We are happy to release the GlobDB r232!

The GlobDB r232 is based on the GTDB release 232 and contains 346,233 genomes, from 26 data sources.
Beyond the increased number of datasets and genomes that are included, there are two major changes in this release of the GlobDB:

1) Dereplication of the datasets is now done using Galah, which takes into account checkM2 quality scores. This means that the species representatives in this version the GlobDB are no longer dependent on the order the datasets are included, but also that a lot of species representative genomes have changed relative to release 226. The taxonomy is thus not transferable between these version, but we will strive to keep this more consistent in future releases.

2) The minimal quality thresholds have been upgraded to the same quality threshold that the GTDB uses (completeness - 5* contamination > 50), but using checkM2 quality scores. This meant a >30K genomes were excluded, but we believe the overall quality of the resulting database is therefore improved. We opted for this change because the quality of the source datasets is less uniform as the number increases.

This release includes the backbone of the GlobDB, but we will release additional data in the coming months. Stay tuned!