Total Pageviews

Thursday, September 6, 2012

mtDNA Haplogroup H5

The interesting thing about the Kurdish individual with the H5a1 haplogroup is that the mtDNA contains SNPs that are characteristic for H5a iand H5b. I discovered it by using the mtDNA haplogroup predictor and GenBank data summarized by Ian Logan:

H2a2a1(rCRS) ⇨ 263G ⇨ H2a2a ⇨ 8860G 15326G ⇨ H2a2 ⇨ 750G ⇨ H2a ⇨ 4769G ⇨ H2 ⇨ 1438G ⇨ H ⇨ 456T ⇨ H5'36 ⇨ 16304C ⇨ H5 ⇨ 4336C ⇨ H5a ⇨ 15833T ⇨ H5a1 ⇨ 3397G 5471A

The definition of mtDNA haplogroup H5b is having a mutation from G to A at postion 5471, a.k.a. G5471A (or short 5471A). Other mtDNA haplogroups with this G5471A mutation are HV7 and N1b. To summarize, this mutation occurred multiple times during human evolution.

Another thing that effects the the midpoint and thus the outcome of the analysis are repeating mtDNA samples in the tree, so I am excluding them, too.

In the H5 branch I am excluding these because they are exactly like [HM625680 Kloss]:
GQ983083(Italy) Santoro
GQ983085(Italy) Santoro
GQ983086(Italy) Santoro
GQ983094(Italy) Santoro

In the H36 branch I am excluding these because they are exactly like [FJ348166 Irene]:
FJ348167 Irene
FJ348168 Irene
FJ348169 Irene

In the H36 branch I am also excluding these because they are exactly like [FJ348151 Irene]:
FJ348152 Irene

 In Figtree, I generated rectangular tree of H5:

Same data of H5 as polar tree:

From the first look at the data, it seems like that the root European H5a originated in the Middle East (bright yellow), which is in agreement with ancient DNA data.

From Wikipedia:

H5 has been dated to around 11,500 BP (9500 BC).[5] It appears to be most frequent and diverse in the Western Caucasus, so an origin there has been suggested, while its subclade H5a appears European.[6] However samples of mtDNA with T16304C in the HVR1 region have been found in four individuals of around 6800 BC from the Pre-Pottery Neolithic B site of Tell Halula, Syria,[7] suggesting that H5 may have arrived in the Caucasus with farmers from the Near East.
   This blunt conclusion needs more thoughts. I will update it.

Wednesday, September 5, 2012

mtDNA Haplogroup H5a1

Today, I want show some data about mtDNA haplogroup H5a1 because there is a Kurdish individual with H5a1 in this Kurdish DNA project.

I analyzed the phylogeny of this haplogroup by using fully sequenced and published mtDNA data from GenBank.

To do so I first downloaded the data from GenBank, then I used CLUSTALW, (mode: slow accurate pairwise alignment) to align the sequences and create a rooted phylogenetic tree with branch length (UPGMA). The data were pasted into CLUSTALW in the Fasta format.

I realized that all the data of Herrnstadt et al. 2002 are lacking the first 577 nucleotides, which is messing up the position of Herrnstadt samples in the trees/network and is messing up the position of the rest. The same effect can be seen with the two samples of Kivisild et al.; they are lacking 236 nucleotides. Thus, I excluded these samples.

Rooted phylogenetic tree with branth length (UPGMA) of H5a1:

The nice thing about CLUSTALW is that it also generates a "dnd file" and an "aln file" of the alignment.

The dnd file can be opened with the Figtree software. In Figtree, I generated another tree of H5a1:

The aln file can be opened with the Splitstree software. In Splitstree4, I generated a network (Convex Hull) of H5a1:

H5a1a (T721C mutation): three individuals in these trees have this mutation [AF346975(Dutch) Ingman; Q983087(Italy) Santoro; HQ659693(Polish) FTDNA]

H5a1b (G11719A mutation): two individuals in these trees have this mutation [AY495167(European) Coble; AY495176(European) Coble]

H5a1c1a (C4095T  G13194A G9055A A2851G mutations): only one individual in these trees has these mutations [HQ663878(Danish) FTDNA]

H5a1d (A8803D mutation): only one individual in these trees has this mutation [AY495171(European) Coble]

H5a1e (A16166G mutation): two individuals from Finland belong to the H5a1e branch [AY339431(Finland) Moilanen; AY339432(Finland) Moilanen].

H5a1f (T961C mutation): one individual in these trees has this mutation [ JN646689(Polish) FTDNA]

H5a1g1 (T16172C, A444G, G9804A, T16311C mutations): two individuals in these trees have these mutations [EU294323 FTDNA; HQ645111(English) FTDNA]. Since HQ645111 has the additional  T1284C and A7517G mutations it belongs to H5a1g1a.

Note: In the Finland DNA project I found one individual (N48161 Mary Anne Bodle, b.1791, Plumstead, Kent) that originated in England and has the same mutations of H5a1g1 HV regions, i.e.  A444G, T16172C, T16311C.

H5a1k (T12864C): two individuals have this mutation [AY495170(European) Coble, GQ983064(Italy) Santoro]

H5a1p (T16093C): three individuals have this mutation [FJ966912 FTDNA, GQ983075(Italy) Santoro, GQ983084(Italy) Santoro]

Sunday, September 2, 2012

Cultural Distance Calculator Part2

This is a follow-up for the Cultural Distance Calculator:

Since there are some cultural data available I decided to use phylogeny software to present the results. This helps to detects groups of population that have similar cultural values and behavior.  In order to visualize the data of the first 4 dimensions.First, I calculated a distance matrix for all populations of the "Old World" (N=69). Then, I had to adjust the IDs to 10 digits to prevent malfunction of the Fitch software. In Fitch, I used 10 randomized runs to improve the results.

The tree of "Old world" cultures:
Europe: red
Middle East and North Africa: green
Asia: green asparagus
New World "Latin America": brown
New World "English-speaking": grey 
Pacific: pink
Africa: black

Some of the surprising and interesting observations:

1. Scandinavians form one cultural cluster.
2. Australia, USA, Canada, New Zealand and South Africa are in the British/Irish cultural cluster.
3. The closest to British/Irish cultural cluster are Central Europeans and Israelis.
4. The "Catholic Cluster" is formed by Argentina, Spain, France, Belgium, Poland, and Malta.
5. The "Arab World Cluster" (Iraq, Saudi-Arabia, UAE, Kuwait) shows cultural similarities to some Latin-Americans (Guatemala, Panama, Surinam, Mexico, Brazil) and some East Europeans, mostly Orthodox Christians (Russia, Serbia, Romania)
6. Culturally, Turkey is more similar to Balkan people (Bulgaria, Croatia) than to the Middle East.
7. Culturally, Albania is more similar to Ecuador, Colombia and Venezuela than to the Balkan
8. Some European countries (Portugal, Greece, Slovenia) form a cultural cluster with Latin Americans (Uruguay, Chile, Costa Rica, El Salvador, Peru) and South Korea. Egypt is not far away from that cluster.
9. Singapore, Hong Kong, China, and Vietnam on one side, and Philippines, Malaysia, and Bhutan on the other side form two closely related clusters.
10. Dominican Republic forms a cultural cluster Ethiopia and Kenya.
11. Culturally, Nepal is more similar to some African countries (Malawi, Zambia, Namibia, Tanzania, Sierra Leone, Senegal)
12. Honduras, Indonesia and the Fiji Islands form a cultural cluster.