If you’ve ever tried to simulate ligand unbinding pathways from proteins, then you’ve likely faced the challenge of defining which atoms belong to the ligand — especially in systems containing cofactors, solvent molecules, or multiple ligands. Making a mistake at this step can compromise the entire simulation, creating misleading energy profiles or broken ligand trajectories.
The Ligand Path Finder app in SAMSON makes it relatively straightforward, but it’s still easy to overlook key details. This post walks you through how to accurately define the ligand atoms, why that matters, and how to verify your selection before launching the search for unbinding pathways.
Why ligand atom selection matters 🎯
In unbinding pathway modeling, the ligand is treated differently from the rest of the system. The ligand is the only part of the system undergoing guided sampling with the ART-RRT algorithm. If non-ligand atoms are accidentally included—or if some ligand atoms are excluded—the ligand won’t behave as expected. This may result in physically unrealistic motions or even cause the app to behave unpredictably.
How to define the ligand in SAMSON’s Ligand Path Finder
In the sample system provided with the tutorial, the ligand is Thiodigalactosid (TDG). You can see it listed in the Document view as TDG. To define the ligand atoms properly:
- In the Document view, select the
TDGnode to automatically select all ligand atoms. (You can also use the viewport selection tools.) - In the Ligand Path Finder App, click the Set button next to the ligand selection step.
After this, check the Advanced information box: you should see a message like “31 atoms set as ligand atoms”. If the atom count seems too low—or too high—it’s worth double-checking that extraneous atoms (e.g., cofactors or solvent) weren’t selected.
Common mistakes to watch for
- Partial selection: If only part of the ligand is selected, it may stretch or collapse unrealistically during motion generation. Always verify the entire ligand is selected.
- Including non-ligand atoms: Sometimes arbitrary molecules in the active site or buffer region are selected by mistake. Use the structure tree view to isolate only the ligand.
- Charge groups and annotations: If your system has specialized atom groupings for charge calculation or descriptors, make sure you’re not misusing those as proxies for ligand definition unless they match the actual molecular identity.
Chasing clean simulations
Correctly defining the ligand in the Ligand Path Finder App sets the foundation for meaningful simulation results. You’ll later choose active atoms, define fixed protein atoms, configure the sampling box, and launch the ART-RRT planner — all of which depend on an accurate atom selection.
Once defined, the ligand atoms will appear in green and red in the sampling box visualization: red for fixed atoms, green for active ARAP atoms, and blue for passive ARAP atoms. Verifying visually can be an extra safeguard before launching a potentially time-consuming run.
If you’re building a pipeline or working with varied PDB files, you may want to script common selection routines or rely on consistently named groups across files.
You can find the complete step-by-step tutorial in the Ligand Path Finder documentation.
SAMSON and all SAMSON Extensions are free for non-commercial use. You can download SAMSON at https://www.samson-connect.net.