Systematic Evaluation of Database Builds in Metaproteomics for the Advancement of Microbiome Research

The microscopic life that inhabits a human shares a unique bond with its host. Microbes perform many functions that are vital to the survival of the human species and have long been shown to regulate the absorption of nutrients and to promote immune function. A lack of exposure to certain microbes early in life, excessive antibiotic usage, and improper diet can perturb human microbiomes and lead to disease. Since the emergence of omic sequencing technologies, it has now become possible to measure and monitor the genes and proteins made by these microorganisms to better understand how they contribute to host health or drive potential disease conditions. Early attempts at studying the genomes and proteomes of these environments have revealed that each person may house a unique community of microbial species, each of which contributes distinct and unique functions that separate healthy vs. disbiotic human gut microbiomes. A subfield of the omics known as metaproteomics is now being used to help characterize all proteins found in an environment, and it has been found to show tremendous potential for describing these microbial systems. However, a great deal remains to be done in this field to improve the accuracy and depth of information gained from these investigations. There is currently no accepted standard for determining which genomic databases are best suitable for searching protein sequencing data and it is difficult to know how/ if false positives are being incorporated into the results. The research for this thesis investigated how database curation affects the number of identified proteins and peptides identified from mice gut metaproteome spectra and explored how functional and taxonomic annotation varied according to the database used. The results indicated that translating a high-depth sequenced and assembled metagenome yielded the highest number of identifications while maintaining a low false discovery rate for mice fecal samples and that each database build identified unique and distinct functional and taxonomic information. The goal of this thesis is to better inform the field of metaproteomics and hopefully guide researchers towards a standard practice of using deep metagenome sequencing for database curation, resulting in more thorough coverage of microbiomes with greater confidence.