maize-contrib#

maize is a graph-based workflow manager for computational chemistry pipelines. This repository contains namespace packages allowing domain-specific extensions and steps for maize. You can find the core maize documentation here.

Installation#

You can install maize-contrib with:

conda env create -f env-users.yml
conda activate maize
pip install -e ./

The first step will install maize and all required dependencies. If you plan on developing, you should use env-dev.yml instead. If you already have maize installed, you can clone add the additional dependencies and install as above:

conda activate maize-dev
conda install -c conda-forge rdkit scipy openbabel
pip install -e ./

Configuration#

Each step documentation will contain information on how to setup and run the node, as well as install the required dependencies. Dependencies can be managed in several ways, depending on the node and workflow you are running:

  • Through a module system:

    Specify a module providing an executable in the config.toml (see Configuring workflows) file. This module will then be loaded in the process running the node.

  • With a separate python environment:

    Some nodes will require custom python environments that are likely to be incompatible with the other environments. In those cases, the node process can be spawned in a custom environment. Note that this environment must still contain maize. Required custom environments can be found in the appropriate node directory.

  • By specifying the executable location and possibly script interpreter. This can also be accomplished using config.toml (see Configuring workflows and the maize example).

Tips#

  • If you’re unsure where to start with custom node, you can copy the example at maize/steps/mai/example as a template.

  • For custom environments it will be easiest to clone the maize-dev environment:

    conda create --name maize-new --clone maize-dev

Roadmap#

  • Glide & Ligprep

  • JSON interface for REINVENT

Indices and tables#