Y-haplogroup Q and Native American origins (Regueiro et al. 2013)

From the paper:

Table 2 provides coalescence time estimations based on 15 Y-STR loci for haplogroup Q-M242 and subhaplogroups Q1a3-M346, Q1a3a-L54 and Q1a3a1-M3. Due to the limitations and assumptions associated with the current calibrations of Y-STR mutation rates (Zhivotovsky et al., 2004; Goedbloed et al., 2009; Ravid-Amir and Rosset, 2010; Burgarella and Navascue/s, 2011), the dates generated in this study should only be taken as relative estimates. However, these relative values may be useful for comparisons among populations. Using the pedigree mutation rate (average mutation rate of 0.0025 per locus per generation; Goedbloed et al., 2009), we obtained coalescence estimates that were approximately three times younger than those calculated with the evolutionary rate (average mutation rate of 0.00069 per locus per generation; Zhivotovsky et al., 2004). In general, the genealogical estimates are more compatible with archeological data than the evolutionary rates 

According to Table 2, the oldest TMRCA for M242 chromosomes is ~11ky using the pedigree rate and ~29ky using the evolutionary rate. In a previous post, I argued that in order to account for the fact that modern-day haplogroups have millions of modern representatives, a fairly high growth rate must be assumed for them, with one estimate of the effective rate being 0.84μ, where μ is the genealogical rate. Ergo, the ~11ky time estimate must be updated to something like ~13ky, which corresponds reasonably well -within the confidence limits- to the first colonization of the Americas.

The paper's conclusion:

Overall, our data are best explained by invoking a single major pre-Holocene migration that proceeded eastward in a trans-continental trek across Beringia and then southward to transverse the length of Americas-a scenario that fits nicely with the South Altaian origin of Native Americans as proposed by Dulik et al. (2012). The subsequent winnowing of the Native American gene pool via repeated founder effects and bottleneck events could have produced the Y chromosome distribution illustrated on the pie map (Fig. 2). The Q haplogroup frequency pattern of the Native Americans features: 1) a dramatic reduction of the ancestral L54 and MEH2 lineages of Central Asia and/or northeast Siberia and 2) a concomitant increase in the derived M3 state, which exerts total domination of the Q landscape in nearly all of South American reference populations examined. We also see evidence of a dramatic Mesoamerican postmigration population growth in the ubiquitous and diverse Y-STR profiles of the Mayan and other Mesoamerican populations in the PCA (Fig. 4), and the M242 and M3 networks (Fig. 5A,D). In the case of the Mayans, this demographic population growth was most likely fueled by the agricultural- and trade-based subsistence adopted during the pre-Classic age of their empire. Our results indicate that the oldest dates for Q-M242 are found in Northeast Siberia followed by populations from Mesoamerica, which is most likely a consequence of demographic expansion as discussed above. The diversity levels observed in the Altaian and Tuvinian regions of Central Asia, the lowest of all populations examined may be the consequence of bottleneck events fostered by the spatial isolation and low effective population size characteristic of a nomadic lifestyle. 

It seems likely that the migration of Q-M242 descendants corresponds mainly to the "First Americans" sensu Reich et al. (2012) which makes up the bulk of Amerindian Y-chromosomes. Interestingly:

The recently sequenced genome of a Paleo-Eskimo _ 4,000 years old, belonging to the Saqqaq culture, provides evidence for a more recent migration from Siberia into the New World some 5.5 kya, independent of the pre-Holocene penetration that gave rise to the modern Native Americans and Inuit (Rasmussen et al., 2010). In addition, the Paleo-Eskimo individual is a member of the haplogroup Q1a*-MEH2 suggesting that this lineage likely traces a population migration originating in Northeast Siberia across the Bering Strait (Rasmussen et al., 2010). 

and:

In the MDS plot, we observe a segregation between Eskimo populations from northeast Siberia and the Native American populations, differentiation likely due to the northeast Siberian presence of the MEH2 mutation which defines the Q1a* haplogroup.  

So, it would appear that the additional "Eskimo" wave may be discernible within Q itself; the third "Na-Dene" wave cannot be distinguished on the basis of Q alone, and probably reflects the later entry of haplogroup C.


Am J Phys Anthropol DOI: 10.1002/ajpa.22207

On the Origins, Rapid Expansion and Genetic Diversity of Native Americans From Hunting-Gatherers to Agriculturalists

Maria Regueiro et al.

Given the importance of Y-chromosome haplogroup Q to better understand the source populations of contemporary Native Americans, we studied 8 biallelic and 17 microsatellite polymorphisms on the background of 128 Q Y-chromosomes from geographically targeted populations. The populations examined in this study include three from the Tuva Republic in Central Asia (Bai-Tai, Kungurtug, and Toora-Hem, n = 146), two from the northeastern tip of Siberia (New Chaplino and Chukchi, n = 32), and two from Mesoamerica (Mayans from Yucatan, Mexico n = 72, and Mayans from the Guatemalan Highlands, n = 43). We also see evidence of a dramatic Mesoamerican post-migration population growth in the ubiquitous and diverse Y-STR profiles of the Mayan and other Mesoamerican populations. In the case of the Mayans, this demographic growth was most likely fueled by the agricultural- and trade-based subsistence adopted during the Pre-Classic, Classic and Post-Classic periods of their empire. The limited diversity levels observed in the Altaian and Tuvinian regions of Central Asia, the lowest of all populations examined, may be the consequence of bottleneck events fostered by the spatial isolation and low effective population size characteristic of a nomadic lifestyle. Furthermore, our data illustrate how a sociocultural characteristic such as mode of subsistence may be of impact on the genetic structure of populations. We analyzed our genetic data using Multidimensional Scaling Analysis of populations, Principal Component Analysis of individuals, Median-joining networks of M242, M346, L54, and M3 individuals, age estimations based on microsatellite variation utilizing genealogical and evolutionary mutation rates/generation times and estimation of Y- STR average gene diversity indices.

Link