Installing gPlace

gPlace runs on 64-bit Linux (x86-64 or ARM64) and macOS, with Python 3.12 or newer.

Note

gPlace is not on PyPI or Bioconda yet: install it from GitHub as below.

Install from GitHub

You need a C compiler and make, which most Linux systems have. If yours does not, conda can provide them (first command below); on macOS, run xcode-select --install.

conda create -n gplace -c conda-forge python=3.12 gcc make    # optional: a fresh environment
conda activate gplace

git clone https://github.com/cmc-aau/gPlace.git
cd gPlace
pip install .            # also builds gPlace's engine: a few minutes, the first time only

Then check that it works:

gplace check_install     # should end with "Installation OK"
gplace test              # places 20 bundled genomes and compares them with pplacer's output

The build downloads one library (GSL 1.16) from a GNU mirror. Without internet access, or to reuse an existing build, see Advanced installation.

What you need to run it

  • Memory: depends on the tree. Up to 44 GiB for the GTDB-Tk class-level trees; for the full GTDB bacterial tree, set a limit such as --max_memory 64G (see Large trees and memory limits). gplace benchmark estimates the memory for your data.

  • CPUs: any number. --cpus auto picks a good number of workers for your computer.

Other ways to install