Fig. 3: Ancient human mtDNA and nuclear DNA isolated from DCP1.

a, The position of DCP1 in a Bayesian tree reconstructed from modern^31,32 and ancient³³ human mtDNA sequences (see Supplementary Information 5 for the full tree). Nodes are labelled with the corresponding posterior probabilities, and the x axis represents years from the present. Identified haplogroups are outlined by the bars on the right. rCRS, revised Cambridge reference sequence. b, X–autosome proportion in DCP1 (using all and deaminated molecules only) in comparison to data from six other ancient hominin individuals^34,35. Circles correspond to the calculated values of the ratios for the number of X to (X + autosomal) fragments for each individual (n (of single-nucleotide polymorphisms (SNPs)) = 20,526, 3,734, 124,862, 85,901, 34,756, 41,632, 34,677 and 72,992 SNPs for each calculation, as ordered on the x axis). The error bars represent 95% binomial CIs of the measurement in each individual. c, Principal component (PC) analysis of non-African modern human genomes³⁶ (grey) with ancient human genomes (coloured) projected on top. DCP1 was analysed twice, using all data versus deaminated fragments only. AG3, Afontova Gora 3; ANA, ancient Native Americans; MA1, Mal’ta 1; Russia_HG, Russian hunter–gatherers; UKY001, Ust Kyakhta; WSHG, West Siberian hunter–gatherers.