Detection of Antimicrobial Peptides from Fecal Samples of FMT Donors Using Deep Learning

Songlin Wei¹Huifang Yin²Xinliang Hu²Yulang Chi^3*Lishan Zhang²Bangzhou Zhang^2,4Kai Qian⁴Wei Xu^2,4*

• ¹School of Information Engineering, Xiamen Ocean Vocational College, Xiamen, China
• ²Xiamen Chenge Biotechnology Co., Ltd, Xiamen, China
• ³College of Oceanology and Food Science, Quanzhou Normal University, Quanzhou, China
• ⁴Basic Medicine College, Yichun University, Yichun, China

Antimicrobial peptides (AMPs) represent a class of short peptides that are widely distributed in organisms and are regarded as an effective means to tackle bacterial resistance, potentially functioning as substitutes for conventional antibiotics. In this study, we employed metagenomics in combination with deep learning to mine AMPs from the 120 fecal microbiota transplantation (FMT) donor metagenome, successfully predicting 2,820,488 potential AMPs. Subsequently, a comprehensive analysis of the candidate AMPs was conducted through metaproteomic cross-validation, solubility analysis, cross-validation with other prediction tools, correlation analysis, and molecular dynamics simulations. Finally, four candidate AMPs were selected for chemical synthesis, and experimental validation identified two with broad-spectrum antimicrobial activity. Furthermore, molecular docking was utilized to further analyze the antimicrobial mechanisms of the candidate AMPs. In brief, our research highlights the potential of mining novel AMPs from the fecal microbiome and provides new insights into the therapeutic mechanisms of FMT.