I don't have any experience in Desmond, but I think Desmond should be easier when preparing ligand for simulation. You need to generate/parametrize your ligands for MD simulation.
I have a little bit experience of with GROMACS, it is free and moderately easy, but for ligand parameters, you need other software (Charm-GUI, ATB, and AmberTools). However, it is fun to learn and use GMX. Plus you can use Gmx trajectories in AmberTools to get the benefit of the ATools analysis modules. For a starting material, you can search Justin's Tutorial, well written for different use cases, and learn how to use Gmx.
Follow literature authored by https://www.researchgate.net/profile/kshatresh-Dubey and https://www.researchgate.net/profile/Sam-Visser and look for the tools and techniques employed by them.
This might be outside the scope of what you're researching. Still wanted to suggest comparing two substances in this model and hear what the difference between them are, because subjectively they feel the same.
Substances to compare:
MDMA
Borax Molly; aka 70 mg 5-MAPB, 20 mg 2-FMA , 2 mg 5-MeO-MiPT (see for more info https://redd.it/l98bop )
PS totally forgot to mention this, There is a lot of anecdotal reporting about how many Arylcyclohexylamines have antimicrobial effects. 3-MeO-PCP supposedly helped kill a stomach/intestine bacteria a friend of mine had that other meds were not able to treat/cure. Perhaps an interesting avenue for your studies?