Fig. 1

Provenience of sub-sampling protocol applied to and microbial taxa detected in the BRS palaeo-faecal specimen. a The location of Bushman Rock Shelter (BRS) in Limpopo Province, South Africa. b Lateral (left) and cross-sectional (right) views of the specimen indicating the sub-sampling protocol applied to facilitate DNA extraction, including ‘sedimentary control’ sample 1 (‘SC1’); faecal samples 2, 3 and 4 and sedimentary control sample 5 (SC2); ¹⁴C AMS dating (¹⁴C); isotope analyses (Iso); intestinal parasitic analyses (Ipa); scanning electron microscopy (Sem); and the preservation of a voucher sample (indicated in green shading). c Non-metric multi-dimensional scaling (NMDS) plot comparing the taxonomic community structure (by weighted Bray-Curtis dissimilarity analysis) of the BRS specimen (i.e. BRS2, BRS3 and BRS4) and the sediment controls (SC1 and SC2) with the ancient (Ötzi) (indicated as SI ‘small intestine’, LPLI ‘lower part of the lower intestine’ and UPLI ‘upper part of the lower intestine’), traditional (Hadza and Malawian) and modern (Italian) IM datasets (taxa were filtered for the occurrence of > 3 in at least 20% of the samples resulting in the inclusion of 371 taxa) (R = 0.6110 indicates ANOSIM analysis which revealed significant differences (p = < 0.001) between the ancient and modern IM samples). d Box-and-whisker plot indicating the relative abundance of intestinal bacterial phyla detected in the BRS specimen (i.e. BRS2, BRS3 and BRS4) (‘other’ comprises phyla with < 0.6% relative abundance). e Bar chart providing an overview of all environmental, commensal and pathogenic genera identified in the BRS specimen (BRS2, BRS3 and BRS4) and information concerning the DNA extraction and library preparation negative controls (E-LPCs) and modern and ancient sedimentary controls (SC1 and SC2) (data derived from Tables 1 and 2 and Table S1) (see the ‘Methods’ section)