ArticlePDF Available

Abstract and Figures

The present-day Sherpa are thought to descend from a small number of ancestors that set-tled in Nepal several centuries ago, coming from the Eastern Tibetan region of Kham. A generally ac-cepted ethnographic theory involves the out-of-Kham migration of four proto-clans between the 15th and 16th centuries. Traditional Sherpa society is still divided into clans, called ru, which are patrilin-early transmitted. Ru therefore roughly correspond to the surnames of Western societies: males of the same ru are expected to share identical Y-chromosome haplotypes. However, multiple origins of ru and/or frequent gene flow of male lineages from neighbouring populations can complicate the genea-logical structure. In the present work, 25 male Sherpas of the Solukhumbu district were typed for the 17 Y-chromosomal short tandem repeats included in the AmpFlSTR ® Yfiler™ kit. Seventeen different haplotypes were found; 12 were unique. A phylogenetic tree was then drawn from the pairwise muta-tional distance matrix with a neighbour-joining algorithm. Branching reliability was also assessed through bootstrap analysis. Two macro-clusters of haplotypes were found, ascribable on the whole to two out of four of the presumed Tibetan proto-ru, the Thimmi and the Minyagpa. However, the Min-yagpa macro-cluster was found to be bipartite in terms of haplogroups, being composed by two distinct haplotype clusters. Clustering of the contributors by birthplace was also performed, suggesting a dif-ferential ru spatial distribution between upper Khumbu and lower Solu. Khumbu seems predominantly populated by newer clans and putative descendants of the Thimmi proto-ru, whereas Solu is mostly inhabited by members of the Minyagpa proto-ru.
Content may be subject to copyright.
1© 2013 The Anthropological Society of Nippon
Vol. advpub(0), 000–000, 2013
Y-chromosome haplotypes and clan structure of the Sherpa of the
Solukhumbu (Nepal): preliminary ethnogenetic considerations
Andrea BOZZATO4, Gianumberto CARAVELLO1
1University of Padua, Department of Molecular Medicine—Public Health Unit, Via Loredan 18, 35131 Padova, Italy
2University of Padua, Deptartment of Molecular Medicine—Legal Medicine Unit, Via Giustiniani 2, 35131 Padova, Italy
3University of Pisa, Department of Physiological Sciences, SS 12 Abetone e Brennero 2, 56127 Pisa, Italy
4University of Padua, Department of Biology, Via Bassi 58/B, 35131 Padova, Italy
Received 14 October 2011; accepted 22 November 2012
Abstract The present-day Sherpa are thought to descend from a small number of ancestors that set-
tled in Nepal several centuries ago, coming from the Eastern Tibetan region of Kham. A generally ac-
cepted ethnographic theory involves the out-of-Kham migration of four proto-clans between the 15th
and 16th centuries. Traditional Sherpa society is still divided into clans, called ru, which are patrilin-
early transmitted. Ru therefore roughly correspond to the surnames of Western societies: males of the
same ru are expected to share identical Y-chromosome haplotypes. However, multiple origins of ru
and/or frequent gene flow of male lineages from neighbouring populations can complicate the genea-
logical structure. In the present work, 25 male Sherpas of the Solukhumbu district were typed for the
17 Y-chromosomal short tandem repeats included in the AmpFlSTR® Yfiler™ kit. Seventeen different
haplotypes were found; 12 were unique. A phylogenetic tree was then drawn from the pairwise muta-
tional distance matrix with a neighbour-joining algorithm. Branching reliability was also assessed
through bootstrap analysis. Two macro-clusters of haplotypes were found, ascribable on the whole to
two out of four of the presumed Tibetan proto-ru, the Thimmi and the Minyagpa. However, the Min-
yagpa macro-cluster was found to be bipartite in terms of haplogroups, being composed by two distinct
haplotype clusters. Clustering of the contributors by birthplace was also performed, suggesting a dif-
ferential ru spatial distribution between upper Khumbu and lower Solu. Khumbu seems predominantly
populated by newer clans and putative descendants of the Thimmi proto-ru, whereas Solu is mostly
inhabited by members of the Minyagpa proto-ru.
Key words: Sherpa, clan structure, Solukhumbu, Y-STRs, AmpFlSTR Yfiler
The Sherpa are a Tibeto-Burman ethnic group from the
Nepalese Himalayas, best known for their exceptional
mountaineering skills as porters and guides. From the genet-
ic point of view, Sherpa constitute a population isolate.
Likely causes of isolation (Arcos-Burgos and Muenke,
2002) are ethnic, cultural, linguistic, religious, and geo-
graphical confinement. Sherpa DNA has been studied in de-
tail in order to investigate the potential association between
genes and adaptation to high altitude (Malacrida et al., 2007;
Dromo et al., 2008). Sherpa villages and communities are
scattered over an extremely wide area, ranging from the
Helambu region, 72 km north-east of Kathmandu, to the
Darjeeling district, in West Bengal. The core of the Sherpa’s
earliest settlement area lies in the neighbouring regions of
Solu and Khumbu (Von Fürer-Haimendorf, 1964), located
in the north-eastern Nepalese administrative districts of
Solukhumbu and Okhaldhunga (Figure 1).
Traditional Sherpa society is divided into patrilineal
clans, although the number of clans is debated: it ranges
from 12 to 24 (Table 1), depending on the author (Von
Fürer-Haimendorf, 1964; Oppitz, 1968; Krämer and Sherpa,
2002; Wangmo, 2005). The Sherpa word for clan is ru,
which means ‘bones,’ since Sherpas actually believe that
bones pass from father to son, thus establishing the natural
propensities of a man. This pseudo-genetic paternal inherit-
ance was considered superior to that of the mother, whose
contribution is only transmitting the flesh, or sha (Wangmo,
2005). Ru are neither castes nor social classes, but rather
wide familial clans with common roots in a certain village or
in a common, and often legendary, ancestor (Fantin, 1973;
Childs, 1997). After all, as in every human society, the addi-
tion of a heritable element facilitates identification and also
marks lineages, providing a label of membership (King and
Jobling, 2009). Sherpa can overall be considered as an endo-
gamic ethnic group, yet exogamy between ru is compulsory:
marrying within the same ru is forbidden and considered in-
cestuous (Presciuttini et al., 2010).
Advance Publication
Brief Communication
* Correspondence to: Irene Amoruso, University of Padua, Depart-
ment of Molecular Medicine—Public Health Unit, Via Loredan 18,
35131 Padova, Italy.
Published online 7 February 2013
in J-STAGE ( DOI: 10.1537/ase.121122
Sherpa tradition tells that all of the present Sherpa ru de-
scend from some ancient Tibetan clans that came to Nepal
from Eastern Tibet about 500 years ago. Many alternative
stories about the very first arrival of the Sherpa to the
Nepalese valleys exist, and most of these should probably be
regarded as mere legends (Brower, 1991). There is not even
agreement about the route of migration from Tibet to
Solukhumbu, nor the exact Nepal entrance point (Von Fürer-
Haimendorf, 1964).
Nevertheless, most Sherpa tales trace back the coming of
their ancestors from the Salmo Gang (Zal mo sgang) area of
Kham, in Eastern Tibet, during the 15th–16th centuries
(Brower, 1991). The Salmo Gang area is presently part of
the modern Ganzi Tibetan Autonomous Prefecture of the
Chinese Sichuan Province. Significantly, the word sherpa
itself is composed of two Tibetan words (shar, east and pa,
person), together meaning ‘Easterners,’ and could thus be
reflecting this migration (Stevens, 1993). The pioneering
work of Oppitz (1968), derived from the analysis of a few
written clan ancestry records retrieved in Solu, provides
firmer historical evidence for the Kham homeland hypothe-
sis, yet still has to face the dubious historical authenticity of
the documents (Wangmo, 2005). On the whole, authors usu-
ally agree the broad terms of Oppitz’s reconstruction (Von
Fürer-Haimendorf, 1964; Brower, 1991; Krämer and
Sherpa, 2002; Wangmo, 2005), which involves the out-of-
Kham migration of four proto-ru at the turn of the 15th to the
16th century (Oppitz, 1968).
The four proto-ru were the Lama-serwa (or simply
Lama), the Chakpa (or Chiawa), the Minyagpa, and the
Thimmi. Probably, each proto-ru gave birth to several sub-
clans through time, in accordance with the Sherpa assertion
of their Tibetan ancestry. Indeed, Sherpa regard some of the
present ru as brother-clans: this is probably due to their com-
mon descent from the same proto-ru. The supposed genealo-
gy of present Sherpa ru and their relationship with the four
proto-clans can be summarized based on a comparison of the
works of the four main works (Von Fürer-Haimendorf,
1964; Oppitz, 1968; Krämer and Sherpa, 2002; Wangmo,
2005) that have already dealt with this issue (Table 1). An
exact genealogy of modern ru cannot be traced for certain,
since written sources are scarce and incomplete and oral tra-
dition sometimes reports contrasting theories. Unfortunate-
ly, the degree of complexity has been increased over the
years by the later arrival of other groups of Tibetan immi-
grants, particularly from the bordering Tingri area, after the
first migration. The members of these more recent groups
are also organized in patrilinear ru and are now considered
as pure Khumbu Sherpa, even if none of them came from
Kham nor had any kind of written tradition concerning its
genealogy (Sherpa, 2007).
Except for its two pseudoautosomal regions found at the
telomeric ends, the Y chromosome is exclusively inherited
from father to son as a entire haplotype of physically linked
loci. The portion of the chromosome which does not
undergo homologous recombination is called the non-
Figure 1. Sherpa main settlement area coincides with Solukhumbu and Okhaldunga Administrative Districts, in North-eastern Nepal.
(Adapted from a GIS map by Cartography and Survey Engineering Consultant Kabindra Joshi, available at
recombining region of the Y (NRY) and comprises male-
specific genes only, known as holandric genes. The recent
characterization of many new Y-chromosome markers has
contributed to increasing their usefulness for studies of ge-
netic anthropology, which, until recently, lagged far behind
mitochondrial DNA and maternal lineage studies (Crawford,
2007). DNA polymorphisms of the NRY have already
proved to be a useful tool for tracing ancient patrilinear lin-
eages (Jobling and Tyler-Smith, 2003; Kwak et al., 2005;
King and Jobling, 2009): all patrilinear relatives share the
same NRY haplotype and thus inheritance, with absence of
recombination, leads to the perpetration of polymorphisms
through a simple paternal transmission pathway (Ljubković
et al., 2008). The most widely used kind of NRY polymor-
phisms for phylogeny of recently diverged paternal lineages
are the short tandem repeats (STRs). Y-STRs are highly
polymorphic markers and thanks to a high level of variabili-
ty due to variation in repeat number they can reveal further
variation within haplogroups (Kwak et al., 2005). Isony-
mous individuals are expected to carry the same Y haplo-
type: observed incongruity can be ascribed to new
mutational events, adoptions, illegitimacy, or polyphyletic
origin of the surname.
In this work, the 17 AmpFlSTR Yfiler polymorphic loci
(Mulero et al., 2006) were amplified in a sample of 25 male
Solukhumbu Sherpa. Haplogroups were also predicted and
the emerging genetic tree was compared with ethnogenea-
logical hypotheses.
In other words, the present research is aimed at a prelimi-
nary definition of the true kinship and clan structure of the
Sherpa population. The approach followed is that of estab-
lishing any existing correspondence between the Y-STR
haplotypes and traditional patrilineal Sherpa clans.
Materials and Methods
Buccal swab samples were collected from 25 male volun-
teers with Whatman® OmniSwabs. Each ejectable head was
put in a 1.5 ml sterile extraction tube. The sampling proce-
dure was approved by the Social and Ethics Committee of
the Department of Environmental Medicine and Public
Health (EMPH), Padua University.
16 samples were collected in spring 2008, during a meet-
ing of the Sherpa community held in Namche Bazaar, the
main village of Khumbu Valley, and 9 samples were collect-
Table 1. Lists of Sherpa ru and their presumed genealogy
Ref. von Führer-Haimendorf
Krämer and Sherpa
No. of listed ru 22 24 12 20
Proto-clan Minyakpa Minyakpa Minyakpa Minyagpa
Subclans Gole Gole Golela Goleg
Pinasa Binasa Binasa Benasa
Thaktu Trakto Takto Tragtho
Gardza Gardza Gartsa
Pankarma Pankarma Pankarma
Shire Shire Shire
Yulgongma Yulgongma Yulgongma
Proto-clan Thimmi Thimmi Thimi Thimmipa
Subclans Salaka Salaka Salaka Zalaka
Paldorje Paldorje Paldorje Paldorje
Goparma Gobarma Gobarma
Khambadze Khambadze Khampaje Khampache
Lakshindu Lakshindo Lakshindo Labushingtog
Proto-clan Lama Serwa Lamaserwa Lama Serwa
Subclans — Lama
Proto-clan Chiawa Chakpa Chakpa Chagpa
Subclans ————
Recent clans (no genealogy available)
Chuserwa Chuserwa Chuserwa
Jongdomba Jungdomba Jongdongpa
Lhukpa Lhukpa — Lhugpa
Mende Mende — Mendewa
Munming Murmin Murmin Tso
Nawa Nawa — Nawa
Shangup Shangup Shangkhug
Sherwa Sherwa — Sherwa
The table lists the genealogical structure of present Sherpa ru. The four proto-clans (Minyagpa, Thimmi, Lamaserwa, and Chakpa)
are followed by their respective subclans. Genealogies proposed by the four main works that have already dealt with the Sherpa ru
history are given for comparison.
ed in spring 2009 in the villages of Namche, Kunde, and
Khumjung. All individuals freely donated their saliva, after
signing an informed consent form that explained the purpose
of DNA analysis in the present research. The volunteers
were questioned about their ru affiliation, age, birthplace,
and dwelling place. All of the contributors reported as being
unrelated to one another and those sharing the same ru were
actually unaware of any kind of biological relationship be-
tween them.
Samples were packaged and transported to Padua (Italy)
within 7 days and processed in the Forensic Genetics Labo-
ratory of the Department of Environmental Medicine and
Public Health (Legal Medicine Unit), Padua University. The
laboratory achieved quality control certification by the Y
chromosome Haplotype Reference Database (YHRD) on 23
November 2000 (YHRD c.n. YC000052).
DNA extraction
DNA extraction from buccal swabs was performed using
the QIAGEN QIAmp® DNA Microkit according to the man-
ufacturer’s instructions (Qiagen, 2010). About 30–50 µg of
DNA were recovered in 30 µl of final solution.
Multiplex PCR
Multiplex polymerase chain reaction (PCR) was per-
formed on 0.5–1.0 ng of target DNA for each sample using
the AmpFlSTR® YfilerTM PCR Amplification Kit, in accor-
dance with the protocols described in its User’s Manual
(Applied Biosystems, 2006). The AmpFlSTR® YfilerTM Kit
amplifies 17 Y-STR loci: they consist of the 9-marker Euro-
pean minimal haplotype (minHt) (DYS19, DYS385a/b,
DYS389I/II, DYS390, DYS391, DYS392, and DYS393),
the 11-marker Scientific Working Group DNA Analysis
Methods (SWGDAM) recommended Y-STR panel (minHt
plus DYS438 and DYS439) and the additional highly poly-
morphic loci DYS437, DYS448, DYS456, DYS458,
DYS635, and Y-GATA-H4.1. Allele nomenclature follows
ISFG recommendations (Gusmao et al., 2006). The locus
DYS385 was excluded from further analysis, as its two twin
loci show indistinguishable alleles.
PCR reaction was performed in a GeneAmp® PCR Sys-
tem 9700 following the manufacturer’s instructions (Ap-
plied Biosystems, 2010) except for minimal modifications
of the final reaction volume, which was 12.5 µl (9.5 µl of
master mix +3.0 µl of DNA solution) instead of 25.0 µl. In
addition, six supplementary amplification cycles were per-
formed for some samples. PCR thermocycling parameters
were the following: initial hot-start incubation step at 95°C
for 11 min, 30 amplification cycles (denaturing at 94°C for
1 min, annealing at 61°C for 1 min, and extension at 72°C
for 1 min) and final extension at 60°C for 80 min.
Amplified STRs were separated by capillary electro-
phoresis using an ABI Prism® 3130 Genetic Analyzer (Ap-
plied Biosystems). GeneScan-500 Internal Lane Size
Standard LIZ-500 was the internal standard of choice. The
size of the PCR products was classified with GeneMapper
v.2 software, with the Yfiler Allelic Ladder (Applied Bio-
systems) as comparison.
Data analysis
The genetic distance for each pair of haplotypes was cal-
culated as the sum, over all loci, of the absolute difference in
the number of STRs. Genetic distances were then reported in
a pairwise distance matrix. Further data analysis was
performed with Arlequin v. software (Excoffier and
Lischer, 2010). Allelic frequencies, allelic ranges, haplotype
diversity (Nei, 1987), and gene diversity per locus (h) were
also calculated. A phylogenetic tree was extrapolated from
the pairwise distance matrix, with a neighbour-joining (NJ)
algorithm. Statistical reliability of branching nodes was as-
sessed through bootstrap (Felsenstein, 1985), performing
100 repetitions. The tree was finally drawn with the TreeGen
tool (CBRG, 2012). Multi-copy marker DYS385 was ex-
cluded from the phylogenetic analysis, as its two loci,
DYS385a and DYS385b, were indistinguishable from each
other and thus the result from these was not informative.
After haplotype typology, haplogroup was predicted for
each sample with the free Haplogroup Predictor tool v. 21
(Athey, 2007). The software identifies the most probable
haplogroup, assigning a prediction confidence score ex-
pressed as a percentage. Haplogroups obtaining the highest
score were chosen and will subsequently be reported.
Results and Discussion
First, statistics of general genetic interest for the Sherpa
population were calculated. 17 different haplotypes were
found out of 25 samples: 5 of these were shared and 12 were
unique. The allelic profiles of the haplotypes are listed in full
in Table 2: neither null alleles nor duplicated loci were
found. DYS448 showed the 20.2 microvariant in samples
M1 and M12. Allelic ranges and frequencies, along with
gene diversity, are also schematically reported (Table 3).
Prediction of haplogroups identified three main groups
(Table 2): 9 individuals belong to haplogroup Q (M10, M11,
M16, M17, M18, M20, M21, M24, M29), 6 to haplogroup
E1b, subclade E1b1 (M4, M6, M13, M23, M26, M27), 9 to
haplogroup R1 (M1, M3, M12, M14, M15, M19, M25 are
R1a; M8, M28 are R1b; M19 had an equal score between
R1a and R1b). Individual M22 alone belongs to haplogroup
H. Probability scores for each predicted haplogroup are re-
ported in Table 2 as well. Overall haplotype diversity for the
Sherpa population assumes a value of 0.9633 ±0.0208.
Haplotype diversities previously reported for a mixed Ne-
palese and two Tibetan population samples are respectively
0.9970 (Parkin et al., 2007), 0.9998 (Tian-Xiao et al., 2009)
and 0.9981 (Gayden et al., 2010). Since haplotype diversity
of Sherpas is not so different from neighbouring popula-
tions, it can be supposed that their ancestors had substantial
diversity in their male lineages.
The NJ tree (Figure 2) represents the presumable kinship
of the 25 typed Sherpa. In addition, the tree reports the stated
ru kinship of each contributor, the proto-ru and the predicted
haplogroup. Unfortunately, contributors M1 and M8 simply
asserted themselves to be ‘Sherpa’ and of ‘Tibetan origin’,
respectively. The topology of the phylogenetic tree clearly
presents two main macro-clusters of haplotypes, branched
with a bootstrap value of 100 (Figure 2). The first macro-
cluster results further partitioned into two minor haplotype
Table 2. The 17 amplified haplotypes and their extended profiles
Table 3. Allelic frequencies, allelic range, gene diversity per locus and G-W index at the 17 Y-STR loci
Haplotype 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Sample id M1 M3 M4 M6 M8 M10 M11 M13 M14 M16 M17 M19 M21 M22 M27 M28 M29
M12 M23 M15 M20 M18
M26 M25 M24
Absolute frequency21311111323 1 11111
Relative frequency 0.08 0.04 0.12 0.04 0.04 0.04 0.04 0.04 0.12 0.08 0.12 0.04 0.04 0.04 0.04 0.04 0.04
DYS19 1515141415141414151414 16 1416141514
DYS389-I1313101014121210131212 14 1212101312
DYS389-II 29 29 27 27 30 28 28 27 29 28 28 30 30 29 28 27 28
DYS390 23 25 23 23 23 23 23 23 25 23 23 23 23 27 23 23 23
DYS391 10 11 10 10 10 10 10 10 10 10 10 10 10 10 10 11 10
DYS392 11 7 11 11 7 14 14 11 7 14 14 7 14 10 11 7 14
DYS393 13 13 14 14 13 13 12 14 13 13 12 13 12 12 14 13 13
DYS437 14 14 14 14 14 16 15 14 14 16 14 14 15 14 14 14 15
DYS 438 11 11 10 10 11 11 11 10 11 11 11 11 11 10 10 11 11
DYS439 12 12 11 12 12 13 12 12 12 12 12 12 12 11 11 12 12
DYS448 20.219191918212019192120 18 2118191920
DYS456 16 16 14 14 17 15 15 13 16 15 15 16 15 15 14 16 15
DYS458 16 16 15 15 17 17 18 15 16 17 18 17 17 15 15 16 18
DYS635 21 21 21 21 21 21 20 20 21 21 20 21 20 20 21 21 20
YGATAH41111111111121111111212 11 1211111112
DYS385 13–15 11–11 15–16 15–16 11–11 14–18 13–19 15–16 11–11 14–18 13–19 11–12 18–19 17–17 15–16 11–11 13–19
Haplogroup R1a R1a E1b1 E1b1 R1b Q Q E1b1 R1a Q Q R1a/R1b Q H E1b1 R1b Q
Prediction % score 42.0 98.9 84.8 96.7 90.3 94.8 95.9 99.3 98.4 94.0 52.8 32.0/30.0 99.2 99.3 98.0 91.1 92.7
DYS 19 389-I 389-II 390 391 392 393 437 438 439 448 456 458 635 GATAH4.1 385 Mean
No. of alleles 3 4 4 3 2 4 3 3 2 3 4 5 4 2 2 8 3.5
Allelic range 2 4 3 4 1 7 2 2 1 2 3 4 3 1 1 8 3.1
Gene diversity h0.553 0.727 0.750 0.347 0.153 0.717 0.640 0.410 0.420 0.397 0.717 0.723 0.777 0.453 0.453 0.853 0.568
clusters, named Cluster I and II, that are mostly composed
by individuals affiliated to ru all descending from the Min-
yagpa proto-clan (Thaktok, 5 individuals; Pinasa, 2 individ-
uals; Gardza, 2 individuals; Gole, 2 individuals; Chuserwa,
1 individual). The Minyagpa proto-clan could therefore have
comprised ancestors with two different Y-chromosome lin-
eages, consistently with the two different haplogroups pre-
dicted, Q and R1, that respectively characterize Cluster I and
Cluster II. Moreover, the Thaktok ru, being common to both
Clusters I and II, is indeed still heterogeneous in terms of Y
haplotypes and haplogroups. A similar polyphyletic origin
for the Minyagpa clan is thus quite plausible. On the other
hand, individual M29 falls as outsiders in Cluster I, since he
belongs to the Chakpa lineage and not to the Minyagpa one.
Similarly, the Tibetan contributor M8 and the Lama ru mem-
ber M19, are part of Cluster II but they do not belong to the
Minyagpa proto-ru. In fact, Lama and Chakpa ru were ex-
pected to form two distinct clusters, since they are numbered
among the four original Tibetan proto-ru and are described
as unsplit clans. Moreover, M25, a Goparma from the Thim-
mi proto-clan, unexpectedly figures in Cluster II. In order to
explain these discrepancies, illegitimacies, adoptions, and
personal data uncertainty should be considered.
The second macro-cluster, named Cluster III, comprises
members of the Salaka (2 individuals) and Paldorje (1 indi-
vidual) ru: both clans belong to the Thimmi proto-ru group
(Table 1). Cluster III also comprises 3 individuals belonging
to the Chuserwa and Tsangup ru, that Oppitz (1968) and
Wangmo (2005) regard as recent clans and unrelated to the
four Kham proto-ru. Nevertheless, the genetic evidence
could be revealing an unreported kinship of these two ru
with the Thimmi proto-clan. Moreover, Cluster III is mainly
composed by haplotypes belonging to the E1b1 haplogroup.
Cluster III could be then regarded as the putative Thimmi
proto-ru tree cluster. In addition, assuming the authenticity
of the reconstructed genealogies (Oppitz, 1968), several
marriages between members of the Minyagpa, Chakpa, and
Thimmi clans seem to have taken place during the 15th cen-
tury (Figure 3). As discussed above, kinship incongruities
observed in some NJ tree branches might thus be linked to
events of haplotype interchange that occurred among the
families of the Sherpa ancestors.
In fact, there is only one clustering outlier represented by
M22, a Pinasa Sherpa, thus coming from proto-ru Minyag-
pa, with the unique property of belonging to the H haplo-
Samples were further grouped by considering the birth-
places of contributors. Home villages were located on a map
(Figure 1) and assigned either to the Khumbu Valley (North
cluster) or to the Solu and Okhaldunga area (South cluster)
(Table 4). This kind of approach was undertaken as a pre-
liminary evaluation of two contrasting hypothesis, the
Figure 2. Neighbour-joining tree (NJ) for the Sherpa population sample. For each of the 25 Sherpa contributors, self-stated ru affiliation has
been listed along with the respective ancestral proto-clan. Abbreviations for the four proto-ru used in the phenogram are: Lama (L), Chakpa (C),
Minyagpa (M), and Thimmi (T). Lower case ‘n’ stands for ‘new’ and represents recent Sherpa clans that are supposed to have reached
Solukhumbu after the pristine out-of-Kham migration. The three main cluster of haplotypes are named I, II and III in the figure. Contributor M1
unfortunately simply claimed to be Sherpa but could not provide his Ru, which in the figure is replaced by a question mark “?.” Moreover, M8
stated to be of Tibetan origin and therefore he had no Sherpa Ru kinship: in the figure this is represented by a slash “/.”
‘Khumbu first’ and the ‘Solu first’ theories, that describe the
history of Sherpa settlement in the Solukhumbu area. Brief-
ly, authors supporting the ‘Khumbu first’ theory agree that
Sherpa entered Nepal from a high mountain pass, notably
the Nangpa La, and immediately settled Khumbu. In partic-
ular, they state that the Thimmi proto-clan originally settled
in the Bhote Kosi valley, in Western Khumbu (Oppitz, 1968;
Krämer and Sherpa, 2002), while Minyagpa families at first
settled in the Phortse village area, also in Khumbu, but later
on moved to the Solu lowlands (Oppitz, 1968; Krämer and
Sherpa, 2002). Moreover, Lama and Chakpa people suppos-
edly went straight to Solu and directly settled there (Krämer
and Sherpa, 2002).
On the other hand, the ‘Solu first’ theory suggests that the
settlement of the Sherpa in Nepal did not even begin in
Khumbu. In fact, the higher Khumbu valley was probably
settled only after Solu, since at the time of their arrival pro-
longed periods of frost still sequestered its cultivable lands.
Frost periods faded only with the ending of Little Ice Age,
due to major climate changes (Muehlich, 1998). Moreover,
in Khumbu only ‘mixed’ Sherpa settlements are found,
whereas in Solu true Sherpa-clan territories and villages can
be identified, even nowadays (Muehlich, 1996). This fact
again strongly supports that the first Sherpa settlement area
was indeed Solu.
Indirectly, the statement of Krämer and Sherpa (2002)
about Lama and Chakpa families heading straight to Solu
supports the ‘Solu first’ hypothesis as well. Genealogies
provided by Oppitz (1968) too suggest that upper Khumbu
was more recently and primarily settled by newer ru, as well
as by Tibetans coming from the bordering Tingri region.
Newcomers were then accepted and integrated by Sherpa al-
ready resident in their society (Muehlich, 1998).
Interpretation of genetic evidence and ru spatial distribu-
tion seems favourable to elements of both theories: it sug-
gests that all but two (71.4%) of the Thimmi cluster
members, which also comprises most of the new ru, were
born in upper Khumbu and eight out of nine (88.9%) mem-
bers of the major Minyagpa cluster were born in the South-
ern Solu and Okhaldhunga areas (Table 4). An ancient
proto-clan settlement area could thus be roughly identified
both for the Thimmi (Khumbu Valley) and for at least one
family branch of the Minyagpa (Solu).
In summary, three main clusters of haplotypes were
found, ascribable on the whole to two of the presumed proto-
clans, the Thimmi and the Minyagpa Tibetan ru. Chakpa and
Lama appeared as sub-clans of the Minyagpa rather then
self-standing families. Ethnogenetic findings about the clan
structure of the Sherpa, do not in fact contradict the reported
theories and also confirm an interesting association between
Y-haplotypes and social kinship. The oral Sherpa tradition
indeed gave birth to a family identification system similar to
the one adopted by societies with a formalized patrilineal
transmission of surnames. A major weakness of the present
preliminary research is the small sample size. As it stands,
members of every ru were not typed and consequently some
clans are still missing from the tree. However, the prelimi-
nary results strongly encourage the ongoing pursuit of eth-
nogenetic research on Sherpa history: it is likely that Sherpa
true clan genealogy and history will be fully elucidated by
future sampling campaigns, aimed at enlarging the sample
size, and in loco investigations. Moreover, among our future
intentions stands out the achievement of a distinct sampling
Figure 3. The Chakpa genealogy: an example of intermarriages
among the four proto-ru of the Sherpa. The picture reproduces part of a
wider and more complex tree of the Chakpa ru genealogy drawn by
Oppitz based on the translation of some Sherpa clan records (Oppitz,
1968). From the tree it can be inferred that various marriages between
members of the four proto-clans took place during the 15th century.
Table 4. Birth-village, ru and proto-ru
Sample id Birthplace Ru Proto-ru MST Cluster
Khumbu area
M4 Nangma Salaka T I (71.4%)
M23 Namche Chuserwa n
M26 Thame Tsangup n
M27 Chukkhung Chuserwa n
M6 Shigatze Paldorje T
M21 Khumjung Chuserwa n II (11.1%)
M1 Namche ? ? III (71.4%)
M3 Namche Thaktok M
M28 Namche Thaktok M
M12 Namche Lhukpa n
M25 Kunde Goparma T
Solu and Okhaldunga area
M13 Okhaldhunga Salaka T I (28.6%)
M22 Okhaldhunga Pinasa M
M20 Goli Thaktok M II (88.9%)
M11 Kerung Pinasa M
M24 Okhaldhunga Garja M
M16 Patle Thaktok M
M17 Patle Thaktok M
M18 Patle Pinasa M
M29 Solu Chawa C
M10 Tapting Garja M
M14 Gepchuka Gole M III (42.8%)
M15 Gepchuka Gole M
M19 Beni Lama L
Samples are grouped both by geographical provenance (Khumbu vs.
Solu and Okhaldunga area) and by NJ cluster affiliation (clusters I, II,
and III). The last column reports percentages of cluster members born
in the corresponding geographical area.
campaign in some Sichuan prefectures, where the legendary
Sherpa homeland should lie, in order to assess potential
genetic similarity between the Solukhumbu Sherpa and the
autochthonous population. Promisingly, in the present
Dartsedo (Kangding) County, in the balkanized area of the
so-called Western Sichuan Ethnic Corridor (Sun, 1990), a
small ethnic group called Minyag is still dwelling (Ikeda,
1998). Some authors sustain that the Minyag people of
Sichuan could be the descendants of some dynasty that from
1038 to 1227 CE ruled a North Minyag kingdom northeast
of Lake Kokonor, also identified as part of Xi-Xia or Tangut
Kingdom, in what later became the Tibetan province of
Amdo (Stein, 1951; Balikci, 2008). Sichuan Minyag possi-
bly migrated to their current dwelling area after the Mongols
conquered the former Minyag kingdom in 1227.
Surprisingly, it is curious to find that the Namgyal dynas-
ty of the Sikkim kings, the Chogyals, also claims to descend
from a Minyag clan member, named Guru Tashi. Oral histo-
ries report that, during the 13th century, Guru Tashi left his
homeland in Kham Minyag and, via Lhasa, eventually
crossed a Himalayan pass and reached the Chumbi Valley in
Sikkim. This strong analogy with Sherpa oral traditions
about their migration route along with the quite common re-
currence of the Minyag appellation in apparently unrelated
ethnic groups, definitely represent an interesting research
trail that should be further investigated.
In conclusion, in the firm belief that in the end only a mul-
tidisciplinary research approach can yield an exhaustive ex-
planation of the true origin of the Solukhumbu Sherpa, we
plan to draw together a team capable of juxtaposing clues
from linguistics, social sciences, and ethnoclimatology to
the genetic data.
Revealing ethnohistory through the juxtaposed investiga-
tion of manuscripts, oral tradition, and genetic pedigrees is a
captivating and engrossing path: the authors wish to thank
everybody who, like them, became fascinated with the
Solukhumbu Sherpa tradition and especially all the Sherpa
volunteers that contributed in person to the present research.
The authors are also grateful to the University of Padua for a
grant (no. CPDR100470).
Applied Biosystems (2006) AmpFlSTR® YfilerTM PCR Amplifica-
tion Kit User’s Manual.
(accessed 3 December 2010)
Applied Biosystems (2010) GeneAmp® PCR System 9700 User’s
Manual. (accessed 3
December 2010)
Arcos-Burgos M. and Muenke M. (2002) Genetics of population
isolates. Clinical Genetics, 61: 233–247.
Athey W. (2007) Haplogroup Predictor ver. 21: Y Haplogroup
Prediction from Y-STR Values.
hapest5/index.html (accessed 19 April 2012)
Balikci A. (2008) Lamas, shamans and ancestors: village religion
in Sikkim. Koninklijke Brill NV, Leiden, the Netherlands.
Brower B. (1991) Sherpa of Khumbu: people, livestock and land-
scape. Oxford University Press, Delhi.
CBRG (2012) TreeGen: Tree generation tool from distance data.
Computational Biochemistry Research Group (CBRG), Swiss
Federal Institute of Technology (ETH), Zurich. www.cbrg. (accessed 8 October 2012)
Childs G. (1997) A note on the Tibetan Origin of the Sherpa Serwa
lineage. Himalayan Research Bulletin, 17: 23–25.
Crawford M.H. (2007) Anthropological Genetics: Theory, Meth-
ods and Applications. Cambridge University Press, Cam-
Dromo Y., Ota M., Hanaoka M., Katsuyama Y., Basnyat B.,
Neupane P., Arjyal A., Pandit A., Sharma D., Ito M., and
Kubo K. (2008) Two hypoxia sensor genes and their associa-
tion with symptoms of acute mountain sickness in Sherpas.
Aviation Space and Environmental Medicine, 79: 1056–1060.
Excoffier L. and Lischer H.E.L. (2010) Arlequin suite ver 3.5: a
new series of programs to perform population genetics analy-
ses under Linux and Windows. Molecular Ecology
Resources, 10: 564–567.
Fantin M. (1973) Sherpa Himalaya Nepal. Arti grafiche Tamari,
Felsenstein J. (1985) Confidence limits on phylogenies: an
approach using the bootstrap. Evolution, 39: 783–791.
Gayden T., Chennakrishnaiah S., La Salvia J., Jimenez S.,
Regueiro M., Maloney T., Persad P.J., Bukhari A., Perez A.,
Stojkovic O., and Herrera R.J. (2010) Y-STR diversity in the
Himalayas. International Journal of Legal Medicine, 124:
Gusmao L., Butler J.M., Carracedo A., Gill P., Kayser M., Mayr
W.R., Morling N., Prinz N., Roewer L., Tyler-Smith C., and
Schneider P.M. (2006) DNA Commission of the International
Society of Forensic Genetics (ISFG): an update of the recom-
mendations on the use of Y-STRs in forensic analysis. Foren-
sic Science International, 157: 187–197.
Ikeda T. (1998) Some phonological features of modern Munya
(Minyak) language. Nairiku Ajia Gengo no Kenkyuu, 13: 83–
91 (in Chinese).
Jobling M.A. and Tyler-Smith C. (2003) The human Y chromo-
some: an evolutionary marker comes of age. Nature Reviews—
Genetics, 4: 598–612.
King T.E. and Jobling M.A. (2009) What’s in a name? Y chromo-
somes, surnames and the genetic genealogy revolution.
Trends in Genetics, 25: 351–360.
Krämer K.H. and Sherpa L.D. (2002) Die Sherpa von Solu-
Khumbu. Nepal Information, 89: 12–13.
Kwak K.D., Jin H.J., Shin D.J., Kim J.M., Roewer L., Krawczak
M., Tyler-Smith C., and Kim W. (2005) Y-chromosomal STR
haplotypes and their applications to forensic and population
studies in east Asia. International Journal of Legal Medicine,
119: 195–201.
Ljubković J., Stipišić A., Sutlović D., Definis-Gojanović M.,
Bučan K., and Anđelinović Š. (2008) Y-chromosomal short
tandem repeat haplotypes in southern Croatian male popula-
tion defined by 17 Loci. Croatian Medical Journal, 49: 201–
Malacrida S., Katsuyama Y., Droma Y., Basnyat B., Angelini C.,
Ota M., and Danieli G.A. (2007) Association between human
polymorphic DNA markers and hypoxia adaptation in Sherpa
detected by a preliminary genome scan. Annals of Human
Genetics, 71: 630–663.
Muehlich M. (1996) Traditionelle Opposition: Individualitaet und
Weltbuild der Sherpa. Pfaffenweiler.
Muehlich M. (1998) Rejoinder and response to ‘Solukhumbu and
the Sherpa’ on Himalayan Research Bulletin 17:2. Himalayan
Research Bulletin, 18:1.
Mulero J.J., Chang C.W., Calandro L.M., Green R.L., Li Y.,
Johnson C.L., and Hennessy L.K. (2006) Development and
validation of the AmpFlSTR® YfilerTM PCR Amplification
Kit: a male specific, single amplification 17 Y-STR multiplex
system. Journal of Forensic Sciences, 51: 64–75.
Nei M. (1987) Molecular Evolutionary Genetics. Columbia Uni-
versity Press, New York.
Oppitz M. (1968) Geschichte und Sozialordnung der Sherpa. Uni-
versitäts Wagner, Innsbruck-München.
Parkin E.J., Kraayenbrink T., Opgenort J.R., Van Driem G.L.,
Tuladhar N.M., De Knijff P., and Jobling M.A. (2007) Diver-
sity of 26-locus Y-STR haplotypes in a Nepalese population
sample: isolation and drift in the Himalayas. Forensic Science
International, 166: 176–181.
Presciuttini S., Caenazzo L., Amoruso I., and Caravello G.U.
(2010) Ru affiliation in the Sherpa of the Khumbu Valley:
ethnographic survey and preliminar analysis of Y-
chromosomes haplotypes. In: Jha P.K. (ed.), Contemporary
Research in Sagarmatha (Mt. Everest) Region. Nepal Acad-
emy of Science and Technology, Lalitpur.
Qiagen (2010) QIAamp® DNA Micro Handbook. http://www. (accessed 12 January 2011)
Sherpa A.T. (2007) Thesis on the origin of Sherpa peopole.
Stein R.A. (1951) Mi-ñag et Si-hia. Géographie historique et
légendes ancestrales. Bulletin de l’Ecole française d’Extrême-
Orient, 44: 223–265.
Stevens S.F. (1993) Claiming the High Ground: Sherpas, Subsis-
tence, and Environmental Change in the Highest Himalaya.
University of California Press, Berkeley.
Sun H. (1990) Languages of the ethnic corridor in Western
Sichuan. Linguistics of the Tibeto-Burman Area, 13: 1–31.
Tian-Xiao Z., Li Y., and Sheng-Bin L. (2009) Y-STR haplotypes
and the genetic structure from eight Chinese ethnic popula-
tions. Legal Medicine, 11: S198–S200.
Von Fürer-Haimendorf C. (1964) The Sherpas of Nepal: Buddhist
Highlanders. Murray, London.
Wangmo J. (2005) The Lawudo Lama. Wisdom Publications,
Boston, MA.
... Notably, genetic researches focused on the Sherpas were mainly performed in the fields of archaeology [11,12], molecular anthropology [7][8][9], medical genetics [3,5], and genetic genealogy [2,4], only a few forensic-related studies focused on the Nepal Sherpas were conducted based on the low-density short tandem repeats (STRs) [13,14]. Insertion and deletion polymorphisms (InDels), the second most abundant polymorphism across the human genome with low mutation rates and small amplicon lengths [15], combining the desirable features of both SNPs and STRs. ...
Sherpa people, one of the high-altitude hypoxic adaptive populations, mainly reside in Nepal and the southern Tibet Autonomous Region. The genetic origin and detailed evolutionary profiles of Sherpas remain to be further explored and comprehensively characterized. Here we analyzed the newly-generated InDel genotype data from 628 Dingjie Sherpas by merging with 4222 worldwide InDel profiles and collected genome-wide SNP data (approximately 600 K SNPs) from 1612 individuals in 191 modern and ancient populations to explore and reconstruct the fine-scale genetic structure of Sherpas and their relationships with nearby modern and ancient East Asians based on the shared alleles and haplotypes. The forensic parameters of 57 autosomal InDels (A-InDels) included in our used new-generation InDel amplification system showed that this focused InDel panel is informative and polymorphic in Dingjie Sherpas, suggesting that it can be used as the supplementary tool for forensic personal identification and parentage testing in Dingjie Sherpas. Descriptive findings from the PCA, ADMIXTURE, and TreeMix-based phylogenies suggested that studied Nepal Sherpas showed excess allele sharing with neighboring Tibeto-Burman Tibetans. Furthermore, patterns of allele sharing in f-statistics demonstrated that Nepal Sherpas had a different evolutionary history compared with their neighbors from Nepal (Newar and Gurung) but showed genetic similarity with 2700-year-old Chokhopani and modern Tibet Tibetans. QpAdm/qpGraph-based admixture sources and models further showed that Sherpas, core Tibetans, and Chokhopani formed one clade, which could be fitted as having the main ancestry from late Neolithic Qijia millet farmers and other deep ancestries from early Asians. Chromosome painting profiles and shared IBD fragments inferred from fineSTRUCTURE and ChromoPainter not only confirmed the abovementioned genomic affinity patterns but also revealed the fine-scale genetic microstructures among Sino-Tibetan speakers. Finally, natural-selection signals revealed via iHS, nSL and iHH12 showed natural selection signatures associated with disease susceptibility in Sherpas. Generally, we provided the comprehensive landscape of admixture and evolutionary history of Sherpa people based on the shared alleles and haplotypes from the InDel-based genotype data and high-density genome-wide SNP data. The more detailed genetic landscape of Sherpa people should be further confirmed and characterized via ancient genomes or single-molecule real-time sequencing technology.
Anthropological genetics is a field that has been in existence since the 1960s and has been growing within medical schools and academic departments, such as anthropology and human biology, ever since. With the recent developments in DNA and computer technologies, the field of anthropological genetics has been redefined. This volume deals with the molecular revolution and how DNA markers can provide insight into the processes of evolution, the mapping of genes for complex phenotypes and the reconstruction of the human diaspora. In addition to this, there are explanations of the technological developments and how they affect the fields of forensic anthropology and population studies, alongside the methods of field investigations and their contribution to anthropological genetics. This book brings together leading figures from the field to provide an introduction to anthropological genetics, aimed at advanced undergraduates to professionals, in genetics, biology, medicine and anthropology.
The recently-developed statistical method known as the "bootstrap" can be used to place confidence intervals on phylogenies. It involves resampling points from one's own data, with replacement, to create a series of bootstrap samples of the same size as the original data. Each of these is analyzed, and the variation among the resulting estimates taken to indicate the size of the error involved in making estimates from the original data, In the case of phylogenies, it is argued that the proper method of resampling is to keep all of the original species while sampling characters with replacement, under the assumption that the characters have been independently drawn by the systematist and have evolved independently. Majority-rule consensus trees can be used to construct a phylogeny showing all of the inferred monophyletic groups that occurred in a majority of the bootstrap samples. If a group shows up 95% of the time or more, the evidence for it is taken to be statistically significant. Existing computer programs can be used to analyze different bootstrap samples by using weights on the characters, the weight of a character being how many times it was drawn in bootstrap sampling. When all characters are perfectly compatible, as envisioned by Hennig, bootstrap sampling becomes unnecessary; the bootstrap method would show significant evidence for a group if it is defined by three or more characters.
In the past 5 years, there has been a substantial increase in the use of Y-short tandem repeat loci (Y-STRs) in forensic laboratories, especially in cases where typing autosomal STRs has met with limited success. The AmpFℓSTR® Yfiler™ PCR amplification kit simultaneously amplifies 17 Y-STR loci including the loci in the “European minimal haplotype” (DYS19, DYS385a/b, DYS389I, DYS389II, DYS390, DYS391, DYS392, and DYS393), the Scientific Working Group on DNA Analysis Methods (SWGDAM) recommended Y-STR loci (DYS438 and DYS439), and the highly polymorphic loci DYS437, DYS448, DYS456, DYS458, Y GATA H4, and DYS635 (formerly known as Y GATA C4). The Yfiler™ kit was validated according to the FBI/National Standards and SWGDAM guidelines. Our results showed that full profiles are attainable with low levels of male DNA (below 125 pg) and that under optimized conditions, no detectable cross-reactive products were obtained on human female DNA, bacteria, and commonly encountered animal species. Additionally, we demonstrated the ability to detect male specific profiles in admixed male and female blood samples at a ratio of 1:1000.