Fix383

[fgrunewald]

@tylerjereddy

I've fixed a bug in CGSmiles that was causing the issue with the atomnames. It is still under review gruenewald-lab/CGsmiles#33 but I made sure it works this time.

Furthermore, I noticed that your CGSmiles string did not include terminal residues. That was a mistake on my side. When we changed the CGSmiles parser we did not fully update gen_ff. In a nutshell, you should always define terminal residues, which in this case would be hydrogen atoms. See repro.py.

Finally, I've added a python script that computes single-point energies between two itp files. That is the best way to check that the itp produced with polyply matches the original one. If all goes well, there should only be minor differences in Coulomb energies because we slightly change the charges. Sometimes the LJSR energies are a little different, which as far as I can tell has to do with GROMACS run initialization not being the same. N.B. the script compare.py requires a bunch of python libraries, so make sure to install the dependencies.

With the ff file you produced you can already get started with the larger polymer. For future reference, I would probably use some CH3 groups as terminal residues to avoid polarization across residues.

[tylerjereddy]

@fgrunewald excellent, thank you, making some progress with this now!