Fig. 1: The integrated mining strategy of AMPs from ancient biological samples.
From: Identification of antimicrobial peptides from ancient gut microbiomes
a The bioinformatic approaches for ancient AMPs mining tasks. Gut microbiome sequences were assembled form ancient biological samples of human stool, and AMPs were mined from these data. Further screened by potential cytotoxicity, the selected candidates were experimentally validated in vitro and in vivo. b The prediction of AMP ORFs from seven well-processed palae-feces samples of Homo sapiens (1000–2000 years ago) with geographical locations. c–f The comparison of molecular features including isoelectric point, aromaticity percentage, molecular weight, charge at pH 7.0 and Hydrophobicity between previously reported AMPs (n = 8794) and Predicted AMPs (n = 723,045) from ancient samples. The statistical tests were conducted by 2-sided Welch T test with the adjustment of Cohen’s d method (p < 0.05 and |Cohen’s d | > 0.8, *, p < 0.01 and |Cohen’s d | > 0.8, **, p < 0.001 and |Cohen’s d | > 0.8, ***). The 2-sided Welch T test p-value in molecular weight comparison is less than 1 × 10−16. g Screen criteria of AMP candidates for further experimental validations. h The experimental validations for antimicrobial activities of the predicted 40 peptides from ancient biological samples towards P. aeruginosa 27853, E. coli 25404, B. subtilis 23857, and S. aureus 6538 in vitro. PC and NC represent positive control and negative control, respectively. The NC represents 1× PBS solutions. The PC represents 100 μM kanamycin for groups of E. coli 25404, B. subtilis 23857, and S. aureus 6538 or 100 μM polymyxin B for the group of P. aeruginosa 27853. BioRender was used during the construction of the Figure, with the issued License agreement number GG294KREHV (Created in BioRender. Chen, S. (2025) https://BioRender.com/7b397jf).
