SBML Showcase

Contributors
Padraig Gleeson (wiki contributor) Padraig Gleeson (wiki & code contributor) Stella (code contributor) Robert Vickerstaff (code contributor) Stella Prins (code contributor) harveymannering (code contributor)

SBML Showcase¶

This repository is intended to highlight and show examples of interactions between the Systems Biology Markup Language (SBML) and Simulation Experiment Description Markup Language (SED-ML) and NeuroML/Open Source Brain.

The repository contains files and instructions for testing validity of Systems Biology Markup Language (SBML) and Simulation Experiment Description Markup Language (SED-ML) files, and their compatibility with different simulations engines. Much of this functionality has been embedded in pyNeuroML.

Testing validity of SBML/SEDML and compatability with simulation engines¶

The tables below show the results of some automated simulation pipelines for testing SBML/SEDML files.

SBML¶

LEMS_NML2_Ex9_FN

BioModels¶

Test Suite¶

Clone the Repository¶

To clone this repository to your local machine using the following command:

git clone https://github.com/OpenSourceBrain/SBMLShowcase.git
cd SBMLShowcase

Conda environment¶

To set up a conda environment for the code in the SBML, BioModels, and test_suite folders.
conda create -n sbmlshowcase -c conda-forge python=3.10 conda activate sbmlshowcase pip install -e . conda activate sbmlshowcase

for developers¶

conda create -n sbmlshowcase-dev -c conda-forge python=3.10
conda activate sbmlshowcase-dev
pip install -e .[dev]
conda activate sbmlshowcase-dev

To install the pre-commit hooks
pre-commit install

Main Folders¶

The repository is organized into several folders each serving a specific purpose. Some of the key folders are:

SBML Contains scripts and files related to a few simple SBML models and testing their compatibility with different BioSimulator engines.
BioModels Includes scripts for parsing and testing the compatibility of BioModels with various BioSimulator engines.
test_suite Holds test cases and scripts for validating the functionality of the repository's tools and models.
utils containing general functions.

SBML¶

SBML contains a simple SBML model LEMS_NML2_Ex9_FN.sbml and it's simulation file LEMS_NML2_Ex9_FN_missing_xmlns.sedml.

test_compatibility_biosimulators adds any missing XML or FBC namespaces to the SED-ML file, wraps up the SBML and SED-ML files into an Open Modeling EXchange format (OMEX) file, submits the OMEX file to different BioSimulator engines either locally via docker (indicated with an L for local in the results table), or remotely through the BioSimulators API (indicated with an R). The results table created by this script can be found in here.

To run the script, change your directory to the SBML folder and run:

python test_compatibility_biosimulators.py

The following scripts test specific components of the process described above and are mainly run as workflows in GitHub Actions.
- test_biosimulators_api
- test_biosimulators_docker
- test_biosimulators_local
- test_biosimulators_remote
- test_compatibility_biosimulators
- test_tellurium

BioModels¶

Overview with all curated BioModels¶

parse_biomodels creates an overview table in BioModels/README.md, testing curated BioModels.

To run the script, change your directory to the BioModels folder and run:

python parse_biomodels.py

Test specific BioModels¶

test_biomodels_compatibility_biosimulators
tests engine compatibility of BioModels listed in biomodel_id_list (BIOMD0000000001, BIOMD0000000138, BIOMD0000000724, BIOMD0000001077) and creates subfolders containing the results. Results can be found in BioModels/BIOMOD_id/tests/results_BioModel_name.md.

For example for BioModel BIOMD0000000001, results can be found here.

test_suite¶

Overview with semantic test suite cases¶

Processes the semantic SBML Test Suite test cases which should all be valid SBML models with deterministic simulation results.

process_test_suite creates an overview table summarising whether the SBML files, SBML units, and SED-ML files are valid. In addition to that it shows whether the XML SBML namespaces (xmlns-sbml) are missing, and whether the models can be run (pass / FAIL) in tellurium natively, or remotely through the BioSimulators API. The table can be found here.

To run the script, change your directory to the test_suite folder and run:

python process_test_suite.py

There are several command line options for testing a limited amount of cases or specific test suite cases, see the test_suite README for more information.

Testing specific test suite cases¶

test_test_suite_compatibility_biosimulators tests engine compatibility of specific semantic test suite cases (00001, 01186) Results can be found in test_suite/test_case_id/tests/results_compatibility_biosimulators.md.

For example for for semantic test case 00001, results can be found here.

To run the script, change your directory to the test_suite folder and run:

python test_test_suite_compatibility_biosimulators.py

There are several command line options in the test_suite README.

Converting NeuroML2/LEMS to & from SBML¶

Most of the interactions between SBML and LEMS/NeuroML showcased here are enabled by:

The SBML to LEMS import function in: org.neuroml.importer
The LEMS/NeuroML to SBML export function in: org.neuroml.export

Note these features can be accessed easily with the pyNeuroML tool. For example:

Load LEMSFile.xml using pyNeuroML, and convert it to SBML format:
```
 pynml LEMSFile.xml -sbml
```
Load LEMSFile.xml using pyNeuroML, and convert it to SBML format with a SED-ML specification for the experiment:
```
 pynml LEMSFile.xml -sbml-sedml
```
Load SBMLFile.sbml using jSBML, and convert it to LEMS format using values for duration & dt in ms
```
pynml -sbml-import SBMLFile.sbml duration dt
```