is intermediate between EHG and AG3, corresponding to geographic position, suggesting "a genetic cline of decreasing ANE-related ancestry stretching from AG3 in Siberia to WHG in Western Europe" - but subst. EA contribution offsets Botai from ANE-WHG cline
 * Yu (2020): Botai is used as proxy for ANE ancestry in Central Steppe MBA populations
 * Zhang (2021): "High-ANE" groups can successfully modeled with Tarim_EMBA1 as ANE source. "Botai _CA shows a similar profile [as West_Siberia_N] but requires an additional Eastern Eurasian contribution": 63.9±6.8% AG3, 17.4±5.7% EHG, 18.6±2.0% Devil's Gate (p=0.147) or 69.5±5.7% Tarim EMBA_1, 24.9±3.4% EHG, 5.6±2.9 Baikal_EBA (p=0.257)


 * Tarim EMBA
 * Zhang (2021): Tarim_EMBA1 has 72% ANE (AG-3), 28% Baikal_EBA; Tarim_EMBA2 = Tarim_EMBA1 with additional 11% Baikal_EBA. F3-values suggest "that the Tarim mummies are currently the best representative of the pre-pastoralist ANE-related population that once inhabited Central Asia and southern Siberia"
 * Kumar (2022): ANE ancestry has persisted in Tarim and Dzungaria into the historical era via Tarim_EMBA1

Modern populations

 * Wong (2017)
 * ANEs MA-1 and AG-2 grouped with Western Siberians (Mansi and Nenets), suggesting a genetic link between ANEs and modern Western Siberians
 * Treemix estimated ANE lineage is 41% of the ancestry of Andean Highlanders
 * Mansi, Khanty, Nenets, and Native Americans substantial ANE ancestry, genetic affinities with MA-1 and AG-2.
 * Mansi people have significant ancestry from ANE (~ 57%)


 * Jeong (2019)
 * ANE-related contribution (AG3) is required for Enets, Selkups, Kets and Mansi
 * Mansi: p-Val 0.112, Nganasan 0.541 ± 0.012, Srubnaya 0.375 ± 0.020, AG3 0.084 ± 0.024
 * Selkup: pval 0.018 Nganasan 0.652 ± 0.011 Srubnaya 0.188 ± 0.018 AG3 0.160 ± 0.023
 * Enets: pval 0.194 Nganasan 0.789 ± 0.016 Srubnaya 0.172 ± 0.025 AG3 0.039 ± 0.031
 * Ket: pval 0.044 Nganasan 0.700 ± 0.011 Srubnaya 0.174 ± 0.019 AG3 0.126 ± 0.025
 * Chelkans: pval 0.521 Srubnaya 0.320 ± 0.022 Ulchi 0.573 ± 0.012 AG3 0.107 ± 0.027
 * Bashkir_south: pval 0.301 Srubnaya 0.562 ± 0.016 Ulchi 0.409 ± 0.008 AG3 0.029 ± 0.019
 * Khakass_Kachins:pval 0.574 Srubnaya 0.221 ± 0.020 Ulchi 0.724 ± 0.010 AG3 0.055 ± 0.024
 * Khakass_Koibals:pval 0.862 Srubnaya 0.198 ± 0.021 Ulchi 0.705 ± 0.012 AG3 0.097 ± 0.024
 * Khakass_Sagai:pval 0.293 Srubnaya 0.276 ± 0.018 Ulchi 0.614 ± 0.010 AG3 0.110 ± 0.022
 * Tatar_Siberian: pval 0.309 Srubnaya 0.475 ± 0.016 Ulchi 0.450 ± 0.008 AG3 0.075 ± 0.018
 * Todzin: pval 0.429 Srubnaya 0.100 ± 0.026 Ulchi 0.835 ± 0.013 AG3 0.065 ± 0.031
 * Tubalar1: pval 0.899 Srubnaya 0.291 ± 0.017 Ulchi 0.572 ± 0.009 AG3 0.137 ± 0.020
 * Tubalar2: pval 0.339 Srubnaya 0.369 ± 0.023 Ulchi 0.538 ± 0.012 AG3 0.093 ± 0.027
 * Tuvinian: pval 0.637 Srubnaya 0.153 ± 0.014 Ulchi 0.806 ± 0.008 AG3 0.041 ± 0.017


 * Flegontov (2016)
 * A global maximum of ANE (MA1) ancestry occurs in Native Americans
 * lower levels in pops w/ more recent Beringian origin, eg Chukotka, Kamchatka, Aleutian Islands, American Arctic
 * In Europe, ANE contribution highest in Baltic region, East European Plain, North Caucasus.
 * Kets "might represent the peak of ANE ancestry in Siberia"
 * model Kets as a two-way mixture of EA and ANE
 * ANE ancestry in Kets estimated c. 27-43%
 * ANE ancestry in various Native American groups c. 25–53%


 * Lazaridis (2014), 'Ancient human genomes suggest three ancestral populations for present-day Europeans' (3-way mixture of EEF, WHG, and ANE)
 * Albanian 12.7%
 * Ashkenazi_Jew 6.9%
 * Basque 11.4%
 * Belarusian 15.1%
 * Bergamo 10.8%
 * Bulgarian 14.1%
 * Croatian 14.5%
 * Czech 16.7%
 * English 14.1%
 * Estonian 18.3%
 * French 13.5%
 * French_South 13%
 * Greek 15.1%
 * Hungarian 17.9%
 * Icelandic 15%
 * Lithuanian 17.2%
 * Maltese 6.8%
 * Norwegian 16.1%
 * Orcadian 15.8%
 * Sardinian 0.8%
 * Scottish 18.2%
 * Sicilian 9.7%
 * Spanish 12.3%
 * Spanish_North 16.3%
 * Tuscan 11.8%
 * Ukrainian 15.1%