- What it is(n't)
- How it was made
- How to get it
- How to view it
- How to use it in research
- Information for users of...
- How to cite it
We present a reference phylogenetic tree (or more precisely, mutiple trees depending on your choice) for bacterial and archaeal genomes that are publicly available from NCBI RefSeq and GenBank. It means to server as a reference for researchers to explore the evolution and diversity of microbes, and to improve the study of microbial communities.
We do not attempt to create a new taxonomy. However we provide annotations (and curations) for the tree and the genome catalog based on either NCBI or GTDB taxonomy.
In brief, we used ASTRAL to generate a consensus tree by summarizing individual trees of 381 single-copy marker genes extracted from 10,575 genomes.
For comparative purpose, we also generated multiple trees using the conventional gene alignment concatenation strategy, and using multiple alternative genome and gene sampling rules. Detailed protocols are provided.
We recommend using this tree as the reference phylogeny for observations and downstream applications, together with the genome metadata.
Multiple trees, built using different input data and methodology, together with the corresponding metadata, curated taxonomy and other information, are provided in this repository. Please browse the data directory for details.
The genome and protein sequences, multiple sequence alignments and other large data files are available at Globus, with endpoint name WebOfLife (owner: [email protected]). For instruction on how to transfer files via Globus, please read this guide.
For readers interested in exploring the evolutionary storying underlying our trees, we provide high-resolution renderings at multiple levels, using NCBI and GTDB taxonomies, as images (PDF format) or compiled packages that can be directly parsed by FigTree and iTOL.
We also provide the Newick files, metadata of taxa and nodes, and code for perform rendering
In addition, our group is actively developing Empress, a novel interactive visualizer for massive trees (with hundreds of thousands of tips). Please stay tuned!
In addition to direct eyeballing, the reference tree can be used to extend the understanding of the composition and diversity of microbial communities. See protocols.
IDs of our genome pool are directly translated from NCBI assembly accessions. Duplicate genomes are merged.
A script is provided to download genomes fresh from the original NCBI server.
we provide mock taxdump files based on curated NCBI and GTDB taxonomy.
Mappings to GTDB genomes IDs are provided in the genome metadata. In the current release, 9,732 (92.03%) of the 10,575 genomes have corresponding GTDB IDs. Annotation (and curation) of our tree using GTDB taxonomy are provided. The relative evolutionary divergence (RED) (Parks et al., 2018) of tree nodes (which are mapped to taxonomic groups) are provided.
Mappings to IMG genome/taxon IDs. are provided in the genome metadata. In the current release, 6,758 (63.91%) of the 10,575 genomes have corresponding IMG IDs.
The reference tree can be used for the diversity analysis of shotgun metagenomes, using phylogeny-aware algorithms such as UniFrac for beta diversity, and Faith's PD for alpha diversity. See protocols.
A derivative for 16S rRNA-based analysis is under development. Please stay tuned.
The tree can replace the NCBI taxonomy hierarchy used in a Kraken analysis to guide the heuristic search. See here for how-to.
Checkout protocol for taxonomy-free profiles
The tree can replace the NCBI taxonomy hierarchy used in a Kraken analysis to guide the heuristic search. See here for how-to.
The 381 marker genes used to build the tree are a curated subsample of the 400 marker genes originally implemented in PhyloPhlAn (Segata et al., 2013). For each marker gene, we provide functional annotation, gene tree and its degree of congruence with the species evolution. Please see data/markers and data/trees/genes.
If you use the data, code or protocols developed in this work, please directly cite our website: https://github.com/biocore/wol. A manuscript detailing this work is current under peer review. Please stay tuned.
Please forward any questions to the project leader: Dr. Qiyun Zhu ([email protected]) or the senior PI: Dr. Rob Knight ([email protected]).