PDA

View Full Version : G25 nMonte - Inferring ANA/Eurasian layers in SSA pops



Angoliga
06-24-2019, 09:26 AM
We're limited on available aDNA but thought we could speculate using nMonte on these interrelated topics:



Do the South Sudanese like the Dinka have West Eurasian/North African admixture? (https://anthrogenica.com/showthread.php?16142-Do-the-South-Sudanese-like-the-Dinka-have-West-Eurasian-North-African-admixture)

Ancient Eurasia K6 - Deducing Traces of Back to Africa Migrations (https://anthrogenica.com/showthread.php?9831-Ancient-Eurasia-K6-Deducing-Traces-of-Back-to-Africa-Migrations)

How closely related are Nilotics to Bantu and West African Niger-Congo peoples? (https://anthrogenica.com/showthread.php?17284-How-closely-related-are-Nilotics-to-Bantu-and-West-African-Niger-Congo-peoples)

Where exactly did nilo saharans originate from? (https://anthrogenica.com/showthread.php?14511-Where-exactly-did-nilo-saharans-originate-from)





After playing with some runs, I found "MWI_Fingira_6100BP" worked as a good crude proxy for SSA -- that includes West-African as well as the Ancient-East-African (Nilotic) component.
Moreover, the sample "MWI_Fingira_2500BP" seemed to correlate nicely as a layer of archaic SSA ancestry (Pygmy/South HG).



MWI_Fingira_6100BP = SSA
MWI_Fingira_2500BP = Arachaic SSA


Using this a basis to tease out deep Eurasian layers, I removed all modern and ancient SSA pops along with modern Saharans while leaving all other G25 pops to see what happens:




[1] "distance%=31.2193"

Yoruba

MWI_Fingira_6100BP,70.6
MAR_Iberomaurusian,15.8
MWI_Fingira_2500BP,8.5
MAR_EN,4.7
BLZ_SakiTzul_7400BP,0.2
CHL_LosRieles_5100BP,0.1
Anatolia_N_Kumtepe_low_res,0.1


[1] "distance%=28.1667"

Gambian

MWI_Fingira_6100BP,68.7
MAR_Iberomaurusian,19
MWI_Fingira_2500BP,7.3
MAR_EN,4.6
Anatolia_N_Kumtepe_low_res,0.3


[1] "distance%=25.7217"

Dinka

MWI_Fingira_6100BP,72.5
MAR_Iberomaurusian,20.7
MAR_EN,4.5
MWI_Fingira_2500BP,2
Eskimo_Sireniki,0.1
BRA_Laranjal_6700BP,0.1
PER_Lauricocha_3500BP,0.1


[1] "distance%=28.2958"

Mandenka

MWI_Fingira_6100BP,69.6
MAR_Iberomaurusian,19.6
MWI_Fingira_2500BP,6.4
MAR_EN,4
Anatolia_N_Kumtepe_low_res,0.2
BLZ_SakiTzul_7400BP,0.1
PER_Laramate_900BP,0.1

[1] "distance%=29.943"

Igbo

MWI_Fingira_6100BP,69.5
MAR_Iberomaurusian,16
MWI_Fingira_2500BP,10.9
MAR_EN,3.3
Anatolia_N_Kumtepe_low_res,0.3


[1] "distance%=23.0629"

UG_CentralSudanic

MWI_Fingira_6100BP,72.3
MAR_Iberomaurusian,16.6
MWI_Fingira_2500BP,7
MAR_EN,3.1
PER_Lauricocha_3500BP,0.4
CHL_LosRieles_5100BP,0.2
USA_Spirit_Cave_11000BP,0.2
BRA_Laranjal_6700BP,0.1
PER_Laramate_900BP,0.1


[1] "distance%=27.867"

Kongo

MWI_Fingira_6100BP,67.8
MAR_Iberomaurusian,16.9
MWI_Fingira_2500BP,14.1
MAR_EN,1
Anatolia_N_Kumtepe_low_res,0.2


[1] "distance%=23.1196"

Biaka

MWI_Fingira_2500BP,74.7
MWI_Fingira_6100BP,20.6
Anatolia_N_Kumtepe_low_res,2
MAR_Iberomaurusian,1.9
MAR_EN,0.7
IRN_Wezmeh_N,0.1


[1] "distance%=11.342"

Ethiopian_Tigray

Levant_Natufian,45.6
MWI_Fingira_6100BP,29.8
MAR_Iberomaurusian,14.9
IRN_Tepe_Abdul_Hosein_N,7.4
IRN_Wezmeh_N,2
MAR_EN,0.3


[1] "distance%=15.1951"

Somali

MWI_Fingira_6100BP,40.1
Levant_Natufian,29
MAR_Iberomaurusian,24.1
IRN_Tepe_Abdul_Hosein_N,5.2
IRN_Wezmeh_N,1.2
Canary_Islands_Guanche,0.3
MAR_EN,0.1



[1] "distance%=1.3255"

MWI_Hora_9000BP

MWI_Fingira_2500BP,66.5
MWI_Fingira_6100BP,26.7
CZE_N,0.9
Scotland_N_o,0.9
IRN_Wezmeh_N,0.6
Koinanbe,0.4
Kosipe,0.4
IRN_Ganj_Dareh_N,0.4
IRN_Tepe_Abdul_Hosein_N,0.3
Iberia_Southwest_EN,0.3
IRN_Seh_Gabi_LN,0.2
ITA_Remedello_BA,0.2
Corded_Ware_CZE_o,0.2
UKR_Globular_Amphora,0.2
Bell_Beaker_HUN_EBA_o,0.2
WHG,0.2
Iberia_Northeast_CA,0.2
Levant_Natufian,0.1
Kurumba,0.1
IRN_Seh_Gabi_C,0.1
Levant_LBN_MA_NE,0.1
IRN_Hajji_Firuz_IA,0.1
Anatolia_Kumtepe_N,0.1
IRL_MN,0.1
Iberia_Southwest_MLN,0.1
Iberia_Southeast_MLN,0.1
DEU_Baalberge_MN,0.1
HUN_Baden_LCA,0.1
HRV_Vucedol,0.1
MAR_Iberomaurusian,0


[1] "distance%=2.0847"

EGY_Late_Period

Levant_ISR_MLBA,26.2
Levant_PPNC,22
Levant_ISR_C,16.6
Levant_Natufian,15.9
TKM_Parkhai_MBA,7.1
MAR_Iberomaurusian,5.5
Anatolia_Kaman-Kalehoyuk_MLBA_low_res,3.2
GEO_CHG,2.4
Syrian,0.5
MWI_Fingira_6100BP,0.4
Nganassan,0.2



1st Modern North-African Attempted:

[1] "distance%=2.2003"

Mozabite

Canary_Islands_Guanche,72.5
MAR_Iberomaurusian,9.5
Anatolia_Kumtepe_N,4.3
MWI_Fingira_6100BP,4.2
Anatolia_Tepecik_Ciftlik_N,2.8
Anatolia_N_Kumtepe_low_res,2.4
Anatolia_Kaman-Kalehoyuk_MLBA_low_res,1.6
HRV_Vucedol,0.9
DEU_MA_o,0.7
TKM_Parkhai_MBA,0.4
MAR_EN,0.4
Levant_ISR_MLBA,0.1
Levant_LBN_MA_NE,0.1
GRC_Peloponnese_N,0.1


...results with Guanche removed:


[1] "distance%=2.937"

Mozabite

MAR_Iberomaurusian,33.6
Anatolia_Kumtepe_N,16
HRV_Vucedol,11.7
MWI_Fingira_6100BP,7.7
Levant_ISR_MLBA,6.3
TZA_Zanzibar_First_Millenium,6
Anatolia_N_Kumtepe_low_res,5.1
HUN_Tisza_LN,4.5
MAR_EN,3.7
DEU_MA_o,3.3
Anatolia_Kaman-Kalehoyuk_MLBA_low_res,1.9
Wales_CA_EBA,0.2

...relatively low "MAR_EN" due to multiple intrusive migrations?





MAR_Iberomaurusian = ANA\BE
MAR_EN = EEF? R1B Carriers?




Observations:



MWI_Fingira_2500BP:


These levels looks out of place for these West-African pops (Gambian, Igbo, Mandenka ...) -- could this be traces of an archaic West-Central African HG?




MAR_Iberomaurusian:


Both West-African and Nilotic pops are getting significant ANA\BE-like ancestry, with the Dinka only slightly higher.

I was expecting the Dinka to have substantially more based on lower ks showing them having ~20% Eurasian ancestry and negligible amounts for YRI -- maybe this will get corrected after we get more aDNA... after all we now know Yourba are ~12% Eurasian but we have yet to see this with ADMIXTURE runs. Since there's not much difference between WAs and Nilotes on pretty much all PCA plots, I suppose the similar ratio of Eurasian ancestry should be expected?

Very marginal pull to Eurasians when WAs and Nilotic pops are compared:
https://i.imgur.com/19xt8hx.png

https://i.imgur.com/kzPk5Mm.png
Would this reinforce the idea that West-Africans are direct offshoots of AEAs?
Could it be an archaic West-Central African HG missing in the puzzle? I wonder if this would account for the embedded West-African ADMIXTURE in the Dinka.


The fits are horrible as expected but it's the ratio of SSA to Eurasian that might give the results some credence. When broken down, these results are holding the ~80/20 (EUR/SSA) split on lower ks for Nilotes.

@K33, based on these numbers -- the Basal ancestry does look skewed for West-Africans having more recent ANA "MAR_EN" and Nilotes having more





MAR_EN:


this appears to be omnipresent in SSA pops except Horners and ancient South-Africans (MWI_Hora_9000BP).


Would this imply, Nilotes like the Dinka acquired MAR_EN during the green sahara period? This theory might have some merit if we agree the origin of Nilotes was in a more West/Central region of the southern-Sahara; @Beyoku I recall you proposing this here (https://anthrogenica.com/showthread.php?14511-Where-exactly-did-nilo-saharans-originate-from&p=412768&viewfull=1#post412768)


@ChadRohlfsen, would this validate your theory of European farmer ancestry entering North Africa? The abstract from the Fregel 18' paper (https://www.pnas.org/content/115/26/6774) seems to confirm it "Among Eurasian ancient populations, Early Neolithic Moroccans are distantly related to Levantine Natufian hunter-gatherers (∼9,000 BCE) and Pre-Pottery Neolithic farmers (∼6,500 BCE).".


It's exclusion among Horners might explain why Nilotes/Central-Sudanics get ~5% "Farmer" ancestry on FTDNA's ancient origins and "Natufian" on calcs like Ancient Eurasia K6 yet fail to get any non-negligible "Western-Semitic"-like ancestry on modern calc's like puntDNAL k8, ancestry, 23andme etc.
For lack of a MAR_EN reference, these calcs are forcing the minor Mediterranean-EEF ancestry of Central-Sudanics (and other nilotes) into "Natufian".
Some more reference to this ~5% Eurasian affinity mentioned here (https://anthrogenica.com/showthread.php?16142-Do-the-South-Sudanese-like-the-Dinka-have-West-Eurasian-North-African-admixture&p=549513&viewfull=1#post549513)


Here's GEDmatch Archaic results for UG_CentralSudanic (T068690), the Eurasian traces are minor but clearly present (0.5cm) for LBK, HN_BR2/NE1 -- as well as a handful of other Eurasian ancients (Usht-Ism, Loschbur...):
https://i.imgur.com/ouuATRO.png
Same goes for the Dinka (#Z022730):
https://i.imgur.com/wWcClOA.png

I tried running modern North-Africans and Berbers to validate a plausible E-W cline for MAR_EN but ran into some trouble, their similar NW-African ancestry had Canary_Island_Guanche eating up "MAR_EN" so I removed it -- and things got complex... wayy too many reference samples for their ADMIXTURE breakdown. I'll have to revisit later -- their results are at the bottom of the code.

Angoliga
06-26-2019, 01:28 AM
If Lazardis' model is correct regarding the ANA-like affinity in the Yoruba (12.5%±1.1%), then these nMonte runs seem to hold some merit.



Moreover, our model predicts that West Africans (represented by Yoruba) had 12.5%±1.1% ancestry from a Taforalt related group rather than Taforalt having ancestry from an unknown Sub-Saharan African source; this may have mediated the limited Neanderthal admixture present in West Africans. (Lazaridis, 2018 (https://www.biorxiv.org/content/biorxiv/early/2018/09/20/423079.full.pdf))




Based on ADMIXTURE, Taforalt and IAM samples (MAR_Iberomaurusian, MAR_EN) have ~about a third SSA-like affinity:

https://i.imgur.com/INQKYIi.png
Fregel et al., PNAS (2018) (https://www.pnas.org/content/115/26/6774)








Yoruba

MWI_Fingira_6100BP,70.6
MAR_Iberomaurusian,15.8
MWI_Fingira_2500BP,8.5
MAR_EN,4.7
BLZ_SakiTzul_7400BP,0.2
CHL_LosRieles_5100BP,0.1
Anatolia_N_Kumtepe_low_res,0.1

15.8 + 4.7 = 20.5
20.5 x .66 = 13.53 vs. 12.5%±1.1%



Graph model from the afmd Lazardis' paper:
https://i.imgur.com/2pgGqP4.png

Jm8
06-26-2019, 07:33 PM
"YRI -- maybe this will get corrected after we get more aDNA... after all we now know Yourba are ~12% Eurasian"

According to Lazaridis the Yoruba (or at least those sampled) are about 12% Iberomaurusian-like/Taforalt-like, and he models Taforalt (the Iberomaurusians from Northern Morocco) as about half (45-50%) ANA and about half (50-55%) Eurasian. The ANA (Ancient North African) population is an African one (albeit one native to North Africa and distinct from other populations) rather than an OOA/Eurasian population, and is more basal than BE/Basal Eurasian. So the Yoruba, according to Lazaridis would seem to be about 6-7% Eurasian (rather than 12%).

It also seems likely that, if Taforalt from northern Morocco was about 50% ANA and half Eurasian, more southerly Iberomaurusian groups (e.g. in parts of North Africa significantly south of the Mediterranean coast region) would likely have had considerably more ANA (and perhaps sometimes other African) ancestry. It would be interesting too see the autosomal and uni-parental results for (from the remains of) a range of Iberomaurusian and putatively Iberomaurusian-related populations from around North Africa and nearby.

Angoliga
06-26-2019, 10:48 PM
"YRI -- maybe this will get corrected after we get more aDNA... after all we now know Yourba are ~12% Eurasian"

According to Lazaridis the Yoruba (or at least those sampled) are about 12% Iberomaurusian-like/Taforalt-like, and he models Taforalt (the Iberomaurusians from Northern Morocco) as about half (45-50%) ANA and about half (50-55%) Eurasian. The ANA (Ancient North African) population is an African one (albeit one native to North Africa and distinct from other populations) rather than an OOA/Eurasian population, and is more basal than BE/Basal Eurasian. So the Yoruba, according to Lazaridis would seem to be about 6-7% Eurasian (rather than 12%).

It also seems likely that, if Taforalt from northern Morocco was about 50% ANA and half Eurasian, more southerly Iberomaurusian groups (e.g. in parts of North Africa significantly south of the Mediterranean coast region) would likely have had considerably more ANA (and perhaps sometimes other African) ancestry.



Valid point -- distinction should be made when referring to non-SSA (ANA\Basal Eurasian) contribution in SSA pops
Given the lack of aDNA, it's still quite fascinating that nMonte's fits manages to mirror Lazardis' model for a non-SSA pop into Yoruba (12.5%±1.1%)



Here's a barebone run for "MAR_EN" (IAM), I was curious to see which modern SSAs nMonte would select:

- I removed modern North-Africans/Saharans and admixed East-Africans modern and ancient



[1] "distance%=7.8563"

MAR_EN

MAR_Iberomaurusian,69.2
Levant_Natufian,20.3
Mandenka,5.1
Gambian,1.3
Dinka,1.3
Levant_PPNC,1
RUS_West_Siberia_N_low_res,0.6
Iberia_Southeast_c.5-8CE,0.4
Koinanbe,0.3
Australian,0.2
Papuan,0.1
USA_NM_Chaco,0.1
AFG_Darra_i_kur_low_res,0.1

Geographically, the Mandenka are a reasonable choice given their West-Central African presence in and around Mali.
At minor levels, nMonte's next SSA choice for neighboring Gambians is also appropriate along with the minor Dinka given their supposed presence further east in the southern-central green sahara.

Maybe one day we'll be able to compare these SSA proportions for Kiffians "southerly Iberomaurusian groups"

Brwn_trd
06-27-2019, 11:32 PM
Do you have a Fulani sample? I’d imagine their numbers would be very interesting given the Fischer component which peaked in them in the recent Tishkoff paper? The Mende are also quite mysterious, they have a larger than expected CHG component, larger than a lot of more easterly groups despite the fact that there is no know history of pygmies that far west in their current teritory

Angoliga
06-28-2019, 02:48 AM
Do you have a Fulani sample?

Unfortunately, there's no G25 cords (https://drive.google.com/file/d/1bPt1Xqy0HwXpJFcug3vaC3MBqFTMvj5A/view) for the Fulani yet.
It would be nice if Davidski got a hold of some -- it looks like the samples from the recent Fulani paper (Vicente, 2019 (https://www.biorxiv.org/content/10.1101/650986v1.full)) are only available for academics"

@Generalissimo, would that SNP count be adequate for G25?


Genome-wide SNP typing
A subset of 55 Fulani and 7 Czech/Slovak individuals were selected for genome-wide genotyping on the Illumina Omni2.5-Octo BeadChip (which contains the T–13910 SNP). The data was aligned to the Human Genome built version 37.

Data management and quality filtering was carried out using PLINK v.1.90 software (Chang et al. 2015). A total of 2,608,742 SNPs were obtained from the 62 individuals. All individuals passed 0.15 data missingness threshold. We subsequently filtered to keep only autosomal SNPs with a SNP missingness filter of 0.1. To account for possible genotyping errors, we applied a Hardy-Weinberg equilibrium filter (HWE) that excluded 90 SNPs (for p ≤ 1e-4). AT and CG SNPs were excluded to prevent strand flipping errors when merging with comparative datasets. Relatedness was measured by identical by state (IBS) analysis and two Fulani individuals were excluded due to potential genetic relatedness. A total of 2,359,821 SNPs and 60 individuals were kept for the study.

The newly generated data will be made available for academic research use through the ArrayExpress database accession number XXXX.



I’d imagine their numbers would be very interesting given the Fischer component which peaked in them in the recent Tishkoff paper?

What "Fischer" component are you referring to -- do you mean the Fuschia colour at K14 from the 09, Tishkoff paper (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2947357/) ? I think that's just ADMIXTURE indicating close relatedness within the Fulani.

Angoliga
06-28-2019, 03:25 AM
"YRI -- maybe this will get corrected after we get more aDNA... after all we now know Yourba are ~12% Eurasian"

According to Lazaridis the Yoruba (or at least those sampled) are about 12% Iberomaurusian-like/Taforalt-like, and he models Taforalt (the Iberomaurusians from Northern Morocco) as about half (45-50%) ANA and about half (50-55%) Eurasian. The ANA (Ancient North African) population is an African one (albeit one native to North Africa and distinct from other populations) rather than an OOA/Eurasian population, and is more basal than BE/Basal Eurasian. So the Yoruba, according to Lazaridis would seem to be about 6-7% Eurasian (rather than 12%).



... not too far off : )

This recent paper (Bergström, 19 (https://www.biorxiv.org/content/biorxiv/early/2019/06/27/674986.full.pdf)) couldn't have come out at a better time :


The West African Yoruba also display a Neanderthal admixture signal, similar in shape but much less 380 pronounced than the signal in non-Africans (Fig. 6D). Other African populations do not clearly display the same behaviour. These results provide evidence for low amounts of Neanderthal ancestry in West Africa, consistent with previous results based on other approaches (15, 19), and we estimate this at 0.18%±0.06% in Yoruba using an f4-ratio (assuming Mbuti has none). The most likely source for this is West Eurasian admixture (37), 385 and assuming a simple linear relationship to Neanderthal ancestry, our estimate implies 8.6%±3% Eurasian ancestry in Yoruba.

*That's corroboration between 3 independent methods

Angoliga
06-28-2019, 04:13 AM
The Mende are also quite mysterious, they have a larger than expected CHG component, larger than a lot of more easterly groups despite the fact that there is no know history of pygmies that far west in their current teritory

Proportionately, the ancient non-SSA (ANA/Basal-Eurasian) in West-Africans might suggests they're directly derived from Ancient East-Africans so could it be this archaic SSA affinity which is the main factor pulling West-Africans closer to the San relative to Nilotes on PCAs? Perhaps this is the case since both the Dinka and Yoruba appear to have similar non-SSA contributions (MAR_EN\MAR_Iberomaurusian) -- it's the difference between their archaic SSA which is most profound.




... we find that the west African Yoruba have a closer relationship to non-Africans than to the central African Mbuti at high allele frequencies but the opposite relationship at low frequencies (Fig. 2 B ), suggesting recent gene flow between Mbuti and Yoruba since the divergence of non-Africans.





1] "distance%=31.1751"

Esan_Nigeria

MWI_Fingira_6100BP,67.1
MAR_Iberomaurusian,15.5
MWI_Fingira_2500BP,12.4
MAR_EN,4.1
RUS_Ust_Ishim,0.4
Anatolia_N_Kumtepe_low_res,0.4

[1] "distance%=29.943"

Igbo

MWI_Fingira_6100BP,69.5
MAR_Iberomaurusian,16
MWI_Fingira_2500BP,10.9
MAR_EN,3.3
Anatolia_N_Kumtepe_low_res,0.3

[1] "distance%=31.2193"

Yoruba

MWI_Fingira_6100BP,70.6
MAR_Iberomaurusian,15.8
MWI_Fingira_2500BP,8.5
MAR_EN,4.7
BLZ_SakiTzul_7400BP,0.2
CHL_LosRieles_5100BP,0.1
Anatolia_N_Kumtepe_low_res,0.1



[1] "distance%=28.1667"

Gambian

MWI_Fingira_6100BP,68.7
MAR_Iberomaurusian,19
MWI_Fingira_2500BP,7.3
MAR_EN,4.6
Anatolia_N_Kumtepe_low_res,0.3

[1] "distance%=28.2958"

Mandenka

MWI_Fingira_6100BP,69.6
MAR_Iberomaurusian,19.6
MWI_Fingira_2500BP,6.4
MAR_EN,4
Anatolia_N_Kumtepe_low_res,0.2
BLZ_SakiTzul_7400BP,0.1
PER_Laramate_900BP,0.1

[1] "distance%=25.7217"

Dinka

MWI_Fingira_6100BP,72.5
MAR_Iberomaurusian,20.7
MAR_EN,4.5
MWI_Fingira_2500BP,2
Eskimo_Sireniki,0.1
BRA_Laranjal_6700BP,0.1
PER_Lauricocha_3500BP,0.1

Chad Rohlfsen
06-28-2019, 04:34 AM
I would say that trying models with Mbuti may help here. I've seen similar stuff with my own runs.

There will always be new stuff coming that changes models. One thing to also consider is R1b and G2 among the Yoruba and Mende respectively. So, it's more than just Iberomaurusian ancestry.

Brwn_trd
06-28-2019, 03:35 PM
What "Fischer" component are you referring to -- do you mean the Fuschia colour at K14 from the 09, Tishkoff paper (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2947357/) ? I think that's just ADMIXTURE indicating close relatedness within the Fulani.

Pardon me, I was on my phone, autocorrect changed it. I did mean fuschia

Angoliga
07-03-2019, 02:17 AM
Regarding the seemingly out of place archaic-SSA ancestry in West-African pops, I found this paper (https://www.biorxiv.org/content/10.1101/285734v2) (Durvasula, 2019) quite relevant:


While introgression from Neanderthals and Denisovans has been well-documented in modern humans outside Africa, the contribution of archaic hominins to the genetic variation of present-day Africans remains poorly understood. Using 405 whole-genome sequences from four sub-Saharan African populations, we provide complementary lines of evidence for archaic introgression into these populations. Our analyses of site frequency spectra indicate that these populations derive 2-19% of their genetic ancestry from an archaic population that diverged prior to the split of Neanderthals and modern humans. Using a method that can identify segments of archaic ancestry without the need for reference archaic genomes, we built genome-wide maps of archaic ancestry in the Yoruba and the Mende populations that recover about 482 and 502 megabases of archaic sequence, respectively. Analyses of these maps reveal segments of archaic ancestry at high frequency in these populations that represent potential targets of adaptive introgression. Our results reveal the substantial contribution of archaic ancestry in shaping the gene pool of present-day African populations.


I ran the Mende out of curiosity since the paper had them cited. Could these results be reflective of ghost archaic introgression? :


[1] "distance%=28.0652"

Mende_Sierra_Leone

MWI_Fingira_6100BP,63.7
MWI_Fingira_2500BP,16.3
MAR_Iberomaurusian,15.2
MAR_EN,3.8
Anatolia_N_Kumtepe_low_res,0.4
RUS_Ust_Ishim,0.3
CHN_Tianyuan,0.1
Jarawa,0.1
VNM_BA_all,0.1

1] "distance%=31.1751"

Esan_Nigeria

MWI_Fingira_6100BP,67.1
MAR_Iberomaurusian,15.5
MWI_Fingira_2500BP,12.4
MAR_EN,4.1
RUS_Ust_Ishim,0.4
Anatolia_N_Kumtepe_low_res,0.4

[1] "distance%=29.943"

Igbo

MWI_Fingira_6100BP,69.5
MAR_Iberomaurusian,16
MWI_Fingira_2500BP,10.9
MAR_EN,3.3
Anatolia_N_Kumtepe_low_res,0.3

[1] "distance%=31.2193"

Yoruba

MWI_Fingira_6100BP,70.6
MAR_Iberomaurusian,15.8
MWI_Fingira_2500BP,8.5
MAR_EN,4.7
BLZ_SakiTzul_7400BP,0.2
CHL_LosRieles_5100BP,0.1
Anatolia_N_Kumtepe_low_res,0.1



[1] "distance%=28.1667"

Gambian

MWI_Fingira_6100BP,68.7
MAR_Iberomaurusian,19
MWI_Fingira_2500BP,7.3
MAR_EN,4.6
Anatolia_N_Kumtepe_low_res,0.3

[1] "distance%=28.2958"

Mandenka

MWI_Fingira_6100BP,69.6
MAR_Iberomaurusian,19.6
MWI_Fingira_2500BP,6.4
MAR_EN,4
Anatolia_N_Kumtepe_low_res,0.2
BLZ_SakiTzul_7400BP,0.1
PER_Laramate_900BP,0.1

[1] "distance%=25.7217"

Dinka

MWI_Fingira_6100BP,72.5
MAR_Iberomaurusian,20.7
MAR_EN,4.5
MWI_Fingira_2500BP,2
Eskimo_Sireniki,0.1
BRA_Laranjal_6700BP,0.1
PER_Lauricocha_3500BP,0.1


*Absent in Somalis

1] "distance%=15.1951"

Somali

MWI_Fingira_6100BP,40.1
Levant_Natufian,29
MAR_Iberomaurusian,24.1
IRN_Tepe_Abdul_Hosein_N,5.2
IRN_Wezmeh_N,1.2
Canary_Islands_Guanche,0.3
MAR_EN,0.1






I would say that trying models with Mbuti may help here. I've seen similar stuff with my own runs.

There will always be new stuff coming that changes models. One thing to also consider is R1b and G2 among the Yoruba and Mende respectively. So, it's more than just Iberomaurusian ancestry.

Unfortunately, we don't have any currently available Mbuti G25 cords for comparison. In the interim, MWI_Fingira_2500BP seems a decent proxy for archaic-SSA.

The minor R1b in Yoruba (~4.8%) is too little to discern anything non-negligible.
In a trans-saharan inverse, it'd be like trying to distill SSA components from modern Sardinians since they have minor A-M13 (~<4.5%).

The likely immediate mediators of R1b to the Yoruba may have some minor clues, hopefully we'll soon have G25 cords of Chadics pops -- aDNA would be ideal of course. I'm curious about Egypt\Sudan from the mid\late-neolthic to the pre-Islamic era.

VytautusofAukstaitija
07-03-2019, 02:19 AM
Regarding the seemingly out of place archaic-SSA ancestry in West-African pops, I found this paper (https://www.biorxiv.org/content/10.1101/285734v2) (Durvasula, 2019) quite relevant:




I ran the Mende out of curiosity since the paper had them cited. Could these results be reflective of ghost archaic introgression? :


[1] "distance%=28.0652"

Mende_Sierra_Leone

MWI_Fingira_6100BP,63.7
MWI_Fingira_2500BP,16.3
MAR_Iberomaurusian,15.2
MAR_EN,3.8
Anatolia_N_Kumtepe_low_res,0.4
RUS_Ust_Ishim,0.3
CHN_Tianyuan,0.1
Jarawa,0.1
VNM_BA_all,0.1

1] "distance%=31.1751"

Esan_Nigeria

MWI_Fingira_6100BP,67.1
MAR_Iberomaurusian,15.5
MWI_Fingira_2500BP,12.4
MAR_EN,4.1
RUS_Ust_Ishim,0.4
Anatolia_N_Kumtepe_low_res,0.4

[1] "distance%=29.943"

Igbo

MWI_Fingira_6100BP,69.5
MAR_Iberomaurusian,16
MWI_Fingira_2500BP,10.9
MAR_EN,3.3
Anatolia_N_Kumtepe_low_res,0.3

[1] "distance%=31.2193"

Yoruba

MWI_Fingira_6100BP,70.6
MAR_Iberomaurusian,15.8
MWI_Fingira_2500BP,8.5
MAR_EN,4.7
BLZ_SakiTzul_7400BP,0.2
CHL_LosRieles_5100BP,0.1
Anatolia_N_Kumtepe_low_res,0.1



[1] "distance%=28.1667"

Gambian

MWI_Fingira_6100BP,68.7
MAR_Iberomaurusian,19
MWI_Fingira_2500BP,7.3
MAR_EN,4.6
Anatolia_N_Kumtepe_low_res,0.3

[1] "distance%=28.2958"

Mandenka

MWI_Fingira_6100BP,69.6
MAR_Iberomaurusian,19.6
MWI_Fingira_2500BP,6.4
MAR_EN,4
Anatolia_N_Kumtepe_low_res,0.2
BLZ_SakiTzul_7400BP,0.1
PER_Laramate_900BP,0.1

[1] "distance%=25.7217"

Dinka

MWI_Fingira_6100BP,72.5
MAR_Iberomaurusian,20.7
MAR_EN,4.5
MWI_Fingira_2500BP,2
Eskimo_Sireniki,0.1
BRA_Laranjal_6700BP,0.1
PER_Lauricocha_3500BP,0.1


*Absent in Somalis

1] "distance%=15.1951"

Somali

MWI_Fingira_6100BP,40.1
Levant_Natufian,29
MAR_Iberomaurusian,24.1
IRN_Tepe_Abdul_Hosein_N,5.2
IRN_Wezmeh_N,1.2
Canary_Islands_Guanche,0.3
MAR_EN,0.1







Unfortunately, we don't have any currently available Mbuti G25 cords for comparison. In the interim, MWI_Fingira_2500BP seems a decent proxy for archaic-SSA.

The minor R1b in Yoruba (~4.8%) is too little to discern anything non-negligible.
In a trans-saharan inverse, it'd be like trying to distill SSA components from modern Sardinians since they have minor A-M13 (~<4.5%).

The likely immediate mediators of R1b to the Yoruba may have some minor clues, hopefully we'll soon have G25 cords of Chadics pops -- aDNA would be ideal of course. I'm curious about Egypt\Sudan from the mid\late-neolthic to the pre-Islamic era.

Yoruba R1b-V88 likely didn't bring 10% Eurasian ancestry. This is deeper stuff than R1b-V88, which is a newcomer to West Africa compared to this much more significant, older Eurasian ancestry and ANA. If someone could check if the Bantu also have it, it will simply nail this in the head as I assume they also carry this Eurasian ancestry. The Yoruba live close to the Bantoid urheimat at the confluence of the Niger-Benue.

Angoliga
07-03-2019, 02:41 AM
Edit: I reran Hadza and KEN_PN, since I had forgotten to remove Saudis and Bedouins -- additional runs also added.


Here's more runs including Bantu, EHGs and ancient East-African Pastoralists... :



[1] "distance%=27.9197"

Kongo

MWI_Fingira_6100BP,61.2
MWI_Fingira_2500BP,20.4
MAR_Iberomaurusian,14.8
MAR_EN,2.6
RUS_Ust_Ishim,0.5
Anatolia_N_Kumtepe_low_res,0.3
LAO_Hoabinhian,0.1
VNM_BA_all,0.1


[1] "distance%=26.3097"

Bantu_Kenya

MWI_Fingira_6100BP,60.5
MAR_Iberomaurusian,17.9
MWI_Fingira_2500BP,17.3
MAR_EN,3.6
RUS_Ust_Ishim,0.7



[1] "distance%=21.4296"

Bantu_S.E.

MWI_Fingira_6100BP,57.4
MWI_Fingira_2500BP,30.5
MAR_Iberomaurusian,8.4
MAR_EN,2.8
RUS_Ust_Ishim,0.7
Anatolia_N_Kumtepe_low_res,0.1
PYF_150BP,0.1



[1] "distance%=24.9105"

Bantu_S.W.

MWI_Fingira_6100BP,61.6
MWI_Fingira_2500BP,21.5
MAR_Iberomaurusian,13.4
MAR_EN,2.9
Anatolia_N_Kumtepe_low_res,0.4
RUS_Ust_Ishim,0.2



[1] "distance%=19.3774"

Kikuyu

MWI_Fingira_6100BP,64.4
MAR_Iberomaurusian,23.8
MAR_EN,5.4
IRN_Wezmeh_N,1.4
IRN_Tepe_Abdul_Hosein_N,1.3
Anatolia_N_Kumtepe_low_res,1
IRN_HotuIIIb_Mesolithic,0.8
RUS_Devils_Gate_low_res,0.5
RUS_Ust_Ishim,0.4
MWI_Fingira_2500BP,0.4
IRN_Seh_Gabi_LN,0.2
IRN_Ganj_Dareh_N,0.2
AFG_Darra_i_kur_low_res,0.1
EGY_Late_Period,0.1



[1] "distance%=12.343"

Hadza

MWI_Fingira_6100BP,66.5
MAR_Iberomaurusian,17.1
MWI_Fingira_2500BP,6.9
RUS_Ust_Ishim,4.4
IRN_HotuIIIb_Mesolithic,1.7
Malaysia_LN,1.2
MAR_EN,0.8
IRN_Tepe_Abdul_Hosein_N,0.6
LAO_Hoabinhian,0.4
PER_Lauricocha_3500BP,0.1
BRA_Laranjal_6700BP,0.1
IRN_Wezmeh_N,0.1
Thailand_IA,0.1


[1] "distance%=14.2441"

Sandawe

MWI_Fingira_6100BP,66
MAR_Iberomaurusian,22.3
IRN_Wezmeh_N,3.7
Anatolia_N_Kumtepe_low_res,2.5
MAR_EN,2.2
RUS_Devils_Gate_low_res,0.9
IRN_Tepe_Abdul_Hosein_N,0.8
EGY_Late_Period,0.7
Anatolia_Kumtepe_N,0.6
IRN_HotuIIIb_Mesolithic,0.1
EGY_Hellenistic,0.1

[1] "distance%=21.2416"

Gumuz

MWI_Fingira_6100BP,67.9
MAR_Iberomaurusian,20.1
MAR_EN,4.9
LAO_Hoabinhian,2.5
MWI_Fingira_2500BP,2.3
RUS_Devils_Gate_low_res,1.1
RUS_Ust_Ishim,0.6
Jarawa,0.3
Malaysia_LN,0.2
Htin_Mal,0.1


[1] "distance%=19.194"

Masai

MWI_Fingira_6100BP,57.7
MAR_Iberomaurusian,29.9
EGY_Late_Period,4.4
IRN_Tepe_Abdul_Hosein_N,3.3
IRN_Wezmeh_N,1.7
EGY_Hellenistic,1.2
RUS_Devils_Gate_low_res,0.8
MAR_EN,0.4
IRN_HotuIIIb_Mesolithic,0.2
Anatolia_Kumtepe_N,0.1
AFG_Darra_i_kur_low_res,0.1
England_Roman_Near_Eastern_o,0.1
RUS_Ust_Ishim,0.1


[1] "distance%=15.0439"

KEN_Pastoral_N

MWI_Fingira_6100BP,40.4
MAR_Iberomaurusian,23.8
Levant_Natufian,19.5
EGY_Late_Period,13.8
IRN_Tepe_Abdul_Hosein_N,1
EGY_Hellenistic,0.9
AFG_Darra_i_kur_low_res,0.2
IRN_Wezmeh_N,0.2
LAO_Hoabinhian,0.2


[1] "distance%=12.6521"

TZA_Luxmanda_3100BP

MWI_Fingira_6100BP,43
Levant_Natufian,23.4
MAR_Iberomaurusian,19.7
EGY_Late_Period,9.5
IRN_Tepe_Abdul_Hosein_N,1.8
RUS_Devils_Gate_low_res,1.1
EGY_Hellenistic,0.8
LAO_Hoabinhian,0.3
PER_Lauricocha_3500BP,0.2
IRN_Wezmeh_N,0.2



[1] "distance%=14.5397"

KEN_Pastoral_N_Elmenteitan

MWI_Fingira_6100BP,39.1
Levant_Natufian,24.2
MAR_Iberomaurusian,20.9
EGY_Late_Period,10.1
EGY_Hellenistic,2.2
LAO_Hoabinhian,1.7
IRN_Tepe_Abdul_Hosein_N,1.1
RUS_Devils_Gate_low_res,0.4
IRN_Wezmeh_N,0.3



[1] "distance%=14.4027"

ETH_4500BP

MWI_Fingira_6100BP,66.9
MAR_Iberomaurusian,18.1
MWI_Fingira_2500BP,7.1
LAO_Hoabinhian,3.8
MAR_EN,1.3
RUS_Ust_Ishim,1.2
PER_Lauricocha_3500BP,0.7
BLZ_SakiTzul_7400BP,0.2
MEX_Huichol,0.1
CHL_Yamana_1000BP,0.1
USA_San_Catalina_Island,0.1
CHL_LosRieles_5100BP,0.1
BRA_Laranjal_6700BP,0.1
Malaysia_LN,0.1
IND_Great_Andamanese_100BP,0.1


[1] "distance%=13.4873"

San

MWI_Fingira_2500BP,68.9
MWI_Fingira_6100BP,21.6
MAR_Iberomaurusian,3.5
MAR_EN,3
RUS_Ust_Ishim,2.9
IRN_Seh_Gabi_LN,0.1



Just to reiterate, the purpose here is to provide an ancient context for the ANA/Eurasian affinities in SSA pops. We're not concerned with the best fit, these runs are to see which pops nMonte selects as a best proxies.


As a recap, here's the general conditions for these runs:


Modern pops with heavy SW-Asian and ANAs have been removed (i.e. modern Egyptians, Saudis, Bedouins, Yemeni..)


Admixed SSA ancients with substantial non-SSA affinities are also removed (e.g. KEN_PN, Luxmanda..). This is to gather a clearer ancestral idea of SSA vs. Eurasian/ANA affinities


All modern SSAs were removed from the pop spreadsheet. SSA proxies are MWI_Fingira_6100BP = modern SSA (main SSA component in Yoruba and Nilotes), and MWI_Fingira_2500BP = Archaic SSA (main component in the Biaka pygmy..)


The remaining 800+ G25 pops have been left to see what pops nMonte selects for non-SSA affinities; these runs take about ~4 minutes each on my comp

VytautusofAukstaitija
07-03-2019, 03:08 AM
Regarding the seemingly out of place archaic-SSA ancestry in West-African pops, I found this paper (https://www.biorxiv.org/content/10.1101/285734v2) (Durvasula, 2019) quite relevant:




I ran the Mende out of curiosity since the paper had them cited. Could these results be reflective of ghost archaic introgression? :


[1] "distance%=28.0652"

Mende_Sierra_Leone

MWI_Fingira_6100BP,63.7
MWI_Fingira_2500BP,16.3
MAR_Iberomaurusian,15.2
MAR_EN,3.8
Anatolia_N_Kumtepe_low_res,0.4
RUS_Ust_Ishim,0.3
CHN_Tianyuan,0.1
Jarawa,0.1
VNM_BA_all,0.1

1] "distance%=31.1751"

Esan_Nigeria

MWI_Fingira_6100BP,67.1
MAR_Iberomaurusian,15.5
MWI_Fingira_2500BP,12.4
MAR_EN,4.1
RUS_Ust_Ishim,0.4
Anatolia_N_Kumtepe_low_res,0.4

[1] "distance%=29.943"

Igbo

MWI_Fingira_6100BP,69.5
MAR_Iberomaurusian,16
MWI_Fingira_2500BP,10.9
MAR_EN,3.3
Anatolia_N_Kumtepe_low_res,0.3

[1] "distance%=31.2193"

Yoruba

MWI_Fingira_6100BP,70.6
MAR_Iberomaurusian,15.8
MWI_Fingira_2500BP,8.5
MAR_EN,4.7
BLZ_SakiTzul_7400BP,0.2
CHL_LosRieles_5100BP,0.1
Anatolia_N_Kumtepe_low_res,0.1



[1] "distance%=28.1667"

Gambian

MWI_Fingira_6100BP,68.7
MAR_Iberomaurusian,19
MWI_Fingira_2500BP,7.3
MAR_EN,4.6
Anatolia_N_Kumtepe_low_res,0.3

[1] "distance%=28.2958"

Mandenka

MWI_Fingira_6100BP,69.6
MAR_Iberomaurusian,19.6
MWI_Fingira_2500BP,6.4
MAR_EN,4
Anatolia_N_Kumtepe_low_res,0.2
BLZ_SakiTzul_7400BP,0.1
PER_Laramate_900BP,0.1

[1] "distance%=25.7217"

Dinka

MWI_Fingira_6100BP,72.5
MAR_Iberomaurusian,20.7
MAR_EN,4.5
MWI_Fingira_2500BP,2
Eskimo_Sireniki,0.1
BRA_Laranjal_6700BP,0.1
PER_Lauricocha_3500BP,0.1


*Absent in Somalis

1] "distance%=15.1951"

Somali

MWI_Fingira_6100BP,40.1
Levant_Natufian,29
MAR_Iberomaurusian,24.1
IRN_Tepe_Abdul_Hosein_N,5.2
IRN_Wezmeh_N,1.2
Canary_Islands_Guanche,0.3
MAR_EN,0.1







Unfortunately, we don't have any currently available Mbuti G25 cords for comparison. In the interim, MWI_Fingira_2500BP seems a decent proxy for archaic-SSA.

The minor R1b in Yoruba (~4.8%) is too little to discern anything non-negligible.
In a trans-saharan inverse, it'd be like trying to distill SSA components from modern Sardinians since they have minor A-M13 (~<4.5%).

The likely immediate mediators of R1b to the Yoruba may have some minor clues, hopefully we'll soon have G25 cords of Chadics pops -- aDNA would be ideal of course. I'm curious about Egypt\Sudan from the mid\late-neolthic to the pre-Islamic era.

Early Nilo-Saharan-like people rich in A3b2 were in the western and Libyan deserts, and nearer the Kharga Oasis and that some corridor from the Nile, perhaps the Laqiya corridor and Wadi Howwar and Wadi el-Malik, brought early Chadics and perhaps some early Cushitic group that carried the E-V5933 E-V32 lineage that some Chadics and Fur, Masalit, and various Nilo-Saharan subgroups carry. There are now Cameroonian E-V32 carriers as well.

@NiloSaharan, can you get the new samples of the ELM/PN and Iron Age Southeast African pastoralists? I really appreciate everything your doing with my endless inquiries.

VytautusofAukstaitija
07-03-2019, 03:08 AM
These take sometime to run, I can continuously add more Bantu pops while editing the post:




[1] "distance%=27.9197"

Kongo

MWI_Fingira_6100BP,61.2
MWI_Fingira_2500BP,20.4
MAR_Iberomaurusian,14.8
MAR_EN,2.6
RUS_Ust_Ishim,0.5
Anatolia_N_Kumtepe_low_res,0.3
LAO_Hoabinhian,0.1
VNM_BA_all,0.1


[1] "distance%=26.3097"

Bantu_Kenya

MWI_Fingira_6100BP,60.5
MAR_Iberomaurusian,17.9
MWI_Fingira_2500BP,17.3
MAR_EN,3.6
RUS_Ust_Ishim,0.7

[1] "distance%=12.3369"

Hadza

MWI_Fingira_6100BP,67.7
MAR_Iberomaurusian,14.6
MWI_Fingira_2500BP,5.7
RUS_Ust_Ishim,4.2
MAR_EN,3
IRN_HotuIIIb_Mesolithic,1.7
Malaysia_LN,1.3
LAO_Hoabinhian,0.6
BedouinA,0.3
IRN_Wezmeh_N,0.3
Yemenite_Dhamar,0.2
PER_Lauricocha_3500BP,0.2
IRN_Tepe_Abdul_Hosein_N,0.1
BRA_Laranjal_6700BP,0.1



Ongoing observations:

As you can see, Bantus like the Kongo have the similar ANA breakdown.
Their affinity for archaic-SSA could likely be combined with more recent Biaka-like affinities and the afmd archaic west-african ghost.

OTOH for the Bantu_Kenya, elevated MWI_Fingira_2500BP could be reflective of additional EHG\Hadza like ancestry
- this looks to not be the case (edit)

The Hadza's results seem to reflect them being closely similar to OoA pops. Maybe that explains nMonte's choice of non-negligible RUS_Ust_Ishim, IRN_HotuIIIb_Mesolithic, Malaysia_LN

Ah, thanks for this, much appreciated @NiloSaharan.

Jm8
07-04-2019, 01:48 AM
Regarding the seemingly out of place archaic-SSA ancestry in West-African pops, I found this paper (https://www.biorxiv.org/content/10.1101/285734v2) (Durvasula, 2019) quite relevant:




I ran the Mende out of curiosity since the paper had them cited. Could these results be reflective of ghost archaic introgression? :


[1] "distance%=28.0652"

Mende_Sierra_Leone

MWI_Fingira_6100BP,63.7
MWI_Fingira_2500BP,16.3
MAR_Iberomaurusian,15.2
MAR_EN,3.8
Anatolia_N_Kumtepe_low_res,0.4
RUS_Ust_Ishim,0.3
CHN_Tianyuan,0.1
Jarawa,0.1
VNM_BA_all,0.1

1] "distance%=31.1751"

Esan_Nigeria

MWI_Fingira_6100BP,67.1
MAR_Iberomaurusian,15.5
MWI_Fingira_2500BP,12.4
MAR_EN,4.1
RUS_Ust_Ishim,0.4
Anatolia_N_Kumtepe_low_res,0.4

[1] "distance%=29.943"

Igbo

MWI_Fingira_6100BP,69.5
MAR_Iberomaurusian,16
MWI_Fingira_2500BP,10.9
MAR_EN,3.3
Anatolia_N_Kumtepe_low_res,0.3

[1] "distance%=31.2193"

Yoruba

MWI_Fingira_6100BP,70.6
MAR_Iberomaurusian,15.8
MWI_Fingira_2500BP,8.5
MAR_EN,4.7
BLZ_SakiTzul_7400BP,0.2
CHL_LosRieles_5100BP,0.1
Anatolia_N_Kumtepe_low_res,0.1



[1] "distance%=28.1667"

Gambian

MWI_Fingira_6100BP,68.7
MAR_Iberomaurusian,19
MWI_Fingira_2500BP,7.3
MAR_EN,4.6
Anatolia_N_Kumtepe_low_res,0.3

[1] "distance%=28.2958"

Mandenka

MWI_Fingira_6100BP,69.6
MAR_Iberomaurusian,19.6
MWI_Fingira_2500BP,6.4
MAR_EN,4
Anatolia_N_Kumtepe_low_res,0.2
BLZ_SakiTzul_7400BP,0.1
PER_Laramate_900BP,0.1

[1] "distance%=25.7217"

Dinka

MWI_Fingira_6100BP,72.5
MAR_Iberomaurusian,20.7
MAR_EN,4.5
MWI_Fingira_2500BP,2
Eskimo_Sireniki,0.1
BRA_Laranjal_6700BP,0.1
PER_Lauricocha_3500BP,0.1


*Absent in Somalis

1] "distance%=15.1951"

Somali

MWI_Fingira_6100BP,40.1
Levant_Natufian,29
MAR_Iberomaurusian,24.1
IRN_Tepe_Abdul_Hosein_N,5.2
IRN_Wezmeh_N,1.2
Canary_Islands_Guanche,0.3
MAR_EN,0.1







Unfortunately, we don't have any currently available Mbuti G25 cords for comparison. In the interim, MWI_Fingira_2500BP seems a decent proxy for archaic-SSA.

The minor R1b in Yoruba (~4.8%) is too little to discern anything non-negligible.
In a trans-saharan inverse, it'd be like trying to distill SSA components from modern Sardinians since they have minor A-M13 (~<4.5%).

The likely immediate mediators of R1b to the Yoruba may have some minor clues, hopefully we'll soon have G25 cords of Chadics pops -- aDNA would be ideal of course. I'm curious about Egypt\Sudan from the mid\late-neolthic to the pre-Islamic era.


Interestingly this 2-19% archaic admixture (or at least some of it) is also shared with Eurasians/non-Africans, meaning that it likely entered the gene pool of modern humans prior to the Out-of-Africa migration of ca. 50,000-65,000 bc.


As is mentioned in the supplemental text (under "Pre- versus post- Out-of-Africa"):

"We inspected the CSFS in the CEU conditioned on the Denisovan allele, since there is little Denisovan related
ancestry in Europeans [Mallick et al., 2016]. If the introgression event is shared in YRI and CEU, we expect to see a U-shaped CSFS in both populations (Figure S39abc). However, if the event is not shared and is specific to African populations only, the signal is expected to only appear in the African CSFS
(Figure S39def). We observe that the CEU CSFS, like the CSFS in the African populations, is U-shaped with an increase in the counts of high frequency alleles. (Figures S4 and S6). Further, we also observe a U-shape in the CSFS computed in the Han Chinese/CHB population (Figure S6). These results suggest that a component of the archaic ancestry that we detect in African populations is shared with non-African populations. We view these results as provisional and additional detailed modeling needs to be done to resolve the history and timing of archaic introgression.” (from page 16 of the supplementary information)


And according to the image in figure 1 (on page 7 of the paper), the 2-19% of archaic “Unknown Hominin” admixture seems to have mixed into the common ancestors of Africans and Europeans/non-Africans (modern humans pre-OOA), implying that it (or some of it) is also present in non-Africans—which fits with the above mentioned finding from the supplements that the admixture, or at least some/a component of the admixture, is shared by both African and non-African populations.

The paper:
https://www.biorxiv.org/content/early/2019/02/28/285734.full.pdf

The "Supplementary Information" (as a download rather than as a webpage/link) can be found by google searching the (entire) quoted text above—I'm not sure whether it's available at the paper link above.

Angoliga
07-05-2019, 05:52 AM
Early Nilo-Saharan-like people rich in A3b2 were in the western and Libyan deserts, and nearer the Kharga Oasis and that some corridor from the Nile, perhaps the Laqiya corridor and Wadi Howwar and Wadi el-Malik, brought early Chadics and perhaps some early Cushitic group that carried the E-V5933 E-V32 lineage that some Chadics and Fur, Masalit, and various Nilo-Saharan subgroups carry. There are now Cameroonian E-V32 carriers as well.

Very likely -- a more central/westward location would allow for the ancient embedded West-African affinities in nilotes like the Dinka.
IMO, the ancestral pop shared by nilotes and horners (Ancient East-African) would've been concentrated more eastward, let's say between the Upper-Nile and the Red-Sea.

I just had a look at E-Y28701 (https://www.yfull.com/tree/E-Y28701/), it bears a lot of similarity with this (https://www.yfull.com/tree/A-FGC38326/) A-M13 branch. This lines is quite interesting actually, you could say their carriers were ancient Nilotes that became North-Africanized, then Mediterraneanized, only to get re-North-Africanized and eventually return back to W/Central Africa albeit further south of the Sahara.

The commonalities between these two paternal sub-lineages are the multiple formation/tmrca dating towards the end of the green-sahara (~5-5k-5k bp); prosperous pastoralism being a likely key factor.



, can you get the new samples of the ELM/PN and Iron Age Southeast African pastoralists? I really appreciate everything your doing with my endless inquiries.




Here's a slew of more samples I ran, those from earlier posts are also grouped here:


*Disclaimer*: this experimental method is more meaningful for SSA pops with minor non-SSA affinities. If you'd like, we can definitely revisit African pops with more substantial\recent non-SSA admixture though I'm sure several other threads have delved into the topic in more meaningful depth. This can be seen clearly with horners... since I kept the ancient Egyptian pops, Late and Hellenistic samples are clearly eating up horner-related ancestry's ancient Iranian affinities. Natufian is the standard for their SW-Asian ancestry but I'd like to leave the pop spreadsheet as-is for consistency at the moment since we're chiefly teasing non-SSA in predominantly SSA pops.






West-Africans:




1] "distance%=31.1751"

Esan_Nigeria

MWI_Fingira_6100BP,67.1
MAR_Iberomaurusian,15.5
MWI_Fingira_2500BP,12.4
MAR_EN,4.1
RUS_Ust_Ishim,0.4
Anatolia_N_Kumtepe_low_res,0.4



[1] "distance%=29.943"

Igbo

MWI_Fingira_6100BP,69.5
MAR_Iberomaurusian,16
MWI_Fingira_2500BP,10.9
MAR_EN,3.3
Anatolia_N_Kumtepe_low_res,0.3



[1] "distance%=28.0652"

Mende_Sierra_Leone

MWI_Fingira_6100BP,63.7
MWI_Fingira_2500BP,16.3
MAR_Iberomaurusian,15.2
MAR_EN,3.8
Anatolia_N_Kumtepe_low_res,0.4
RUS_Ust_Ishim,0.3
CHN_Tianyuan,0.1
Jarawa,0.1
VNM_BA_all,0.1



[1] "distance%=31.2193"

Yoruba

MWI_Fingira_6100BP,70.6
MAR_Iberomaurusian,15.8
MWI_Fingira_2500BP,8.5
MAR_EN,4.7
BLZ_SakiTzul_7400BP,0.2
CHL_LosRieles_5100BP,0.1
Anatolia_N_Kumtepe_low_res,0.1



[1] "distance%=28.1667"

Gambian

MWI_Fingira_6100BP,68.7
MAR_Iberomaurusian,19
MWI_Fingira_2500BP,7.3
MAR_EN,4.6
Anatolia_N_Kumtepe_low_res,0.3



[1] "distance%=28.2958"

Mandenka

MWI_Fingira_6100BP,69.6
MAR_Iberomaurusian,19.6
MWI_Fingira_2500BP,6.4
MAR_EN,4
Anatolia_N_Kumtepe_low_res,0.2
BLZ_SakiTzul_7400BP,0.1
PER_Laramate_900BP,0.1






Bantu:




[1] "distance%=27.9197"

Kongo

MWI_Fingira_6100BP,61.2
MWI_Fingira_2500BP,20.4
MAR_Iberomaurusian,14.8
MAR_EN,2.6
RUS_Ust_Ishim,0.5
Anatolia_N_Kumtepe_low_res,0.3
LAO_Hoabinhian,0.1
VNM_BA_all,0.1


[1] "distance%=26.3097"

Bantu_Kenya

MWI_Fingira_6100BP,60.5
MAR_Iberomaurusian,17.9
MWI_Fingira_2500BP,17.3
MAR_EN,3.6
RUS_Ust_Ishim,0.7



[1] "distance%=21.4296"

Bantu_S.E.

MWI_Fingira_6100BP,57.4
MWI_Fingira_2500BP,30.5
MAR_Iberomaurusian,8.4
MAR_EN,2.8
RUS_Ust_Ishim,0.7
Anatolia_N_Kumtepe_low_res,0.1
PYF_150BP,0.1



[1] "distance%=24.9105"

Bantu_S.W.

MWI_Fingira_6100BP,61.6
MWI_Fingira_2500BP,21.5
MAR_Iberomaurusian,13.4
MAR_EN,2.9
Anatolia_N_Kumtepe_low_res,0.4
RUS_Ust_Ishim,0.2



[1] "distance%=19.3774"

Kikuyu

MWI_Fingira_6100BP,64.4
MAR_Iberomaurusian,23.8
MAR_EN,5.4
IRN_Wezmeh_N,1.4
IRN_Tepe_Abdul_Hosein_N,1.3
Anatolia_N_Kumtepe_low_res,1
IRN_HotuIIIb_Mesolithic,0.8
RUS_Devils_Gate_low_res,0.5
RUS_Ust_Ishim,0.4
MWI_Fingira_2500BP,0.4
IRN_Seh_Gabi_LN,0.2
IRN_Ganj_Dareh_N,0.2
AFG_Darra_i_kur_low_res,0.1
EGY_Late_Period,0.1





EHGs:




[1] "distance%=14.4691"

Ethiopian_Ari_cultivator

MWI_Fingira_6100BP,58
MAR_Iberomaurusian,23.7
EGY_Late_Period,6.6
Levant_Natufian,5.1
LAO_Hoabinhian,3.2
IRN_Tepe_Abdul_Hosein_N,1.4
EGY_Hellenistic,1
IRN_Wezmeh_N,0.4
RUS_Devils_Gate_low_res,0.2
AFG_Darra_i_kur_low_res,0.1
Malaysia_LN,0.1
RUS_Ust_Ishim,0.1
MAR_EN,0.1


[1] "distance%=14.6177"

Ethiopian_Ari_blacksmith

MWI_Fingira_6100BP,59.4
MAR_Iberomaurusian,24.6
EGY_Late_Period,5.9
Levant_Natufian,3.4
LAO_Hoabinhian,2.3
IRN_Wezmeh_N,1.2
IRN_Tepe_Abdul_Hosein_N,1.1
EGY_Hellenistic,1
RUS_Devils_Gate_low_res,0.9
LAO_BA,0.1
RUS_Ust_Ishim,0.1


[1] "distance%=12.343"

Hadza

MWI_Fingira_6100BP,66.5
MAR_Iberomaurusian,17.1
MWI_Fingira_2500BP,6.9
RUS_Ust_Ishim,4.4
IRN_HotuIIIb_Mesolithic,1.7
Malaysia_LN,1.2
MAR_EN,0.8
IRN_Tepe_Abdul_Hosein_N,0.6
LAO_Hoabinhian,0.4
PER_Lauricocha_3500BP,0.1
BRA_Laranjal_6700BP,0.1
IRN_Wezmeh_N,0.1
Thailand_IA,0.1


[1] "distance%=14.4027"

ETH_4500BP

MWI_Fingira_6100BP,66.9
MAR_Iberomaurusian,18.1
MWI_Fingira_2500BP,7.1
LAO_Hoabinhian,3.8
MAR_EN,1.3
RUS_Ust_Ishim,1.2
PER_Lauricocha_3500BP,0.7
BLZ_SakiTzul_7400BP,0.2
MEX_Huichol,0.1
CHL_Yamana_1000BP,0.1
USA_San_Catalina_Island,0.1
CHL_LosRieles_5100BP,0.1
BRA_Laranjal_6700BP,0.1
Malaysia_LN,0.1
IND_Great_Andamanese_100BP,0.1


[1] "distance%=14.2441"

Sandawe

MWI_Fingira_6100BP,66
MAR_Iberomaurusian,22.3
IRN_Wezmeh_N,3.7
Anatolia_N_Kumtepe_low_res,2.5
MAR_EN,2.2
RUS_Devils_Gate_low_res,0.9
IRN_Tepe_Abdul_Hosein_N,0.8
EGY_Late_Period,0.7
Anatolia_Kumtepe_N,0.6
IRN_HotuIIIb_Mesolithic,0.1
EGY_Hellenistic,0.1





NiloSaharans:




[1] "distance%=21.2416"

Gumuz

MWI_Fingira_6100BP,67.9
MAR_Iberomaurusian,20.1
MAR_EN,4.9
LAO_Hoabinhian,2.5
MWI_Fingira_2500BP,2.3
RUS_Devils_Gate_low_res,1.1
RUS_Ust_Ishim,0.6
Jarawa,0.3
Malaysia_LN,0.2
Htin_Mal,0.1


[1] "distance%=19.194"

Masai

MWI_Fingira_6100BP,57.7
MAR_Iberomaurusian,29.9
EGY_Late_Period,4.4
IRN_Tepe_Abdul_Hosein_N,3.3
IRN_Wezmeh_N,1.7
EGY_Hellenistic,1.2
RUS_Devils_Gate_low_res,0.8
MAR_EN,0.4
IRN_HotuIIIb_Mesolithic,0.2
Anatolia_Kumtepe_N,0.1
AFG_Darra_i_kur_low_res,0.1
England_Roman_Near_Eastern_o,0.1
RUS_Ust_Ishim,0.1


[1] "distance%=25.5201"

Luo

MWI_Fingira_6100BP,63.2
MAR_Iberomaurusian,17.6
MWI_Fingira_2500BP,15.4
MAR_EN,3
RUS_Ust_Ishim,0.6
VNM_BA_all,0.1
RUS_Devils_Gate_low_res,0.1
ROU_Iron_Gates_N,0


[1] "distance%=24.71"

Ethiopian_Anuak

MWI_Fingira_6100BP,67
MAR_Iberomaurusian,19.7
MWI_Fingira_2500BP,7.6
MAR_EN,4.5
RUS_Devils_Gate_low_res,0.8
LAO_Hoabinhian,0.2
Htin_Mal,0.1
RUS_Ust_Ishim,0.1


[1] "distance%=25.7217"

Dinka

MWI_Fingira_6100BP,72.5
MAR_Iberomaurusian,20.7
MAR_EN,4.5
MWI_Fingira_2500BP,2
Eskimo_Sireniki,0.1
BRA_Laranjal_6700BP,0.1
PER_Lauricocha_3500BP,0.1


[1] "distance%=23.0629"

UG_CentralSudanic

MWI_Fingira_6100BP,72.3
MAR_Iberomaurusian,16.6
MWI_Fingira_2500BP,7
MAR_EN,3.1
PER_Lauricocha_3500BP,0.4
CHL_LosRieles_5100BP,0.2
USA_Spirit_Cave_11000BP,0.2
BRA_Laranjal_6700BP,0.1
PER_Laramate_900BP,0.1





Ancient Pastoralists:




[1] "distance%=15.0439"

KEN_Pastoral_N

MWI_Fingira_6100BP,40.4
MAR_Iberomaurusian,23.8
Levant_Natufian,19.5
EGY_Late_Period,13.8
IRN_Tepe_Abdul_Hosein_N,1
EGY_Hellenistic,0.9
AFG_Darra_i_kur_low_res,0.2
IRN_Wezmeh_N,0.2
LAO_Hoabinhian,0.2


[1] "distance%=12.6521"

TZA_Luxmanda_3100BP

MWI_Fingira_6100BP,43
Levant_Natufian,23.4
MAR_Iberomaurusian,19.7
EGY_Late_Period,9.5
IRN_Tepe_Abdul_Hosein_N,1.8
RUS_Devils_Gate_low_res,1.1
EGY_Hellenistic,0.8
LAO_Hoabinhian,0.3
PER_Lauricocha_3500BP,0.2
IRN_Wezmeh_N,0.2


[1] "distance%=14.5397"

KEN_Pastoral_N_Elmenteitan

MWI_Fingira_6100BP,39.1
Levant_Natufian,24.2
MAR_Iberomaurusian,20.9
EGY_Late_Period,10.1
EGY_Hellenistic,2.2
LAO_Hoabinhian,1.7
IRN_Tepe_Abdul_Hosein_N,1.1
RUS_Devils_Gate_low_res,0.4
IRN_Wezmeh_N,0.3





Horners:




[1] "distance%=13.063"

Ethiopian_Wolayta

MWI_Fingira_6100BP,43
MAR_Iberomaurusian,21.7
Levant_Natufian,17.5
EGY_Hellenistic,9.9
EGY_Late_Period,3.3
IRN_Tepe_Abdul_Hosein_N,2.5
LAO_Hoabinhian,0.9
RUS_Devils_Gate_low_res,0.7
IRN_Wezmeh_N,0.5


[1] "distance%=11.3349"

Ethiopian_Tigray *excluding ancient egyptians and near eastern romans

Levant_Natufian,45.4
MWI_Fingira_6100BP,30.1
MAR_Iberomaurusian,14.4
IRN_Wezmeh_N,5.4
AFG_Darra_i_kur_low_res,3.4
IRN_Tepe_Abdul_Hosein_N,1
MAR_EN,0.3


[1] "distance%=11.239"

Ethiopian_Tigray

MWI_Fingira_6100BP,29.9
Levant_Natufian,26.8
MAR_Iberomaurusian,16.8
EGY_Late_Period,16.7
EGY_Hellenistic,5.6
IRN_Tepe_Abdul_Hosein_N,2.4
IRN_Wezmeh_N,1.6
AFG_Darra_i_kur_low_res,0.2


[1] "distance%=15.1904"

Somali *excluding ancients egyptian and near eastern romans

MWI_Fingira_6100BP,40.4
Levant_Natufian,29.1
MAR_Iberomaurusian,23.6
AFG_Darra_i_kur_low_res,2.7
IRN_Tepe_Abdul_Hosein_N,1.9
IRN_Wezmeh_N,1.8
RUS_Devils_Gate_low_res,0.2
MAR_EN,0.2
LAO_Hoabinhian,0.1


[1] "distance%=15.1448"

Somali

MWI_Fingira_6100BP,40
MAR_Iberomaurusian,25.4
Levant_Natufian,14
EGY_Hellenistic,10.2
EGY_Late_Period,7.7
IRN_Tepe_Abdul_Hosein_N,1.6
LAO_Hoabinhian,0.8
RUS_Devils_Gate_low_res,0.2
MAR_EN,0.1










Berbers:




[1] "distance%=3.3876"

Saharawi

MAR_Iberomaurusian,37.2
Anatolia_Kumtepe_N,18.5
HUN_Tisza_LN,9.5
MWI_Fingira_6100BP,9
EGY_Late_Period,8.6
EGY_Hellenistic,5.5
HRV_Vucedol,4.3
MAR_EN,4.3
Anatolia_N_Kumtepe_low_res,2.9
CZE_N,0.2


[1] "distance%=3.612"

Berber_MAR_ERR

MAR_Iberomaurusian,35.4
Anatolia_Kumtepe_N,22.3
MWI_Fingira_6100BP,9.9
HRV_Vucedol,8.7
EGY_Hellenistic,5.2
Anatolia_N_Kumtepe_low_res,4.7
MAR_EN,4.4
HUN_Tisza_LN,4.2
TZA_Zanzibar_First_Millenium,3.8
Wales_CA_EBA,1.4


[1] "distance%=2.937"

Mozabite

MAR_Iberomaurusian,33.6
Anatolia_Kumtepe_N,16
HRV_Vucedol,11.7
MWI_Fingira_6100BP,7.7
Levant_ISR_MLBA,6.3
TZA_Zanzibar_First_Millenium,6
Anatolia_N_Kumtepe_low_res,5.1
HUN_Tisza_LN,4.5
MAR_EN,3.7
DEU_MA_o,3.3
Anatolia_Kaman-Kalehoyuk_MLBA_low_res,1.9
Wales_CA_EBA,0.2





Others:




[1] "distance%=13.4873"

San

MWI_Fingira_2500BP,68.9
MWI_Fingira_6100BP,21.6
MAR_Iberomaurusian,3.5
MAR_EN,3
RUS_Ust_Ishim,2.9
IRN_Seh_Gabi_LN,0.1


[1] "distance%=1.3255"

MWI_Hora_9000BP

MWI_Fingira_2500BP,66.5
MWI_Fingira_6100BP,26.7
CZE_N,0.9
Scotland_N_o,0.9
IRN_Wezmeh_N,0.6
Koinanbe,0.4
Kosipe,0.4
IRN_Ganj_Dareh_N,0.4
IRN_Tepe_Abdul_Hosein_N,0.3
Iberia_Southwest_EN,0.3
IRN_Seh_Gabi_LN,0.2
ITA_Remedello_BA,0.2
Corded_Ware_CZE_o,0.2
UKR_Globular_Amphora,0.2
Bell_Beaker_HUN_EBA_o,0.2
WHG,0.2
Iberia_Northeast_CA,0.2
Levant_Natufian,0.1
Kurumba,0.1
IRN_Seh_Gabi_C,0.1
Levant_LBN_MA_NE,0.1
IRN_Hajji_Firuz_IA,0.1
Anatolia_Kumtepe_N,0.1
IRL_MN,0.1
Iberia_Southwest_MLN,0.1
Iberia_Southeast_MLN,0.1
DEU_Baalberge_MN,0.1
HUN_Baden_LCA,0.1
HRV_Vucedol,0.1
MAR_Iberomaurusian,0


[1] "distance%=23.1196"

Biaka

MWI_Fingira_2500BP,74.7
MWI_Fingira_6100BP,20.6
Anatolia_N_Kumtepe_low_res,2
MAR_Iberomaurusian,1.9
MAR_EN,0.7
IRN_Wezmeh_N,0.1


[1] "distance%=11.8492"

KEN_LSA

MWI_Fingira_6100BP,63.9
MAR_Iberomaurusian,14.6
MWI_Fingira_2500BP,14.5
LAO_Hoabinhian,4.3
Malaysia_LN,1
RUS_Ust_Ishim,0.5
MAR_EN,0.4
IND_Great_Andamanese_100BP,0.3
Jarawa,0.2
Onge,0.2
IRN_HotuIIIb_Mesolithic,0.1


[1] "distance%=9.0645"

MAR_Iberomaurusian

MAR_EN,69
Levant_Natufian,14.7
MWI_Fingira_6100BP,6.9
MAR_LN,5.9
MWI_Fingira_2500BP,1.8
Luzon,1
NPL_Mebrak_2125BP,0.3
PYF_150BP,0.2
JPN_Jomon,0.1
Onge,0.1


[1] "distance%=8.076"

MAR_EN

MAR_Iberomaurusian,69.3
MAR_LN,16.2
Levant_Natufian,8.8
MWI_Fingira_6100BP,4.5
MWI_Fingira_2500BP,0.5
Papuan,0.2
BLZ_MayahakCabPek_9300BP,0.1
BHS_Taino,0.1
Australian,0.1
Koinanbe,0.1
VUT_150BP_all,0.1



[1] "distance%=3.3716"

MAR_LN

GRC_N,31.3
DEU_Esperstedt_MN,25.4
MAR_EN,15.9
Levant_PPNC,12.9
MAR_Iberomaurusian,10.6
Levant_Natufian,1.5
MWI_Fingira_6100BP,0.9
Nganassan,0.5
Iberia_Southeast_CA,0.3
Yakut,0.3
IRN_Seh_Gabi_LN,0.2
Iberia_Northeast_CA,0.1
MKD_N,0.1


[1] "distance%=2.0847"

EGY_Late_Period

Levant_ISR_MLBA,26.2
Levant_PPNC,22
Levant_ISR_C,16.6
Levant_Natufian,15.9
TKM_Parkhai_MBA,7.1
MAR_Iberomaurusian,5.5
Anatolia_Kaman-Kalehoyuk_MLBA_low_res,3.2
GEO_CHG,2.4
Syrian,0.5
MWI_Fingira_6100BP,0.4
Nganassan,0.2






Some observations:



The archaic ancestry in West-Africans seems to be concentrated between the Windward coast and the Bight of Biafra


Elevated archaic ancestry among the Bantu appears to be a combination of the West-African archaic introgression and recent Pygmy/EHG. The levels of archaic ancestry among Nilo-saharan speakers demonstrates them intrusive to EHGs (Ik, Benet, Ogiek etc.); these pops would've largely inhabited the region between the Congo basin and the Gregory Rift. The Luo migrated smack-through this region, so them having the highest archaic makes perfect sense.


The Kikuyu interestingly have negligible archaic ancestry, as Northeast Bantu speakers maybe this confirms their northerly migration route into modern-day Kenya:


"The exact place that the Northeast Bantu speakers migrated from after the initial Bantu expansion is uncertain. Some authorities suggest that the Kikuyu arrived in their present Mount Kenya area of habitation from earlier settlements further to the north and east,[3] while others argue that the Kikuyu, along with their closely related Eastern Bantu neighbors the Embu, Meru, Mbeere, and Kamba moved into Kenya from points further north." (some references here (https://en.wikipedia.org/wiki/Kikuyu_people#cite_note-Bindloss-3))

EHGs are getting a lot of non-negligible ancient noise (RUS_Ust_Ishim, Malaysia_LN,LAO_Hoabinhian), perhaps due to them being most akin to the small-subset of OoAs. ETH_Gumuz has the same minor noise, more than other Nilosaharan-speakers. This could be indicative of the Gumuz' known to having higher Omotic ancestry. Similarly, KEN_LSA and MWI_Hora_9000BP are also getting a lot of noise. Since Ust’-Ishim is more closely related to modern East Asians, this would explain Malaysia/Lao popping up. Among EHGs, the Sandawe are the exception to this Eurasian-noise, likely due to their heavy Cushitic ancestry.


MAR_EN seems validated by being negligible in the Biaka Pygmies, Horners, KEN_LSA and ancient South-Africans (MWI_Hora_9000BP) -- pops whose general location wouldn't have been situated in N\W-Central Africa during the green-sahara period.

Angoliga
07-05-2019, 05:57 AM
Interestingly this 2-19% archaic admixture (or at least some of it) is also shared with Eurasians/non-Africans, meaning that it likely entered the gene pool of modern humans prior to the Out-of-Africa migration of ca. 50,000-65,000 bc.


As is mentioned in the supplemental text (under "Pre- versus post- Out-of-Africa"):

"We inspected the CSFS in the CEU conditioned on the Denisovan allele, since there is little Denisovan related
ancestry in Europeans [Mallick et al., 2016]. If the introgression event is shared in YRI and CEU, we expect to see a U-shaped CSFS in both populations (Figure S39abc). However, if the event is not shared and is specific to African populations only, the signal is expected to only appear in the African CSFS
(Figure S39def). We observe that the CEU CSFS, like the CSFS in the African populations, is U-shaped with an increase in the counts of high frequency alleles. (Figures S4 and S6). Further, we also observe a U-shape in the CSFS computed in the Han Chinese/CHB population (Figure S6). These results suggest that a component of the archaic ancestry that we detect in African populations is shared with non-African populations. We view these results as provisional and additional detailed modeling needs to be done to resolve the history and timing of archaic introgression.” (from page 16 of the supplementary information)


And according to the image in figure 1 (on page 7 of the paper), the 2-19% of archaic “Unknown Hominin” admixture seems to have mixed into the common ancestors of Africans and Europeans/non-Africans (modern humans pre-OOA), implying that it (or some of it) is also present in non-Africans—which fits with the above mentioned finding from the supplements that the admixture, or at least some/a component of the admixture, is shared by both African and non-African populations.

The paper:
https://www.biorxiv.org/content/early/2019/02/28/285734.full.pdf

The "Supplementary Information" (as a download rather than as a webpage/link) can be found by google searching the (entire) quoted text above—I'm not sure whether it's available at the paper link above.

The estimates of introgression might be revealing (~50ky bp):

https://i.imgur.com/ANhEYgK.png

It fits well with the beginning of the Mousterian Pluvial (~50ky bp); the last green-sahara period prior to the recent Neolithic Sub-pluvial.

If the afmd models imply West-African is derived from Ancient-East-African, could it have been the "greening" of the Sahara or other environmental ramifications during the Mousterian Pluvial that allowed for a westerly corridor from northern East-Africa to these West-African archaics?

Angoliga
07-05-2019, 08:29 AM
The estimates of introgression might be revealing (~50ky bp):

https://i.imgur.com/ANhEYgK.png

It fits well with the beginning of the Mousterian Pluvial (~50ky bp); the last green-sahara period prior to the recent Neolithic Sub-pluvial.

If the afmd models imply West-African is derived from Ancient-East-African, could it have been the "greening" of the Sahara or other environmental ramifications during the Mousterian Pluvial that allowed for a westerly corridor from northern East-Africa to these West-African archaics?


TMRCAs for both E-M5479 (https://www.yfull.com/tree/E-M5479/) and E-M75 (https://www.yfull.com/tree/E-M75/) have Ancient-East-African-rich pops (Amhara and Dinka respectively) and West-Africans (Gambians, YRI) stretching back to within the Mousterian Pluvial which spanned between ~50-30 ky bp.

The recurrence of the same formation dates (53,200 ybp) for subclades E-P147PF, E-M5479 and E-P147 might be signaling star-like expansions initiated by the Mousterian Pluvial.

It would be similar to what we're seeing 20k years later during the neolithic-subpluvial for Ancient-East-African-rich pops under A-M13 (https://www.yfull.com/tree/A-M13/) (A-Y23655, A-YP4751, A-Y23655 ).

Jm8
07-05-2019, 12:40 PM
The estimates of introgression might be revealing (~50ky bp):

https://i.imgur.com/ANhEYgK.png

It fits well with the beginning of the Mousterian Pluvial (~50ky bp); the last green-sahara period prior to the recent Neolithic Sub-pluvial.

If the afmd models imply West-African is derived from Ancient-East-African, could it have been the "greening" of the Sahara or other environmental ramifications during the Mousterian Pluvial that allowed for a westerly corridor from northern East-Africa to these West-African archaics?

Interesting. 50ky bp (according to recent estimates, like Haber's recent 2019 study on Ydna-D0) is also about (roughly) the time recently estimated for the Out-of-Africa migration (of the ancestors of modern non-African populations)—estimated at about 49-59ky bp. (much of) The archaic introgression seems likely to have happened soon before the OOA (possibly with a little happening in some regions of Africa soon after)—with the the admixture (or some of it) being shared by non-Africans (and some portion perhaps not being shared by them).

Unfortunately the study does not seem to estimate the likely percentages for each group (2-19% is a wide range).

(It's also too bad that the interesting finding/fact that some of this quite divergent archaic admixture is likely also shared by non-Africans is buried in the hard-to-find supplements and only subtly indicated in the paper, in figure 1, without explanation.)

Jm8
07-05-2019, 05:24 PM
Deleted

Jm8
07-05-2019, 05:25 PM
Deleted

Jm8
07-05-2019, 05:27 PM
Deleted

Angoliga
07-05-2019, 05:44 PM
Interesting. 50ky bp (according to recent estimates, like Haber's recent 2019 study on Ydna-D0) is also about (roughly) the time recently estimated for the Out-of-Africa migration (of the ancestors of modern non-African populations)—estimated at about 49-59ky bp. (much of) The archaic introgression seems likely to have happened soon before the OOA (possibly with a little happening in some regions of Africa soon after)—with the the admixture (or some of it) being shared by non-Africans (and some portion perhaps not being shared by them).

Unfortunately the study does not seem to estimate the likely percentages for each group (2-19% is a wide range).

(It's also too bad that the interesting finding/fact that some of the archaic admixture is likely also shared by non-Africans is buried in the hard-to-find supplements and only subtly indicated in the paper, in figure 1, without explanation.)

The admixture percentages are in Fig. 2 b) of the above post "Admixture Fraction"... decimals on the Y-axis show the percentage of archaic in each pop ESN, GWD, MSL, YRI (Esan_Nigeria, Gambian_Western_District, Mende_Sierre_Leone,Yoruba_In_Ibadan).


From the nmonte run, Iberomaurusian was rendered with minor archaic affinity -- it's absent completely among horners:


MAR_Iberomaurusian

MAR_EN,69
Levant_Natufian,14.7
MWI_Fingira_6100BP,6.9
MAR_LN,5.9
MWI_Fingira_2500BP,1.8
Luzon,1
NPL_Mebrak_2125BP,0.3
PYF_150BP,0.2
JPN_Jomon,0.1
Onge,0.1


Ethiopian_Wolayta

MWI_Fingira_6100BP,43
MAR_Iberomaurusian,21.7
Levant_Natufian,17.5
EGY_Hellenistic,9.9
EGY_Late_Period,3.3
IRN_Tepe_Abdul_Hosein_N,2.5
LAO_Hoabinhian,0.9
RUS_Devils_Gate_low_res,0.7
IRN_Wezmeh_N,0.5



Ethiopian_Tigray *excluding ancient egyptians and near eastern romans

Levant_Natufian,45.4
MWI_Fingira_6100BP,30.1
MAR_Iberomaurusian,14.4
IRN_Wezmeh_N,5.4
AFG_Darra_i_kur_low_res,3.4
IRN_Tepe_Abdul_Hosein_N,1
MAR_EN,0.3



Ethiopian_Tigray

MWI_Fingira_6100BP,29.9
Levant_Natufian,26.8
MAR_Iberomaurusian,16.8
EGY_Late_Period,16.7
EGY_Hellenistic,5.6
IRN_Tepe_Abdul_Hosein_N,2.4
IRN_Wezmeh_N,1.6
AFG_Darra_i_kur_low_res,0.2



Somali *excluding ancients egyptian and near eastern romans

MWI_Fingira_6100BP,40.4
Levant_Natufian,29.1
MAR_Iberomaurusian,23.6
AFG_Darra_i_kur_low_res,2.7
IRN_Tepe_Abdul_Hosein_N,1.9
IRN_Wezmeh_N,1.8
RUS_Devils_Gate_low_res,0.2
MAR_EN,0.2
LAO_Hoabinhian,0.1



Somali

MWI_Fingira_6100BP,40
MAR_Iberomaurusian,25.4
Levant_Natufian,14
EGY_Hellenistic,10.2
EGY_Late_Period,7.7
IRN_Tepe_Abdul_Hosein_N,1.6
LAO_Hoabinhian,0.8
RUS_Devils_Gate_low_res,0.2
MAR_EN,0.1




Maybe some minor archaic-SSA admixed with Ancient North-Africans following the Mousterian Pluvial (~<50ky bp); it would've tagged along with the modern-SSA.
From North-Africa the minor affinity could've made it's way across the Mediterranean into Eurasia. That wouldn't be far fetched since the introgression dating for West-Africans and the archaic is ~50ky bp. It'd be a replay of the trans-saharan SSA during the more recent green-sahara period.

Jm8
07-05-2019, 05:49 PM
I see the percentages now, thank you.

sum1
07-07-2019, 02:31 AM
https://genomebiology.biomedcentral.com/articles/10.1186/s13059-019-1684-5#MOESM1

This paper might help in regards to archaic introgression in African Populations.