This would allow discrimination between these hypotheses. With whole genome autosomal data, we could investigate the whole population samples of both females and males, including either carriers of C3* chromosomes or of other Y haplogroups, since any admixture would affect the whole population. 31 samples from Ecuador (12 male and 10 female Waorani, 7 male and 2 female Kichwa) with DNA concentrations between 0.05 and 1 ng/μl were chosen for whole-genome amplification (WGA). Between 10 and 25 μl (depending

on the DNA concentration) were concentrated in a Speed Vac (Thermo Scientific) to increase the DNA concentration to at least 1 ng/μl. Afterwards the samples were whole-genome amplified using click here the Illustra GenomiPhi HY DNA Amplification Kit (GE Healthcare). The protocol was adapted to a final volume of 20 μl as follows: 1 μl of the sample DNA and 9 μl sample buffer were mixed and denatured for 3 min at 95 °C. Samples were then cooled to 4 °C. In the next step, 9 μl find more reaction buffer and 1 μl enzyme

mix were added to the sample and incubated as 30 °C for 4 h. The WGA reaction was inactivated at 65 °C for 10 min. 11 JPT samples with concentration of 10 ng/μl [12] were amplified using the same protocol. The quality of the resulting DNA was tested by a PCR reaction using the AmpFlSTR® Exoribonuclease NGM™ PCR Amplification Kit. WGA samples were diluted

100 or 200 times depending on initial DNA concentration. For all WGA-treated samples, full profiles were obtained which were concordant with the DNA profiles of the original unamplified sample DNA. This study, together with the informed consent, was approved by the ethics committee of the Institute of Legal Medicine and Forensic Sciences (Charité-Universitätsmedizin, Berlin, Germany) under the accession number 11-2010/02. 31 individuals from Ecuador (22 Waorani and 9 Kichwa) and 11 from Japan (JPT) were genotyped using the Illumina HumanOmni2.5-8 (Omni2.5) BeadChip. Genotypes across these samples were called using Gencall (http://www.illumina.com/Documents/products/technotes/technote_gencall_data_analysis_software.pdf) via the Sanger standard genotype-calling pipeline, then merged with available genotypes from the HGDP population panel [13]. We then removed individuals with a low genotyping rate (>20% missing data) and with high relatedness (PI_HAT > 0.5); for the HGDP data, we used the subset of individuals recommended [14]. The final dataset consists of 207,321 single nucleotide markers (SNPs) with an average genotyping rate >99.7% in 967 individuals.