Cohorts

The 2011 Nepalese census recorded 125 distinct ethnic groups [4]. Using this census data and descriptors of the “Nepal Federation of Indigenous Nationalities”, we selected the most common indigenous Nepalese ethnic groups that collectively represent approximately 50% of the population of the Eastern Mountain and Hill Region of Nepal (Additional file 1: Table S1) [4, 33]. The four Nepalese populations selected based on census data were as follows; Rai (n = 20), Magar (n = 10), Tamang (n = 18) and Newar (n = 17). We also included the Indo-Aryan (n = 11) considering their close ethnic relation to the Newar [34]. For the purpose of downstream genetic analysis these Nepalese ethnic groups were merged to form the “Nepal” cohort. Nepalese ethnic groups were recruited from regions in close proximity to the Sherpa including Lukla, Solukhumbu and Kunde. For additional information on the Nepalese cohort see Additional file 3 [35–40].

We established a cohort of 118 Sherpa from three distinct recruitment efforts, including 49 previously genotyped individuals [15]. The Sherpa were sampled from seven high altitude villages in the Khumbu region of Eastern Nepal, namely Thame (n = 43), Khumjung (n = 30), Namche (n = 19), Lukla (n = 9), Khunde (n = 7), Forte (n = 3) and Solukhumbu (n = 7). Details of two of these recruitment efforts have been described previously [15, 41]. We will refer to these samples collectively as the “Sherpa” cohort.

For fine scale population analysis we have catagorised the Sherpa by village and the Nepalese by their ethnic group (see Additional file 1: Table S2).

We extended our cohort to include other populations representing the greater Himalayan region. We included formerly genotyped Tibetans from Lhasa (n = 29), Yunnan (n = 35), Tuo Tuo river (n = 46), Maduo (31), and Tibetans resident in the UK (n = 27) [42–45].

Previously genotyped individuals representing populations along the Silk Road were also included, considering their cultural and economic ties with populations of Central Asia [46]. These consisted of individuals from Alga and Almaty in Kazakhstan (n = 59), Bukhara, Karshi, and Tashent in Uzbekistan (n = 83), Shing, Zeravshan, Kalaikhum, Khorog, and Rushan in Tajikistan (n = 83) and Krakoi and Kara-koo in Kirghizstan (n = 184). We refer to these populations collectively as the “Pamir” cohort.

We also included genotype data for 331 individuals from 18 mainland Indian populations kindly provided to us by Basu et al. [18] and individuals from the Human Genome Diversity Project to include the Han (Beijing China, n = 44), and individuals from Lahore (Pakistan, n = 168). See Additional file 1: Table S2 for detailed information of recruitment sites for all individuals included in the analysis.