Skip to content
/ de_sim Public

Python-based object-oriented discrete-event simulation tool for complex, data-driven modeling

License

Notifications You must be signed in to change notification settings

KarrLab/de_sim

Repository files navigation

PyPI package Documentation Test results Test coverage Code analysis License DOI Analytics

DE-Sim: a Python-based object-oriented discrete-event simulator for modeling complex systems

DE-Sim is an open-source, Python-based object-oriented discrete-event simulation (DES) tool that makes it easy to use large, heterogeneous datasets and high-level data science tools such as NumPy, Scipy, pandas, and SQLAlchemy to build and simulate complex computational models. Similar to Simula, DE-Sim models are implemented by defining logical process objects which read the values of a set of variables and schedule events to modify their values at discrete instants in time.

To help users build and simulate complex, data-driven models, DE-Sim provides the following features:

  • High-level, object-oriented modeling: DE-Sim makes it easy for users to use object-oriented Python programming to build models. This makes it easy to use large, heterogeneous datasets and high-level data science packages such as NumPy, pandas, SciPy, and SQLAlchemy to build complex models.
  • Stop conditions: DE-Sim makes it easy to terminate simulations when specific criteria are reached. Researchers can specify stop conditions as functions that return true when a simulation should conclude.
  • Results checkpointing: DE-Sim makes it easy to record the results of simulations by using a configurable checkpointing module.
  • Reproducible simulations: To help researchers debug simulations, repeated executions of the same simulation with the same configuration and same random number generator seed produce the same results.
  • Space-time visualizations: DE-Sim generates space-time visualizations of simulation trajectories. These diagrams can help researchers understand and debug simulations.

Projects that use DE-Sim

DE-Sim has been used to develop WC-Sim, a multi-algorithmic simulator for whole-cell models.

Examples

  • Minimal simulation: a minimal example of a simulation
  • Random walk: a random one-dimensional walk which increments or decrements a variable with equal probability at each event
  • Parallel hold (PHOLD): model developed by Richard Fujimoto for benchmarking parallel DES simulators
  • Epidemic: an SIR model of an epidemic of an infectious disease

Tutorial

Please see sandbox.karrlab.org for interactive tutorials on creating and executing models with DE-Sim.

Template for models and simulations

de_sim/examples/minimal_simulation.py contains a template for implementing and simulating a model with DE-Sim.

Installation

  1. Install dependencies

    • Python >= 3.7
    • pip >= 19
  2. Install this package using one of these methods

    • Install the latest release from PyPI

      pip install de_sim
      
    • Install a Docker image with the latest release from DockerHub

      docker pull karrlab/de_sim
      
    • Install the latest version from GitHub

      pip install git+https://github.com/KarrLab/de_sim.git#egg=de_sim
      

API documentation

Please see the API documentation.

Performance

Please see the DE-Sim article for information about the performance of DE-Sim.

Strengths and weaknesses compared to other DES tools

Please see the DE-Sim article for a comparison of DE-Sim with other DES tools.

License

The package is released under the MIT license.

Citing DE-Sim

Please use the following reference to cite DE-Sim:

Arthur P. Goldberg & Jonathan Karr. (2020). DE-Sim: an object-oriented, discrete-event simulation tool for data-intensive modeling of complex systems in Python. Journal of Open Source Software, 5(55), 2685.

Contributing to DE-Sim

We enthusiastically welcome contributions to DE-Sim! Please see the guide to contributing and the developer's code of conduct.

Development team

This package was developed by the Karr Lab at the Icahn School of Medicine at Mount Sinai in New York, USA by the following individuals:

Acknowledgements

This work was supported by National Science Foundation award 1649014, National Institutes of Health award R35GM119771, and the Icahn Institute for Data Science and Genomic Technology.

Questions and comments

Please submit questions and issues to GitHub or contact the Karr Lab.