A ligand’s bound structure shows where it sits, but not how it might leave. For molecular modelers, exploring that exit can be difficult: the ligand needs to move through a crowded protein environment, and the surrounding structure may need to adjust along the way. A pathway search can help generate candidate routes to inspect—provided the setup reflects the question you want to ask.
SAMSON’s Ligand Path Finder uses ART-RRT, combining T-RRT pathway search with As-Rigid-As-Possible (ARAP) modeling to generate ligand motion. Constrained minimization adapts the protein as the ligand moves. The result is a set of candidate unbinding paths to examine in the molecular scene, not a guarantee that a path is biologically preferred.
Start with a prepared, minimized system
Before searching, make sure the system is minimized. The tutorial uses the Universal Force Field (UFF) as the interaction model and FIRE as the state updater. If you are starting from your own structure, preparation may also be needed—for example, removing alternate locations, waters, or ions, and adding hydrogens. SAMSON’s Home > Prepare tools and the Protein Preparation & Validation tutorial are useful starting points.
The tutorial’s sample system is lactose permease with the ligand thiodigalactoside (TDG). It includes a minimized bound conformation named bound_minimized. Using a known bound conformation gives the search a clear starting state; for another system, create and select an appropriate conformation before running the planner.
Three setup choices shape the search
1. Identify the ligand precisely. Select the ligand atoms in the Document view and use Set in the app. The remaining atoms are treated as protein atoms. Check the reported atom count in Advanced information to catch an incomplete or overly broad selection before spending time on a search.
2. Choose active and fixed ARAP atoms thoughtfully. Active ligand atoms control the motion, while the other ligand atoms follow them. Fixed protein atoms provide positional anchors so the protein does not simply drift with the ligand. In the example, sulfur atom S1 in TDG is active, and a backbone CA atom in HIS 205 is fixed. These are tutorial choices, not universal defaults: fixed atoms should come from regions expected to remain relatively static, and the selection can influence the pathways found.
3. Treat the sampling box as a scientific choice. The box defines where active atoms are sampled. Its position and dimensions bias ligand motion, so a box that excludes a plausible exit route can prevent that route from being explored. The sample box is oriented to bias motion toward the periplasmic side. Before running, check the green box in the viewport and consider whether it covers the routes relevant to your question.
Make runs interpretable and reproducible
Search settings balance exploration with computation. In the tutorial example, two runs are used to seek up to two paths, with 20 ARAP modeling iterations and 20 constrained minimization iterations per newly sampled state. The maximum ligand displacement is 40 Å, and the RRT extension step is 1 Å. These are example values, not guarantees that the same settings suit every molecule or question.
If you need to repeat a search, enable Use seed and record the seed; the planner increments it after each run. Without that option, it uses a random seed. Also note the maximum time per run, which stops a run when the time limit is reached. During a search, Advanced information reports elapsed time, tree nodes, run number, and paths found—useful context when comparing runs or deciding whether to pause, resume, or stop.
Read the results as candidates, not conclusions
The Results tab lists paths and their energy summaries. Selecting a path displays its energy curve; moving the slider shows the corresponding conformation in SAMSON. You can select multiple paths to compare curves. The table includes the minimum and maximum path energies, the energy difference between them (Saddle), and the difference between the maximum and first-state energies (Barrier), along with the first and last energies and the number of states.
These values describe the computed path under the selected model and settings. They should not be mistaken for experimental measurements or, by themselves, proof of a real-world unbinding rate. Inspect the structures along the route as well as the curve, and record choices such as the starting conformation, fixed atoms, sampling box, and seed when documenting a comparison.
Paths can be exported as trajectories, and conformations can be exported at a chosen interval. For further refinement, the documentation points to the P-NEB app; Pathlines can help visualize ligand center-of-mass motion. These are useful next steps when a candidate route merits closer inspection.
For the complete workflow and interface details, see the Ligand Path Finder tutorial.
Note: SAMSON and all SAMSON Extensions are free for non-commercial use. Get SAMSON.