Approach for building G-quadruplex models with 3'-3' and 5'-5'polarity inversion

[muha]

Dear all,
I am working on the structural modeling of non-canonical G-quadruplexes (G4). Specifically, I am trying to build a monomolecular G4 model characterized by inversions of polarity (alternating 5'-3', 3'-3', 3'-5', and 5'-5' linkages) within the G-tracts.

I am looking for some guidance on how to build the starting structure: While DSSR is excellent at analyzing and rebuilding standard G4 topologies, i am not sure if it can handle these specific phosphodiester bond inversions.

Does DSSR have a built-in way to recognize and validate a chain that contains 3'-3' or 5'-5' linkages, or will it treat them as broken chains during analysis?
Are there specific commands in the latest DSSR releases (or via the mutate module) that allow for inverting the direction of a specific block of residues while maintaining the G-tetrad stacking geometry?

I am looking for a general methodology for handling these "inverted" backbone topologies in a way that remains compatible with subsequent Molecular Dynamics setups.
Once I find a stable workflow, I will be happy to summarize the steps for the community as per the forum guidelines.
Any suggestion is welcome!
Thank you for your time and for this useful resource.

Best