A ligand leaving a binding site is not necessarily following one obvious route. For a molecular modeler, the practical challenge is to explore plausible exit pathways without letting the search setup quietly decide the answer in advance. In SAMSON, the Ligand Path Finder app can search for candidate unbinding routes and display their conformations and energies. The key is to treat the result as a product of both the molecular model and the choices made before the search begins.
The app uses ART-RRT: T-RRT for pathway search and ARAP modeling to generate ligand motion. Constrained minimization adapts the protein as the ligand moves. This makes the setup more than a matter of pressing Run: the starting conformation, selected atoms, and sampling box all shape what the planner explores.
Start with a prepared, minimized model
Before searching, make sure the system is minimized and that the required Ligand Path Finder app and FIRE state updater are available. The tutorial’s sample is lactose permease with the ligand thiodigalactosid (TDG), and is a useful first system if you have not used the workflow before. For your own structure, preparation may include removing alternate locations, water, and ions, adding hydrogens, and fixing issues as needed. The SAMSON Protein Preparation & Validation tutorial covers that preparation step.
For energy evaluation in the documented workflow, select Universal Force Field (UFF) and FIRE. The tutorial specifies FIRE with a step size of 1 fs and one step. The system should already be minimized; if it is not, minimize it before starting the search.
Make the search geometry intentional
Choose the intended bound conformation as the starting state, then explicitly define the ligand atoms. In the sample, the conformation is bound_minimized, and the ligand is TDG. The remaining atoms are treated as protein atoms.
Next, choose active ARAP atoms to control ligand motion, and fixed ARAP atoms to keep the protein from drifting along with the ligand. The sample uses the ligand’s sulfur atom, S1, as active and the protein’s CA atom in HIS 205 as fixed. For a different protein, choose fixed atoms from a region expected to remain relatively static. This choice can influence the pathways found, so it is worth checking the selection rather than accepting it by default.
Orientation matters too. The sampling region is a box in Cartesian coordinates, so orient and align the system before defining it. A box that is poorly placed may spend search effort in directions that are not useful for the exit route you want to explore.
The sampling box is part of the hypothesis
The sampling box defines where the active atoms can be sampled. Its size and position therefore bias ligand motion and the resulting pathways. In the tutorial, the box is set to favor motion toward the periplasmic side of the protein. That is a deliberate choice—not a neutral setting. If you want to explore a different direction, adjust the box accordingly and keep a record of the choice so you can interpret or reproduce the search.
Use search settings to control the scope
The tutorial’s example uses two runs, 20 ARAP-modeling iterations, and 20 constrained-minimization iterations for each sampled state. It sets a maximum ligand displacement of 40 Å and an RRT extension step size of 1 Å. The maximum time per run also limits how long each search continues. These parameters determine the scope and granularity of the exploration; they are not a guarantee that every possible route will be found.
If reproducibility matters, use a specified seed. The planner increments the seed by one after each run; with the seed option unchecked, it uses a random seed. Record the sampling box and parameters alongside the seed, since the seed alone does not describe the full setup.
Read paths as candidates, not verdicts
When a route is found, the Results tab reports the number of states, minimum and maximum energies, saddle, barrier, and other details. Selecting a path and moving the slider lets you inspect its conformations in the structural model and view the corresponding UFF energy. Comparing several paths can help you see how their energy profiles and geometries differ; the table also allows remarks and path-energy export.
A computed route is a candidate shaped by the model, force field, atom assignments, and sampling region. Inspect the motion and energy profile, and be cautious about comparing results produced with different setups. The tutorial also describes exporting paths and refining them with the P-NEB app.
For the complete workflow and interface details, see the Ligand Path Finder documentation.
Note: SAMSON and all SAMSON Extensions are free for non-commercial use. Get SAMSON at samson-connect.net.