Salkhit Pleistocene sample with High denisovan and Eurasian
|
Salkhit Pleistocene sample with High denisovan and Eurasian
|
03-27-2024, 01:06 PM
(03-27-2024, 02:12 AM)tipirneni Wrote: I see you are treating the single sample ancient culture Siberian Denisovan as a human admix. Unfortunately I don't get paid for educating such analysis. When I get paid to speak so I will definitely reach out to you seeking monetary compensation You don't seems to understand F-stat analysis. You can't educate others for something that you don't understand. 
03-27-2024, 09:48 PM
(03-25-2024, 08:42 AM)TanTin Wrote: I did some F4 tests and I can't confirm such High Denisovan components in Salkhit. Most samples will have way more Neanderthal than Denisovan so that stat will mostly just show Neanderthal compared to Ust'-Ishim. Maybe you can think up an f4-ratio or something to calculate Denisovan. But I also get positive with Chimp in place of Denisova, contamination or lot of damage? 
03-29-2024, 01:12 AM
The Salkhit genome is a little screwy. It's not really worth trying to analyze. I've already looked at this with Harvard. It acts like it has some kind of ancient and unknown archaic that branches before Denisovan. It looks like there's some serious type of issue here with the data.
03-29-2024, 02:57 AM
(03-29-2024, 01:12 AM)Chad Wrote: The Salkhit genome is a little screwy. It's not really worth trying to analyze. I've already looked at this with Harvard. It acts like it has some kind of ancient and unknown archaic that branches before Denisovan. It looks like there's some serious type of issue here with the data. The unknown archaic is exactly what I am searching. Salkhit is one of the individuals where we may confirm more or less for such unknown archaic components. However the branching is much earlier in my estimations. 
03-29-2024, 10:37 AM
I did variant calls on the Salkhit genome a while ago.
https://www.mediafire.com/file/zo1rj1she...9.zip/file
03-29-2024, 02:08 PM
No, it would be parallel to that found in West Africans. It's almost certainly just noise. Issues with processing.
03-30-2024, 05:22 AM
(03-29-2024, 10:37 AM)Norfern-Ostrobothnian Wrote: I did variant calls on the Salkhit genome a while ago. Thanks for sharing your data. I did a test projection on PCA - there is some difference with the officially provided genome (in V52 HO dataset), however the difference is not so critical. I think Salkhit is a very good example of early modern HS. The missingness for this sample is near 86 %. That means we have only 14% of its data. We can't expect miracle from such data. 
03-30-2024, 03:39 PM
Target: Mongolia_Salkhit_UP.SG
 alkhit1_noUDG.SG__BC_32923__Cov_12.87%Distance: 3.7828% / 0.03782780 | ADC: 0.25x RC 21.8 Russia_Yana_UP.SG 17.8 Bulgaria_BachoKiro_LatePleistocene 16.0 China_Guangxi_Longlin_Epipaleolithic 13.4 Japan_Honshu_EarlyJomon.SG 12.4 Hungary_Hun 5.8 Turkey_Kumtepe_N_lc.SG 5.6 Vanuatu_200BP_oPolynesian 4.0 Hungary_Conqueror_Elite 3.2 South_Africa_1900BP.SG Target: Mongolia_Salkhit_UP.SG alkhit1_noUDG.SG__BC_32923__Cov_12.87%Distance: 5.6397% / 0.05639683 | ADC: 0.25x RC 47.0 Jarawa 13.0 Han 12.0 TUR_Barcin_N 10.4 Yamnaya_RUS_Samara 5.6 ETH_4500BP 5.4 WHG 5.0 Nganassan 1.6 BRA_LapaDoSanto_9600BP Code: Distance to: Mongolia_Salkhit_UP.SG:salkhit1_noUDG.SG__BC_32923__Cov_12.87%Very large distances... but I guess to be expected. Distance to: Mongolia_Salkhit_UP.SG alkhit1_noUDG.SG__BC_32923__Cov_12.87%0.74718867 Spy_final_provisional.SG 0.74788442 Mezmaiskaya1_final_provisional.SG 0.75341334 VindijaG1_final_provisional.SG 0.75588754 Mezmaiskaya2_final_provisional.SG 0.77009335 Les_Cottes_final_provisional.SG 0.77409512 Denisova11.SG 0.77760547 Goyet_final_provisional.SG 0.78088775 Ancestor.REF 0.78645009 Vindija_snpAD.DG 0.78677914 Chagyrskaya.SG 0.78776323 Altai_snpAD.DG 0.79078078 Chimp.REF 0.79332972 Denisova_snpAD.DG
03-30-2024, 05:57 PM
- G25 is not really meant for paleolithic samples
- Salkhit appears to have ~1/3 modern European contamination Other than that from a superficial glance it didn't look like anything terribly implausible. Mostly Yana + Tianyuan, maybe some sort of proto-East/Asian or modern East/Southeast-Asian contamination also. But yeah, unfortunately it's a case of 'nothing to see here'.
03-30-2024, 07:08 PM
Could you check out if my version would be of better quality?
03-30-2024, 07:23 PM
(This post was last modified: 03-30-2024, 07:23 PM by Norfern-Ostrobothnian.)
I found coordinates made on my files
Code: Mongolia_Salkhit_UP:salkhit1,-0.001138,-0.169593,-0.039221,0.061693,0.017234,-0.01506,0.017391,-0.023999,0.035178,-0.004009,0.000325,-0.003897,0.007284,-0.004679,-0.004479,-0.001326,0.007302,0.006968,0.014455,0.015632,0.001123,-0.001484,-0.007148,-0.00735,-0.017843(03-30-2024, 07:23 PM)Norfern-Ostrobothnian Wrote: I found coordinates made on my files Target: Mongolia_Salkhit_UP alkhit1Distance: 4.7563% / 0.04756322 45.4 Jarawa 13.2 Yamnaya_RUS_Samara 12.4 Han 6.6 Nganassan 6.2 ETH_4500BP 6.2 Levant_PPNB 4.8 WHG 2.6 BRA_LapaDoSanto_9600BP 1.4 TUR_Barcin_N 0.6 IRN_Ganj_Dareh_N 0.6 IRN_Shahr_I_Sokhta_BA2_I8728 As been noted, this probably doesn't make any sense because the sample should predate those of the populations being compared to.
03-30-2024, 11:30 PM
Hypothetically, I would estimate a non-contaminated Salkhit to, in G25, take mostly Yana, Tianyuan, and maybe random other populations in no-man's land, e.g. Onge, Ust-Ishim, BachoKiro, Kostenki, etc. It producing large amounts of Yamnaya and WHG is highly suspect.
03-30-2024, 11:46 PM
I don't see any contamination with Salkhit. The connections that we see to many populations appear for a reason. You may also test to Japan, TSI, Ancient greeks and WHG.. Salkhit is something very special. It has the highest levels of some archaic markers and the lowest levels of later African related markers. That makes it very special.
|
|
« Next Oldest | Next Newest »
|
![[Image: Salkhit.png]](https://i.postimg.cc/4dkwMrnD/Salkhit.png)