Tag Archives: reich

ANI-ASI Admixture Dating

Similar to an earlier conference poster, Reich Lab's Priya Moorjani et al have another poster at SMBE. Here's the abstract:

Estimating a date of mixture of ancestral South Asian populations
Linguistic and genetic studies have demonstrated that almost all groups in South Asia today descend from a mixture of two highly divergent populations: Ancestral North Indians (ANI) related to Central Asians, Middle Easterners and Europeans, and Ancestral South Indians (ASI) not related to any populations outside the Indian subcontinent. ANI and ASI have been estimated to have diverged from a common ancestor as much as 60,000 years ago, but the date of the ANI-ASI mixture is unknown. Here we analyze data from about 60 South Asian groups to estimate that major ANI-ASI mixture occurred 1,200-4,000 years ago. Some mixture may also be older—beyond the time we can query using admixture linkage disequilibrium—since it is universal throughout the subcontinent: present in every group speaking Indo-European or Dravidian languages, in all caste levels, and in primitive tribes. After the ANI-ASI mixture that occurred within the last four thousand years, a cultural shift led to widespread endogamy, decreasing the rate of additional mixture.

I bolded the portion which seems new compared to the previous abstract.

Related Reading:

HarappaWorld Ancestral South Indian

Using the same method as I used for reference 3 admixture, I decided to guesstimate the Ancestral South Indian proportions, as given by Reich et al, for my HarappaWorld admixture run.

Basically, I used the 92 (out of the 96 samples Reich et al used) to find population averages for the South Indian component. Then, I used linear regression between the South Indian component average and Reich et al's estimate of Ancestral South Indian (ASI) ancestry. Since Reich et al actually list Ancestral North Indian percentages in their paper but their model is a two-ancestry ANI+ASI one, I simply calculated the ASI percentages as 100% minus ANI.

The correlation between Reich et al ASI and my HarappaWorld South Indian component for the relevant populations turns out to be 0.99277086.

And the linear regression fit for the data is:

ASI = 2.5218942 + 0.8104836 * S_INDIAN

where both ASI (Reich et al) and S_INDIAN (HarappaWorld) are given in percentages.

Of the individuals in HarappaWorld, I kept only those who had a South Indian component of at least 20% for computing the ASI proportions.

The resulting ASI percentages can be seen in a spreadsheet.

Please note that in the Group sheet, the averages are based on the samples which met the 20% South Indian component threshold. Thus, the 20% ASI in the Romanians is the average of the two Romanians who met the threshold out of a total of 16 Romanian samples.

The individual results are available in the Individual sheet. These results are a little different from the estimates using reference 3. Thus, I would point out that these should be taken only as a rough estimate.

Related Reading:

Indian Cline III

I have been working on creating 100% ASI (Ancestral South Indian) samples recently. So it was really interesting that Dienekes did similar experiments:

I am going about creating the "pure" allele frequencies somewhat differently, so that would be a useful exercise.

Anyway, I thought you guys would be itching for some new results. So here's a PCA plot:

This used the same Principal Component Analysis as the one here using the 96 Indian Cline samples, Utahn Whites and Onge. However, I projected three extra "populations" on this plot.

These three populations are simulated genetic data of 25 individuals using the allele frequencies from Reference 3 Admixture results.

  1. Onge11 is generated from the Onge (C2) component from K=11 admixture for Reference 3.
  2. SA11 is generated from the South Asian (C1) component from the same K=11 admixture.
  3. SA12 is generated from the South Asian (C1) component from the K=12 admixture.

As you can see, the SA12 population lies between 100% ASI and the Indian Cline samples.

The Onge11 generated samples are a bit beyond 100% ASI on the first principal component, but they are also shifted towards the real Onge on pc2.

Related Reading:

Misuse of Correlation

I have been misusing correlation in computing Ancestral South Indian percentages from PCA/ADMIXTURE and Reich et al population-level averages.

I have tried to make it clear that just looking at the correlation is not enough, that an admixture component is not similar to ASI just because it correlates well with Reich et al's ASI averages for the 18 Indian cline populations. Even when the correlation is higher than 0.99. To illustrate what I mean, let's look at the Ref4C admixture runs.

I calculated the mean for each admixture component from the K=2 to K=12 runs for the 18 Indian cline populations and then computed the correlation between that and the Reich et results. Let's take a look:

K Component Correlation
2 C1 Euro-Afro -0.9941887
3 C2 East Asian 0.9955347
4 C3 European -0.993933
5 C3 European -0.993277
6 C1 South Asian 0.9675099
7 C1 South Asian 0.993081
8 C1 South Asian 0.9932762
9 C1 South Asian 0.9914145
10 C1 South Asian 0.9918095
11 C1 South Asian 0.9919097
12 C1 South Asian 0.9918594

Where do you see the highest correlation? At K=3 ancestral populations, the East Asian component is very highly correlated with ASI for the Indian cline populations. Does that mean that we could use that to compute ASI? No, not at all. While it is expected that at K=3, ASI would be a little closer to East Asian than to European, East Asian is not a good proxy for ASI at all since we cannot extrapolate to other individuals and populations.

Related Reading:

Indian Cline II

One thing I forgot in the post yesterday about the Indian cline was to try to extrapolate from the PCA results to 100% ANI (Ancestral North Indian) and 100% ASI (Ancestral South Indian).

This is a simple linear extrapolation which should be okay since PCA is linear.

The "N" denotes the extrapolated position of ANI and "S" denotes the ASI. The points to the left of "N" are all Utahn Whites while the Onge are on the bottom right of the graph.

As you can see, the ASI is about the same as Onge in terms of eigenvector 1 (which represents the Indian cline approximately), but ASI is far from Onge on the 2nd eigenvector. That is expected since the Onge have been separated from the mainland populations for a long time.

The more interesting thing is that the extrapolated position of ANI is a little to the right of all the Utahn Whites.

We'll need a similar analysis of the Indian cline with more populations to see which one the ANI is closest to.

PS. I should point out that I am using correlation between a limited number of population statistics to find a relationship between the 1st principal component and Reich et al's ASI estimate. This has a number of drawbacks. It would be much better to compute ASI directly.

Related Reading:

Indian Cline

I had used linear regression to estimate Ancestral South Indian (ASI) component from Reference 3 K=11 admixture run. Now here are a couple more exercises along the same lines but much simpler.

Just using the 96 Indian cline samples from Reich et al to compute PCA or admixture doesn't work as the Chenchu separate out in both analyses from the rest. So I added the Utahn White (CEU) samples from HapMap and the Onge from Reich et al.

First, I ran supervised admixture with two ancestral components, Utahn Whites and Onge. Here's the Onge component plotted against Reich et al's ASI estimate along with a linear regression estimate. The correlation between the two is 0.9908.

Second, I ran Principal Component Analysis (PCA) on the Indian cline samples plus Utahn Whites and Onge. Here are the first two PCA dimensions plotted. The first eigenvector explains 4.04% of the total variation and the 2nd explains 1.94%.

The first principal component is mostly along the Indian cline while the second one basically separates the Onge from everyone else.

Using the 1st principal component to estimate ASI, here's the plot with Reich et al's ASI estimate along with a regression line. The correlation between pc1 and ASI is 0.9929.

Note that both these methods work only if the samples are on the Indian cline, i.e., they don't have any other admixture.

And now for comparison, here's the linear regression for the Reference 3 K=11 admixture Onge component and ASI. The correlation here is 0.9949. Note that this is a little different than my previous analysis since I calculated the population averages using only the 96 samples recommended by Reich et al.

Here's a spreadsheet containing the data for these three runs.

There are a couple more tricks I have to figure out some things regarding Ancestral South Indian admixture. Let's hope they provide us some insight.

Related Reading:

Reference 3 Admixture K=11

Continuing with the admixture analysis with our new reference 3 dataset.

Here's the results spreadsheet for K=11.

You can click on the legend to the right of the bar chart to sort by different ancestral components.

You don't know how excited I am to see the Onge (C2) component. Let's compare the Onge component with Reich et al's ASI (Ancestral South Indian):

Reich ASI % Onge Component %
Mala 61.2 39.9
Madiga 59.4 37.9
Chenchu 59.3 38.6
Bhil 57.1 37.5
Satnami 57 36.4
Kurumba 56.8 39.5
Kamsali 55.5 35.5
Vysya 53.8 34.4
Lodi 50.1 31.8
Naidu 49.9 32.1
Tharu 49 32.2
Velama 45.3 28.9
Srivastava 43.6 27.8
Meghawal 39.7 25.4
Vaish 37.4 23.8
Kashmiri-Pandit 29.4 17.6
Sindhi 26.3 13.4
Pathan 23.1 10.6

Let's plot that with a linear regression:

How do you like that?

Now let's take all the reference populations with an Onge component between 10% to 50% and use the equation above to calculate their ASI percentage. The results are in a spreadsheet. There are several populations with an even higher Ancestral South Indian than any of the Reich et al groups, with Paniya being the highest at 67.4%.

Fst divergences between estimated populations for K=11 in the form of an MDS plot.

I guess you might want to see the Fst dendrogram too. Just remember it's not a phylogeny.

And the numbers:

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10
C2 0.165
C3 0.121 0.122
C4 0.090 0.161 0.152
C5 0.071 0.152 0.137 0.048
C6 0.134 0.144 0.067 0.163 0.143
C7 0.184 0.224 0.216 0.179 0.186 0.232
C8 0.210 0.209 0.205 0.235 0.223 0.228 0.286
C9 0.175 0.207 0.139 0.208 0.178 0.141 0.281 0.290
C10 0.261 0.304 0.294 0.257 0.261 0.311 0.123 0.367 0.364
C11 0.150 0.195 0.187 0.143 0.148 0.203 0.059 0.260 0.252 0.133

Related Reading:

Introducing Reference 3

Having collected 12 datasets, I have gone through them and finally selected the samples and SNPs I want to include in my new dataset, which I'll call Reference 3.

It has 3,889 individuals and 217,957 SNPs. Since this is a South Asia focused blog, there are a total of 558 South Asians in this reference set (compared to 398 in my Reference I).

You can see the number of SNPs of various datasets which are common to 23andme version 2, 23andme version 3 and FTDNA Family Finder (Illumina chip).

The following datasets had more than 280,000 SNPs common with all three platforms and hence were included in Reference 3:

  1. HapMap
  2. HGDP
  3. SGVP
  4. Behar
  5. Henn (Khoisan data)
  6. Rasmussen
  7. Austroasiatic
  8. Latino
  9. 1000genomes

Reich et al had about 100,000 SNPs in common with 23andme (v2 & v3 intersection) and 137,000 with FTDNA, but there was not a great overlap. Only 59,000 Reich et al SNPs were present in all three platforms. Since I really wanted Reich et al data in Reference 3, I included it but the SNPs used for FTDNA comparisons won't be the same as for the 23andme comparisons.

Of the datasets I could not include, I am most disappointed about the Pan-Asian dataset since it has a good coverage of South and Southeast Asia. Unfortunately, it has only 19,000 SNPs in common with 23andme v2 and 23,000 with 23andme v3. I am going to have to do some analyses with the Pan-Asian data but it just can't be included in my Reference 3.

I am also interested in doing some analysis with the Henn et al African data with about 52,000 SNPs for personal reasons.

Xing et al has about 71,000 SNPs in common with 23andme v3, so some good work could be done with that, though I'll have to use only 23andme version 3 participants.

The information about the populations included in Reference 3 is in a spreadsheet as usual.

Related Reading:

Dienekes on ANI/ASI

Dienekes has a word of caution about choosing reference populations and admixture results.

Consider a sample of 25 Mexicans from the HapMap and 25 Yoruba from the Hapmap, 25 Iberian Spanish from the 1000 Genomes Project, and 25 Pima from the HGDP as parental populations. We obtain for our Mexican sample:

  • 59.7% European
  • 36.9% "Native American"
  • 3.4% African

Let's run a final experiment with just the Mexicans, Spanish, and Yoruba, i.e., with no Native American samples. At K=3 we obtain:

  • 70% "Native American"
  • 29.7% European
  • 0.4% African

The "Native American" component has increased again! The explanation is simple: as we exclude less admixed Native American groups, Mexicans appear (comparatively) more Native American. The "Native American pole" has shifted, and so has the relative position of populations between them.

In other terms, what is labeled "Native American" in the three experiments is not the same: in the first one it is anchored on the more unadmixed Pima, in the last one in the more admixed Mexicans.

Thus, it seems that unadmixed reference samples are much more useful in getting good results from Admixture.

Then he runs Admixture on the Reich et al dataset for South Asians and tries to estimate the relationship between the Ancestral North Indian percentage computed by Reich et al and his K=2 admixture results on the same data.

Dienekes then included South Asian Dodecad participants in the analysis and ran a K=4 admixture analysis on Reich et al + Dodecad South Asian data, including Yoruba and Beijing Chinese from the HapMap to catch any African or East Asian ancestry.

Here are the admixture results for the reference populations:

The R2 correlation between the West Eurasian admixture component and the Reich et al ANI component is 0.98 which is good. His relationship equation comes out to:

ANI = 0.779*WestEurasian + 39.674

Using this relationship, he calculates the ANI and ASI (Ancestral South Indian) components for Dodecad project members. My results (DOD128) are as follows:

East Eurasian 0.0%
African 3.5%
Ancestral North Indian 75.9%
Ancestral South Indian 20.6%

I should point out that due to my recent Egyptian ancestry, my ANI result is wrong since it's collecting all of the non-African Egyptian in there too.

Also, in the case of Razib, I don't think his East Asian 14.4% should be separated out from his ANI-ASI like that. At least some of it should form part of his ASI percentage in my opinion.

Otherwise, this seems like a very good exercise by Dienekes.

Related Reading:

Reich et al and Pan-Asian Datasets

I got access to the Reich et al (Nature 2009) dataset used in their paper "Reconstructing Indian population history".

It has the following populations:

Aonaga Aus Bhil
Chenchu Great_Andamanese Hallaki
Kamsali Kashmiri_Pandit Kharia
Kurumba Lodi Madiga
Mala Meghawal Naidu
Nysha Onge Sahariya
Santhal Satnami Siddi
Somali Srivastava Tharu
Vaish Velama Vysya

There are 141 individuals with 587,753 SNPs in their dataset which conveniently is in PED format.

Also, Blaise pointed me to the Pan-Asian SNP data used in the Dec 2009 Science paper "Mapping Human Genetic Diversity in Asia".

It includes the following 71 populations:

Maya Auca Quechua Karitiana Pima
Ami Atayal Melanesians Zhuang Han_Cantonese
Hmong Jiamao Jinuo Han_Shanghai Uyghur
Wa Alorese Dayak Javanese Batak_Karo
Lamaholot Lembata Malay Mentawai Manggarai
Kambera Sunda Batak_Toba Toraja Andhra_Pradesh
Karnataka Bengali-Assamese Rajasthan Uttaranchal Uttar Pradesh
Haryana Spiti Bhili Marathi Japanese
Ryukyuan Korean Bidayuh Jehai Kelantan
Kensiu Temuan Ayta Agta Ati
Iraya Minanubu Mamanwa Filipino Singapore_Chinese
Singapore_Indian Singapore_Malay Hmong (Miao) Karen Lawa
Mlabri Mon Paluang Plang Tai_Khuen
Tai_Lue H'tin Tai_Yuan Tai_Yong Yao
Hakka Minnan

It has 1,719 individuals with 54,794 SNPs. I wish it had more SNPs considering the wealth of populations.

Also, the Pan-Asian data is in the form of minor allele counts, so I need to convert that back to A/C/G/T. Since there are some HapMap populations included in the dataset, that shouldn't be too hard.

I am going to include both these datasets into my big reference set.

Related Reading: