Sun, Zhongzhi2025-07-242025-07-242025-07-24http://hdl.handle.net/10393/50688https://doi.org/10.20381/ruor-31268The human gut microbiome has a complex and dynamic relationship with its host and is closely associated with human health and disease states. As an emerging high-throughput omics technique, metaproteomics measures the presence and abundance of proteins in the microbial community and serves as a powerful tool for investigating interactions between the human gut microbiome and the host. However, metaproteomics faces unique challenges due to the complexity of microbial communities. Limited proteome coverage in metaproteomics has recently been highlighted, suggesting it could obscure important biological discoveries. This issue is further compounded by information loss during protein inference. As mass spectrometry-based (meta)proteomics directly measures peptides rather than proteins, information loss occurs due to the difficulty in precisely attributing peptides to their parent proteins. Despite this, most current metaproteomic studies still rely on protein-level analyses, and there are fewer applicable tools for peptide-level analysis. To address these challenges, this thesis enhances traditional protein-level metaproteomics with peptide-level analysis, develops practical tools to reduce information loss. First, we explored the potential of traditional protein-level metaproteomics analysis by investigating the effects of commonly used sugar substitute sweeteners on ex vivo human gut microbiome. Traditional protein-level metaproteomics analysis revealed that the functional profiles of the microbiomes grouped these sweeteners into two clusters, the noncaloric artificial sweetener (NAS) cluster and the carbohydrate (CHO) cluster. Notably, prebiotics fructooligosaccharide (FOS) and kestose (KES) clustered with the CHO cluster. Protein-level analysis further revealed that the sugar substitute sweeteners in the CHO cluster could modulate the metabolism of Clostridia. Next, to facilitate peptide-level human gut metaproteomics analysis, we constructed a core peptide database by extracting and reanalyzing data from published human gut metaproteomics studies. Raw files from fifteen publicly available human gut metaproteomics projects were reanalyzed and the identified peptide-spectrum matches (PSMs) were used to construct a core peptide database, MetaPep. The constructed MetaPep database enabled rapid and accurate identification of peptides for human gut metaproteomics. Additionally, annotation and reanalysis of peptides from the MetaPep database revealed the current landscape of human gut metaproteomics research. Finally, to advance peptide-centric analysis in metaproteomics studies, we explored peptide abundance correlations to enhance taxonomic and functional characterization of the human gut microbiome. Specifically, we analyzed peptide abundance correlations in a metaproteomics dataset derived from six in vitro cultured human gut microbiomes exposed to over one hundred drug treatments. Peptides originating from the same protein or taxon exhibited correlated abundance changes. Peptides from the same taxon were clustered within a peptide correlation map visualized with t-SNE. Leveraging these correlations, we enabled genome-level taxonomic assignments for a larger number of peptides. In addition, by focusing on single-species subsets, we calculated taxon-based normalized peptide abundances (TNPA) and constructed peptide abundance correlation networks, which revealed functional linkages among peptides and provided novel insights into the roles of previously uncharacterized microbial proteins. Altogether, this thesis demonstrated the utility of protein-level metaproteomics analysis by studying the functional effects of sugar-substitute sweeteners on the human gut microbiome. Furthermore, by constructing a core peptide database, analyzing peptides in this database, and studying peptide abundance correlations in a large metaproteomics dataset, this thesis provided practical tools for peptide-level analysis, which in turn advanced taxonomic and functional analysis in human gut metaproteomics.enAttribution-NonCommercial 4.0 Internationalhttp://creativecommons.org/licenses/by-nc/4.0/MetaproteomicsHuman gut microbiomePeptide-level analysisPeptide abundance correlationTaxonomic and functional annotationThesis