Our findings of high diversity and high frequency of X haplogroup concentrated among the Galilee Druze provides new insights regarding both sampling methodologies in population genetics, and the understanding of the effects of sociological patterns on the population genetic landscape of the Near East.
The findings provide a formal demonstration that in an endogamous and consanguineous population, calculated mtDNA genetic diversity is significantly lower using random sampling than using paternal or maternal household directed sampling. Therefore, in order to obtain a comprehensive phylogenetic picture of lineage diversity in a sample set from the Druze population or populations with similar socio-demographic patterns, it is necessary to either collect a very large random sample, or to use a directed sampling strategy to ensure coverage of all of the regionally endogamous cryptic subpopulations. Random sampling taken from a regional and small sample within an endogamous subpopulation, will be relevant only for that specific subpopulation as might have been the case in the Macaulay et.al
sample, and cannot necessarily be extrapolated to the population as a whole. This is also reflected in the diversity parameters, which we have obtained using the directed sampling approach, and which do not support a recent population bottleneck or founder effect for the Druze population as a whole. It should be noted though, that directed sampling might miss localized founder effects and bottlenecks, since it does not provide the real haplotype frequencies, but rather is designed to uncover as many diverse lineages as possible for a given sample size. The decision of which sampling method is more appropriate depends on the scientific question being addressed. With regard to population substructure and genetic distances directed sampling is more appropriate to uncover cryptic diversity and regional substructure in populations such as the Druze, characterized by high levels of endogamy and local founder effects. In studies designed to uncover population allele frequencies, or population based association studies with complex clinical phenotypes, the first step will have to be elucidation of the population substructure by directed sampling, followed by random sampling from a regional subpopulation.
The absence of a recent demographic bottleneck or population founder effect for the Druze population is consistent with their oral tradition of heterogeneous ancestral origins, and also is consistent with the pattern of Mendelian genetic disease in the population. Rather than finding an excess of genetic disease with monogenic inheritance distributed throughout the population, rare monogenic recessive diseases are kindred or community specific and causative mutations have been successfully mapped in many cases using homozygosity mapping strategies
. This contrasts with the pattern observed in populations with recent bottlenecks, such a Ashkenazi Jews, in whom founder mutations have been mapped, which are distributed throughout all communities of the population
Lineage analysis within the mtDNA X-haplogroup was particularly enlightening. It should be noted that the estimated coalescence times for the major mtDNA X subhaplogroups X1 and X2 are 42,900±18,100 and 17,900±2,900 respectively
. It is striking that those different lineages (from the same parental haplogroup) whose genetic divergence date back more than ten thousand years would remain so concentrated within such a small geographic region. Mutation rates for the mtDNA coding region
, are not consistent with the possibility that this number of different coding region defined lineages within haplogroup X could have resulted from the recent expansion of a monophyletic clade within the past 1000 years. Rather this combination among the Druze, of a large number of lineages, together with a high frequency of the haplogroup in which these lineages are found, suggests descent from an ancestral population, in which the X haplogroup was more abundant than it is in the contemporary Near East, and which reflects the prevailing Near East genetic landscape at that time, antedating the establishment of the Druze religion in 1017 A.C.E. This supports the notion that the Druze represent a refugium of the population genetic structure from the time period prior to the “Dawa”, and also confirms the hypothesis of high endogamy among the Druze. The refugium hypothesis based on mtDNA haplogroup X analysis was corroborated by the finding of high diversity for the Druze mtDNA haplogroups H and K, with the added finding of novel lineages not shared with nearby populations.
Furthermore, the formal rejection of the alternate hypothesis relating to immigration to the region of individuals sharing the same mtDNA haplogroup but with lineages that diverged in antiquity, further strengthens the Druze refugium model. Although, we cannot exclude the possibility that some ratio between non-random migration and colonization did occur during and following the “Dawa” period and generated the differences in haplogroup frequencies among the current Druze subregions, and between the Druze and other populations, this explanation is highly unlikely in face of the demographic modeling results. The overall low migration rate between the Druze and all other nearby populations () cannot explain the high diversity and high frequency of X haplogroup lineages in the Galilee region. Low migration rates were also evident between the Galilee Druze and Druze from other subregions. The finding of the enrichment of the NRY haplogroup K among the Galilee Druze with no detection in samples from other subregions, further supports the relative isolation of this region, even among the Druze. Taken together these findings support the hypothesis that the Galilee Druze are a further more isolated subpopulation of the Druze, who in turn represents a refugium of the population genetic architecture of the Near East in antiquity.
Demographic modeling can also provide estimates of divergence times for populations with shared ancestries. The demographic modeling in the current study indicates most recent divergence of the Druze from an ancestral population shared with Egyptians, Ashkenazi Jews, Adygeis and Greeks (). The Egyptian shared ancestry is also consistent with Druze oral tradition. The migration rates of the Druze with these populations are exceedingly low, and this can be attributed to endogamy and geographic isolation following divergence. It should be kept in mind however, that the computational algorithms used for the demographic modeling are designed for models involving an ancestral population which split and maintained constant migration rate among the two daughter populations. Such a simplified model does not take into account the effect of shared party migration, and therefore would tend to overestimate the migration rate, and underestimate the divergence time
. Therefore the Druze would seem to have an even greater degree of genetic isolation, than indicated by these results of the demographic modeling.
The historical events and time frame for the loss or dilution of haplogroup X individuals is consistent with the population upheavals and patterns of migration that have characterized the Near East during the past two millennia at least 
. The preservation of this refugium of mtDNA lineage diversity among the Druze, mainly due to genetic isolation may be the result of their location in relatively more defensible mountainous regions, and the practice of conciliation with governing authorities and dissimulation called the “Taqiyya”
, or due to other factors which facilitated preservation of societal integrity during periods of demographic and political change in the region.
Our findings suggest that the Near East maternal genetic landscape differed substantially in the past from its current structure, and was enriched in diverse lineages of the mtDNA X haplogroup. These findings have been uncovered due to the unique demographic features of the Druze population, and the adjusted sampling method employed in the current study. The combination of a high frequency and diversity of the Druze mtDNA haplogroup X lineages, in a confined geographic region, and the low migration rate with nearby populations make it unlikely that this diversity was imported. It is thus likely that the global diversity of this haplogroup evolved in the Near East and adjacent regions of western Eurasia, during a long incubation period coinciding with and following the most recent out of Africa expansion as dated by mtDNA coalescence simulations
. The Druze population of the Galilee represents a contemporary refugium of this past genetic landscape.