Examples
These examples cover most of the calculations and analyses available in gmx_MMPBSA. Although each example focuses on a specific case, gmx_MMPBSA can process systems containing several components (e.g., metalloprotein-ligand or protein-DNA-ligand complexes). A single run can also combine multiple calculation types, such as GB with alanine scanning and per-residue decomposition, or PB with interaction entropy and pairwise decomposition.
Maintaining example documentation¶
Example README files under examples/ are the canonical copies used by gmx_MMPBSA_test and GitHub browsing. The MkDocs site reads published copies under docs/examples/.
When you edit an example README:
- Change the file under
examples/only. - Run from the repository root:
- Commit both the
examples/change and the synceddocs/examples/copy.
CI runs python scripts/sync_example_docs.py --check and fails if the docs copies are stale.
The examples available through gmx_MMPBSA_test are defined in GMXMMPBSA/data/gmx_MMPBSA_test_manifest.json. When adding, removing, or renaming a testable example, update the manifest and run:
Jupyter notebooks¶
Two Jupyter notebooks are available for interactive use:
- Google Colab notebook: installs a conda-based CPU environment, runs bundled examples, supports uploaded user files, and displays results through the Python API.
- Local notebook: runs or loads local results, extracts data with the Python API, and plots energy terms with seaborn.
Systems¶
The following examples represent systems that gmx_MMPBSA can process and analyze. The list is not exhaustive; gmx_MMPBSA can also process other systems with compatible input structures.
- Protein-protein123
- Protein-ligand12
- Protein-DNA123
- Protein-glycan123
- MMPBSA with membrane proteins12
- Metalloprotein-ligand12
- Multicomponent system (Comp_receptor)123
Analyses¶
This section covers the analyses available in gmx_MMPBSA. Although each example focuses on a specific case, one run can combine several calculation types (e.g., GB with alanine scanning and per-residue decomposition, or PB with interaction entropy and per-residue decomposition).
- Single Trajectory Protocol123
- Multiple Trajectory Protocol1
- Binding free energy calculations
- Alanine scanning123
- Decomposition analysis123
- Entropy
- Stability calculations123
- QM/MMGBSA calculations1
- Correlation
- Python API extraction
- Local API/seaborn notebook
Input formats and force-field workflows¶
The examples below demonstrate preparation workflows for different input formats and force fields. The molecular systems shown are representative examples, not a list of supported receptor-ligand compositions. For example, a protein-protein tutorial under native AMBER or PSF/DCD does not mean that the workflow is restricted to protein-protein complexes. The same preparation principles apply to other systems when the resulting topologies, structures, trajectories, and molecular selections are compatible with gmx_MMPBSA or amber_MMPBSA.
Individual calculation models can impose narrower requirements. Review each tutorial's topology-conversion notes and the restrictions of the selected energy, entropy, decomposition, or membrane method.
Native AMBER inputs¶
Use native AMBER topologies and trajectories directly with amber_MMPBSA.
GROMACS topologies prepared with CHARMM¶
These examples use supplied GROMACS topologies containing CHARMM parameters, including specialized membrane and LPH workflows.
- Representative protein-ligand example12
- Specialized membrane protein-ligand example (CHARMM-GUI)1
- Specialized ligand example with LPH virtual sites1
GROMACS topologies prepared with OPLS¶
Converting PSF/DCD simulations¶
PSF and DCD files are preparation sources, not files read directly by gmx_MMPBSA. Convert them into a compatible topology, structure, trajectory, index, and receptor/ligand selections before analysis.
Created: November 9, 2020 14:28:26