Fig. 3

Contamination estimates on artificial mixtures of Neandertal and present-day human DNA sequences. Estimates from AuthentiCT are shown with red dots, whereas the estimates from two other methods are shown in grey [17, 22]. Each point represents the difference between the estimated proportion of contamination and the proportion of present-day human DNA sequences introduced in the corresponding Neandertal dataset (different panels). Added contamination (x-axis) ranges from 5 to 95% in steps of 5% (see Additional file 1: Figure S3 for a narrower range from 1 to 10% in steps of 1%). Error bars correspond to 95% CI. There were not enough observations with 10,000 sequences for the method relying only on terminal deamination [3]. We note that the Neandertal datasets, particularly Spy 94a, may contain some present-day human DNA contamination in addition to the human contamination we introduced (see Additional file 1: Table S1 for the libraries used and the associated contamination estimates)