shadie is a Python wrapper around the evolutionary simulation program SLiM that provides a user-friendly API interface. In addition to implementing standard SLiM operators, shadie includes a rich library of functions to generate genome structure (chromosomes), mutation types, and selection scenarios. In particular, this includes the shadie.reproduction module for creating simulations of alternating hapliod/diploid lifecycles.

shadie has been updated for compatibility with SLiM4.

Docs (under construction): https://elissasoroj.github.io/shadie/

We are working on integrating a more rigorous testing framework into shadie. In the mean time, you can verify your analyses by checking that the generated SLiM code matches your expectations. Please don't hesitate to start a discussion if you have any questions.

To install shadie and all of its dependencies we recommend using conda. This will install all required Python packages in addition to slim.

# recommended installation method (installs shadie and slim>=4.2)
conda install shadie -c conda-forge

An alternative strategy that may be useful for developers is to install all dependencies with conda and then to install a development version of shadie from source (or a fork):

conda install shadie -c conda-forge --only-deps
git clone https://github.com/elissasoroj/shadie.git
cd ./shadie/
pip install -e . --no-deps

If you follow one of these instructions then shadie will automatically find the SLiM binary. If you wish to call a different version of SLiM installed locally on your machine this can be done easily when running shadie code.

~

Dependencies:

• numpy>=1.26.2
• pandas>=2.1.4
• toytree>=3.0.0
• toyplot>=1.0.3
• altair>=5.2.0
• pyslim>=1.0.2
• msprime>=1.3.0
• tskit>=0.5.6
• loguru>=0.6.0

~

Planned implementation:

• additional models (currently implementing: red algae, Vittaria appalachiana)
• convert user-provided phylogeny into SLiM3-compatible subpopulation demography.
• further extensions of the post-sim analysis module