Skip to content

Latest commit

 

History

History
46 lines (29 loc) · 1.6 KB

README.md

File metadata and controls

46 lines (29 loc) · 1.6 KB
Raw

build gtdb databases for sourmash

First, test your installation etc.

  1. Install sourmash v2 or v3.

  2. Download the GTDB-Tk reference data (currently release89)

  3. Edit config.yml in this directory to point to release89/taxonomy/gtdb_taxonomy.tsv.

  4. Make a directory that contains a ~dozen of the genomes from release89/fastani/datbase, and set genomes_location in config.yml to that directory.

  5. Set sigs_output_location to a local directory (you can leave it as-is if you like).

  6. Run snakemake. You can use -j 8 if you like.

At the end you should have one *.sig file in your sigs_output_location for each genome in genomes_location, and six output files in the current directory:

gtdb-release89-k21.lca.json.gz
gtdb-release89-k31.lca.json.gz
gtdb-release89-k51.lca.json.gz
gtdb-release89-k21.sbt.json
gtdb-release89-k31.sbt.json
gtdb-release89-k51.sbt.json

The first three are standalone LCA databases for sourmash, and the last three are SBT databases for sourmash; see also the .sbt.* hidden directories.

Build the full databases

  1. Copy config.yml to config-release89.yml

  2. Comment out prefix with a # at the beginning of the line.

  3. Change genomes_location to point to release89/fastani/database.

  4. Change sigs_output_location to point to another directory.

  5. Run snakemake.

With snakemake -j 8, building the 25,000 signatures takes about 3 hours on the farm cluster at UC Davis, and each database takes about an hour to build after that. Memory requirements are far under 100 GB total.