Fig 1.

Creation of the GSR database. (A) Merging algorithm to create the GSR database. The algorithm takes as an input a Reference database (R) and a Candidate database (C). Entries from the Candidate database are susceptible to being added to the Reference database after being evaluated. (B) Database merging workflow to obtain the GSR full-length 16S database. It has a final size of 90,408 entries with the following source composition: 22.29% RDP, 58.15% SILVA, 19.41% Greengenes, and 0.15% NCBI. Merging steps were performed using the merging algorithm described in panel A. (C) Sequence length distribution of the GSR databases. Databases for the variable regions are clustered at 100% identity.