Metaproteomics can provide a direct means to identify microbial proteins in microbiota for compositional and functional characterization. However, in-depth and accurate metaproteomics is still limited due to the extreme complexity and high diversity of microbiota samples. We proposed an accurate taxonomic annotation pipeline for genes from metagenomic data, namely contigs directed gene annotation (ConDiGA), and used this method to build protein sequence databases for metaproteomic analysis. ConDiGA can tackle the current taxonomic annotation reliability problem in metagenomics-derived protein sequence database and can promote the in-depth metaproteomic analysis of microbiome. Moreover, when metagenomics-derived database is not available, public databases are frequently employed in metaproteomics. Nevertheless, utilizing large public databases for metaproteomic analysis can cause an expansion in the search space, prolong the search duration, and diminishing the sensitivity of identification. To address this issue, we present a high-abundance protein-guided hybrid spectral library strategy for in-depth data independent acquisition (DIA) metaproteomic analysis (HAPs-hyblibDIA). HAPs-hyblibDIA achieves effective filtering of public databases whose protein and peptide identification numbers are comparable to metagenomics-derived databases.