Thoughts on an extremely ancient root of the human Y tree

I was recently interviewed by Alan Boyle at NBC to comment on: 

 An African American Paternal Lineage Adds an Extremely Ancient Root to the Human Y Chromosome Phylogenetic Tree 

Fernando L. Mendez, Thomas Krahn, Bonnie Schrack, Astrid-Maria Krahn, Krishna R. Veeramah, August E. Woerner, Forka Leypey Mathew Fomine, Neil Bradman, Mark G. Thomas, Tatiana M. Karafet and Michael F. Hammer

The American Journal of Human Genetics, 28 February 2013

doi:10.1016/j.ajhg.2013.02.002

I do think the paper is very exciting. The identification of a new Y lineage is always interesting, and this one appears to be very long-lived. However, after a more careful reading, and some thought, I am not sure I agree with the way the TMRCA (Time to the Most Recent Common Ancestor) of the Y chromosomes was computed. And the ancient TMRCA depends quite a bit on the TMRCA. I have written up my thoughts and submitted them to the American Journal of Human Genetics, AJHG. The AJHG has a pretty strict pre-print policy (emphasis is mine):

"Work intended for submission to AJHG, currently under consideration at AJHG, or in press at AJHG may not be discussed with the media before publication. Providing preprints, granting interviews, discussing data with members of the media, or participating in press conferences in advance of publication without prior approval from the AJHG editorial office may be grounds for rejection."

But, I have gotten permission from the editor to discuss my thoughts about the submitted manuscript with colleagues (and in blog form). I am sharing the full AJHG manuscript with Mendez, but want to summarize here: 

Mendez et al. identify a Y chromosome haplotype that has not been characterized before and, with more work, they determine that it is nearly identical to a small group of Y chromosomes from Cameroon. They also estimate the TMRCA for the Y haplotype phylogeny, including this new Y chromosome and find it to be at least twice as large as anyone else, and as noted by the authors themselves, this TMRCA is inconsistent with what is known in the human fossil record.  While the new Y haplotype does increase the diversity, and thus the TMRCA, the TMRCA calculation is extremely sensitive to the mutation rate used. Mendez et al. advocate for using a mutation rate from human pedigree data instead of from comparative genomics. They then derive a mutation for the human Y chromosome from the mutation rate estimated from autosomal pedigree data. The equation they use assumes a linear correlation between the mutation rate on the autosomes, and the mutation rate on the Y chromosome. 

The authors themselves note that the TMRCA they suggest is incompatible with the known human fossil record. I present a case in my response that: 1) it is not appropriate to assume a linear correlation between the mutation rate on the autosomes and the mutation rate on the Y chromosome; 2) the mutation rate Mendez et al. computed for the Y from autosomal data is an order of magnitude lower than the mutation rate that was measured for the Y chromosome from a pedigree analysis in 2009; 3) the resulting TMRCA is inconsistent with what is known about diversity on the mtDNA, autosomes and X chromosome. Further, our own research suggests that selection is acting to reduce diversity on the Y chromosome relative to the autosomes, X, and mtDNA, which would make an extremely high TMRCA on the Y even more incompatible with observed data. 

As such, I am curious why the mutation rate measured from Y chromosomes in a pedigree analysis was not used. I think the results would still be quite exciting and novel. Given what we expect to be strong purifying selection acting to reduce diversity on the Y, the same arguments, of ancient population structure or even archaic introgression may still apply to this unique Y haplotype. 

Cross-posted at the Neilsen Lab. Please direct your comments on this article there.