(A) The influence of the ORF finding approach on metagenomic databases. Venn diagrams indicate overlap among non-redundant peptide identifications obtained using ORFs found by FragGeneScan (FGS) or six-frame translation (6FT) as sequence databases. Percentage increase related to the use of a 6FT database in addition to the corresponding FGS database is shown on the bottom-right of each Venn diagram. Metagenomic databases derive from a 6-Mbps metagenome. Peptide identifications were obtained at 5Â % FDR, using three different bioinformatic platforms (left, MetaProteomeAnalyzer; middle, MaxQuant; right, Proteome Discoverer) with the corresponding (and above indicated) search engines and peptide validation tools. (B) Comparison between reads- and contigs-based metagenomic databases at different sequencing depths. Venn diagrams indicate overlap among non-redundant peptide identifications obtained using reads (R) or assembled contigs (C) as sequence databases. ORFs from both reads and contigs were found by FragGeneScan (F). Metagenome sequencing depth was 18 (top), 6 (middle), or 3 (bottom) Mbps. Percentage increase related to the use of contigs-based database in addition to the corresponding reads-based database is shown in red on the bottom-right of each Venn diagram, while percentage increase related to the use of reads-based database in addition to the corresponding contigs-based database is shown in blue on the bottom-left of each Venn diagram. Peptide identifications were obtained at 5Â % FDR, using three different bioinformatic platforms (left, MetaProteomeAnalyzer; middle, MaxQuant; right, Proteome Discoverer) with the corresponding (and above indicated) search engines and peptide validation tools. (TIF 2050 KB)