Protocols#
openfe provides several protocols for calculating free energy differences, each suited to different scientific problems. This page provides an overview of the available protocols, guidance on choosing between them, and their main limitations.
Available protocols#
Relative binding free energies#
Relative protocols compute the difference in binding free energy (\(\Delta\Delta G\)) between two ligands.
Transforms one ligand into another using a hybrid topology defined by an atom mapping.
Can also calculate relative hydration free energies.
Inserts one ligand while removing the other, using orientational restraints.
No atom mapping required.
Absolute binding free energies#
Absolute protocols compute the binding free energy (\(\Delta G\)) of a single ligand.
Decouples the ligand from the binding site and solvent, using orientational restraints in the complex.
Other protocols#
Absolute Solvation Protocol: absolute hydration free energy of a small molecule.
Plain MD Protocol: conventional (non-alchemical) molecular dynamics, for example to check the stability of a binding pose before running free energy calculations.
Choosing a protocol#
Different binding free energy protocols are suited to different applications. The table below summarises what each protocol is best suited for and its main limitations. More than one protocol may be suitable for a given problem.
Protocol |
Best suited for |
Limitations |
Relative cost |
|---|---|---|---|
|
|
Lowest per edge; higher for net charge changes |
|
|
|
Higher per edge |
|
|
|
Higher per ligand |
Cost estimates are based on using the default settings. To see the exact settings
for your installed version, inspect the protocol’s default_settings().
Domain of applicability#
The openfe binding free energy protocols are primarily designed to calculate the noncovalent binding of small molecules to proteins. Not all molecular systems or alchemical transformations are supported.
System types and transformations#
The following tables group molecular systems and transformations according to the available testing and validation evidence. These categories indicate the current level of experience with openfe, rather than a guarantee of accuracy for a given system.
Support also depends on the selected force fields and on system preparation.