February 10, 2009

Y chromosomes and surnames in Britain

From the paper:
Some haplogroups that are rare (less than 10%) or absent in the controls exist at high frequencies within particular surnames: examples are hgA1a in R., E1a in Bray, G in Wadsworth, J2 in Ketley, T in Feakes, Q* in Mallinson, R1* in Northam, and R1a in Swindlehurst (Figure 2a). Attenborough provides the clearest signal of coancestry, with 87% of chromosomes belonging to hgE1b1b1, which is present at only 1% in controls.
Of interest:
Direct analysis of Y-STR haplotypes in father-son pairs gives mutation rate estimates around 2.1 x 10-3 per STR per generation (Gusmão et al. 2005), while an
‘evolutionary’ rate based on diversity accumulated in specific lineages within
populations (Zhivotovsky et al. 2004) provides a rate some three times lower,
at 6.9 x 10-4.


We therefore chose to estimate a mutation rate by typing the 17 Y-STRs in a set of deep-rooting pedigrees totalling 274 transmissions of the Y chromosome, and with a mean pairwise separation within all pedigrees of 5.6 generations (Supplementary Figure 2). This gave a rate of 1.5 x 10-3 per STR per generation. Figure 5 shows
the mean and standard deviations of ages for a total of 74 clusters based on
these parameters.

One would need to look at the specific set of markers to derive a relationship between this "King & Jobling" (KJ) mutation rate, but it appears to be ~0.75 of the germline mutation rate. In the first post of my Y-STR series, I argued that Y-STR variance (not identical, but related to the ρ measure used here) accumulates at near the germline rate. The KJ estimate seems closer to the germline rate than to the ~0.3 (slow) Zhivotovsky et al. mutation rate. Note also that the age estimates for relatively young groups (first figure in this post) tend to be underestimates (1:1.17 for the youngest data point), which further supports the thesis that Y-STR diversity accumulates at near (but not exactly) the germline rate.

Note also that the time depth of British surnames is not too widely different from the populations used by Zhivotovsky et al. (Bulgarian Gypsies, Maori) for calibration of their evolutionary mutation rate. Therefore, it appears that the Zhivotovsky et al. rate is inconsistent with British surnames, and this underscores the difficulties with archaeological calibration of the mutation rate I talked about here.

Molecular Biology and Evolution doi:doi:10.1093/molbev/msp022

Founders, drift and infidelity: the relationship between Y chromosome
diversity and patrilineal surnames

Turi E. King and Mark A. Jobling


Most heritable surnames, like Y chromosomes, are passed from father
to son. These unique cultural markers of coancestry might therefore have a
genetic correlate in shared Y chromosome types among men sharing
surnames, although the link could be affected by mutation, multiple
foundation for names, nonpaternity, and genetic drift. Here, we demonstrate
through an analysis of 1678 Y-chromosomal haplotypes within 40 British
surnames a remarkably high degree of coancestry that generally increases as
surnames become rarer. On average, the proportion of haplotypes lying
within descent clusters is 62%, but ranges from zero to 87%. The shallow timedepth of many descent clusters within names, the lack of a detectable effect of
surname derivation on diversity, and simulations of surname descent suggest
that genetic drift through variation in reproductive success is important in
structuring haplotype diversity. Modern patterns therefore provide little
reliable information about the original founders of surnames some 700 years
ago. A comparative analysis of published data on Y diversity within Irish
surnames demonstrates a relative lack of surname frequency dependence of
coancestry, a difference probably mediated through distinct Irish and British
demographic histories including even more marked genetic drift in Ireland.



Gioiello said...

But a mutation rate of 0,0015 calculated on 700 years may become 0,00069 if calculated on 7,000 or 14,000 years. Within 700 years we haven't the time for back mutations and so on. What I have said in the past remains valid: we must calculate the mutation rate for long distances of time not around the modal but from the more distant markers from the modal.

Dienekes said...

But a mutation rate of 0,0015 calculated on 700 years may become 0,00069 if calculated on 7,000 or 14,000 years.

The 0.00069 rate was calculated in the 2004 paper by Zhivotovsky et al. for a much shorter time span (about 1ky). In the 2006 letter by Zhivotovsky, Underhill, and Feldman, it was attempted to explain this rate, but the explanation made sense only for very small haplogroups that never grew to large numbers.

Currently, there is absolutely no reason to prefer the 0.00069 rate for any molecular age estimation.

Gioiello said...

You said that for times longer than 700 years, for example x10, the mutation rate (of Vizachero, Nordtvedt, etc. ) is understimated at a percentage of 40%. O,00069, i.e. 0,0007, is half of the rate of this paper (0,0015), i.e. for about a 100%. It seems to me that the difference isn't insurmountable and for putting my R1b1b2/ht 35 in the Italian Refugium at the Younger Dryas time I need of much less.

Dienekes said...

You said that for times longer than 700 years, for example x10, the mutation rate (of Vizachero, Nordtvedt, etc. ) is understimated at a percentage of 40%

I think you are referring to the mtDNA paper I linked to, and that is for a different time frame than 700+ years.

Gioiello said...

Yes, you were referring to mtDNA, but don't you think that the same calculation should be done for YDNA? The back mutations are very rare in mtDNA, but are very frequent in YDNA. I would think that for YDNA the understimate is therefore much much strong.

Anonymous said...

With the surname and haplogroup study, it is a starting point to find the real origins of those surnames, their first entry into the books so to speak, and their spread to the present day. I am sure David and Richard Attenborough will be please to know they have a high chance of being E1b1b. Oddly the surname does sound foreign, the Atten bit. Sort of Egyptian, lol.

I will be please when the Eurocentric tendency to pick ever finer subdivisions of common as muck haplogroups in Europeans stops. There is nothing particularly European about R1b except its commonality. Probably says something about the moral nature and sexual proclivities of R1b men than anything to do with the original peopling of Europe in the Paleolithic, the LGM, the Neolithic or the Bronze Age I.E language spread. Face it you R1b men, you are just a bunch of male tarts.

I am not a fan of Mr. Zhivotovsky and his germ line theory. I will always go by observation and data, which can be replicated numerous times than someone's well meaning hypothesis which is at best coincidental with data. The use of the Maoris, Bolgar intrusion into Europe or the Romani are just too pat, too neat. We basically know from historical documents certain things like the arrival of the Romani or mention of their arrival in certain parts of the world outside Hindustan. We know from linguistics, archeology, and other references the approximate time the Maoris arrived in New Zealand. Similarly we know about the arrival of people from Indonesia to Madagascar or Yehudis to Europe and elsewhere. That is not in contention. The finding of the estimated dates by Mr. Zhivotovsky using those historic timings is too pat, and appears contrived.

The fact is this: mutations occur and cannot be timed by any chromometer or any formulae other than one based on Chaos Theory. Reminds me of Goldberg's character in Jurassic Park. Studying real changes in real people many times over can provide an approximate rate that mimics the true chaotic rate of mutations.

One thing that is never studied is this: Why do new SNPs occur when humans leave a geographical region after long habitation to live in another geographical region? Why did M89 occur in the Middle East? What happened to M89 man for him to eventually spawn men containing other downward SNPs. Haplogroup F still exists, haplogroup IJ does not. There must be something in the geographic regions that caused new SNPs or mutations to allele values of STRs to occur. Mating practices or sexual preferences or genetic drift or inbreeding or bottlenecks cannot account for new SNPs or mutations. Even the vast array of SNPs used in modern, up-to-date dna studies show a geographic bias. Most of those SNPs code for nothing, some just indicate how coding sections of dna may behave, and having green eyes compared with blue eyes is not really of any genetic importance to human survival or extinction prospects. Most people's individual genome is identical to every other human, just minor differences based on geography. Geography is the key to diversity, and the underlying reasons for new SNPs or mutations to STRs. Geography leads to separation which leads to genetic drift and founder effects. The reason we look different is just that our ancestors spent long periods apart from other human groups and develop in their own way more in accordance to their environment which leads back to geography.