Preparing several copies of the same protein in one structure can seem straightforward: duplicate a chain, move the copy, and run coarse-graining. But duplicated chains may retain the same residue and chain identifiers. Those repeated identifiers can cause problems when Martinize2 generates topologies. The key preparation step is to make each replica unambiguous before conversion.
This matters whenever your input contains multiple copies of a protein—for example, a system assembled from several replicas of ubiquitin. Martinize2 treats each top-level structural model as a separate input model, so first check how your copies are organized in the SAMSON document. If the replicas are chains within one structural model, they are part of the same input model.
Build the replicas without losing track of them
You can create copies with the Molecular Box Builder, a Python script, or manually. For a manual workflow, first make sure the atoms are visible. Select the chain or chains to copy, use Ctrl/Cmd + C followed by Ctrl/Cmd + V, then move the newly copied chain with a move editor. The copy is initially placed at the same position as its source, so move it before proceeding. Repeat until you have the desired number of replicas.
Copying and moving multiple chains together can be useful when each replica is a multi-chain assembly. The important point is to identify which chains belong to each copy and ensure the final structure contains the intended replicas before generating the coarse-grained model.
Make chain and residue identifiers unique
After creating the copies, check both residue numbering and chain names. A duplicated chain can carry the same residue IDs and chain identity as the original. The resulting structure may look correct in the viewport, yet still contain ambiguous identifiers for topology generation. Visual separation is not a substitute for unique naming and numbering.
- Renumber residue IDs. Right-click the structural model and choose Structural model > Renumber residues and structural groups. In the dialog, keep the default starting value of 1 and click OK.
- Renumber chain IDs. Right-click the structural model and choose Structural model > Renumber chain IDs. Keep the default starting value of 0 and click OK.
- Rename the chains. Make sure every chain has a unique name. You can select a chain and press F2, use Rename from the context menu, or rename it through the Inspector.
Think of these as separate checks: renumbering residues does not rename chains, and renumbering chain IDs does not ensure that the visible chain names are unique. Review all three before continuing. It is also a good idea to save the prepared system so you can return to the version with clearly identified replicas.
Then generate the coarse-grained models
Once the replicas have unique numbering and names, continue with the usual Martinize2 workflow: select the structural model, set it in the extension, review the force-field and other options, choose a results folder, and create the coarse-grained models. The output includes coarse-grained structures and GROMACS topology files. For a single input structure, the generated coarse-grained model is loaded in SAMSON.
A quick preflight checklist can prevent avoidable confusion: Are the intended chains present? Are residue IDs unique? Are chain IDs and names unique? Is the system saved? Checking these details before conversion makes it easier to interpret the generated files and trace each coarse-grained result back to its input.
For the complete workflow and option descriptions, see the Martinize2 tutorial.
Note: SAMSON and all SAMSON Extensions are free for non-commercial use. Get SAMSON at samson-connect.net.