Files provided with GTDB R05-RS95. Analogous files are provide for Archaea. Both compress and uncompressed files are available in some cases for convenience. * METHODS Overview of methods used to form the GTDB taxonomy. * RELEASE_NOTES Overview of changes specific to this release of the GTDB. * VERSION Version identifier for this release of the GTDB. * bac120_taxonomy_.tsv GTDB taxonomy for all bacterial genomes assigned to a GTDB species cluster. * bac120_.tree Bacterial reference tree inferred from the concatenation of 120 proteins and spanning the representative genomes for each bacterial species cluster. This tree is used to curate the GTDB taxonomy. The provided tree is in Newick format, decorated with the GTDB taxonomy, and contains non-parametric bootstrap support values. * bac120_metadata_.tar.gz Metadata for all bacterial genomes including GTDB and NCBI taxonomies, completeness and contamination estimates, assembly statistics, and genomic properties. * Files in genomic_files_reps are specific to GTDB species representatives: * bac120_marker_genes_reps_.tar.gz Untrimmed and unaligned marker genes used in the concatenated alignment used to infer the bacterial reference tree. Genomes without a marker gene indicate either multiple hits or no hits were found for that gene. Genes are provided as both nucleotide and amino acid sequences. Trimmed and aligned hits can be found in bac120_msa_marker_genes_reps_.tar.gz. * bac120_msa_marker_genes_reps_.tar.gz Trimmed and aligned marker genes for GTDB representative for each of the 120 bacterial proteins. * bac120_msa_reps_.tar.gz Multiple sequence alignment used to infer the bacterial reference tree. * bac120_ssu_reps_.tar.gz FASTA file of 16S rRNA gene sequences identified within the set of bacterial representative genomes. The longest identified 16S rRNA sequence is selected for each representative genomes. The assigned taxonomy reflects the GTDB classification of the genome. Sequences are identified using nhmmer with the 16S rRNA model (RF00177) from the RFAM database. Only sequences with a length >=200 bp and an E-value <= 1e-06 are reported. In a small number of cases, the 16S rRNA sequences are incongruent with this taxonomic assignment as a result of contaminating 16S rRNA sequences. * gtdb_proteins_aa_reps_.tar.gz Amino acid FASTA files for all protein-coding gene sequences predicted with Prodigal. * gtdb_proteins_nt_reps_.tar.gz Nucleotide FASTA files for all protein-coding gene sequences predicted with Prodigal. * gtdb_genomes_reps_.tar.gz FASTA files for each GTDB representative genome. * Files in genomic_files_all cover all genomes passing the QC criteria of the GTDB: * bac120_marker_genes_all_.tar.gz Untrimmed and unaligned marker genes for all GTDB genomes. Genomes without a marker gene indicate either multiple hits or no hits were found for that gene. Genes are provided as both nucleotide and amino acidsequences. Trimmed and aligned hits can be found in bac120_msa_marker_genes_all_.tar.gz. * bac120_msa_marker_genes_all_.tar.gz Trimmed and aligned marker genes for all GTDB genomes for each of the 120 bacterial proteins. * ssu_all_.tar.gz FASTA file containing 16S rRNA sequences identified across the set of GTDB genomes passing QC. The assigned taxonomy reflects the GTDB classification of the genome. Sequences are identified using nhmmer with the 16S rRNA model (RF00177) from the RFAM database. All sequences with a length >=200 bp and an E-value <= 1e-06 are reported. * Files in auxillary_files: * bac120_.sp_labels.tree Synonymous to the bac120_.tree, except with species labels appended to each genome. * bac120_msa_mask_.txt Mask indicating which columns were trimmed from the 120 bacterial protein concatenated alignment. * bac120_marker_info_.tsv Information about each of the 120 bacterial proteins used to infer the bacterial reference tree. The order of proteins in this file indicates the order in which they are concatenated in the MSA. * gtdb_.dic List of all taxa with and without rank prefixes in GTDB. This can be used as a dictionary in word processing programs to indicate the correct spelling of GTDB taxa. See: https://www.officetooltips.com/word_2016/tips/how_to_share_the_custom_dictionary_in_word.html * gtdb_vs_ncbi__bacteria.xlsx Correspondence between GTDB and NCBI taxa ordered by degree of polyphyly. * gtdbtk_data.tar.gz Reference data required by the companion tool GTDB-Tk (https://github.com/Ecogenomics/GTDBTk) for classifying genomes according to the GTDB. This includes the genomic FASTA files for the GTDB reference genomes. * hq_mimag_genomes_.tsv List of isolates, MAGs, and SAGs meeting the MIMAG high-quality genome criteria (Bowers et al., Nat Biotechnol, 2017): - estimated completeness >90% - estimated contamination <5% - presence of the 5S, 16S, and 23S rRNA genes with a minimum length of 80 bp for 5S, 900 bp for archaeal 16S, 1200 bp for bacterial 16S, and 1900 bp for 23S - at least 18 tRNAs [An updated version of this file was released on October 6, 2020 that implements the above length testing of 5S, 16S, and 23S rRNA genes.] * metadata_field_desc.tsv Description of each field in the above metadata file and indication of the originating source of the metadata. * ncbi_vs_gtdb__bacteria.xlsx Correspondence between NCBI and GTDB taxa ordered by degree of polyphyly. * sp_clusters_.tsv Metadata file indicating the representative genome of each GTDB species cluster, the set of genomes assigned to the species cluster, and the average nucleotide identity radius used to circumscribe the species cluster. * synonyms_.tsv List of species considered synonyms in the GTDB taxonomy. * gtdb_taxa_not_in_lit_.xlsx List of GTDB families to phyla with Latin names without prior publication that have been introduced to achieve monophyly and rank normalization. * qc_failed.tsv List of genome assemblies failing the internal GTDB QC criteria.