CMAP support

[vedranmiletic]

ff19SB supports (per-amino-acid) CMAPs, but cpptraj doesn't print their energies. It would be nice to have them supported, just like other bonded interactions are.

To give credit where it's due, this issue was also brought up in #842 that has since been closed.

[drroe]

This is a good idea, and one that has been on my agenda for a while. I actually have a branch that does an overhaul of how topology information is processed and handled (https://github.com/drroe/cpptraj/tree/prepareforleap.remote.buildatom) that I want to get merged into the master branch before adding this support, but recent upheavals at my current place of employment over the past few months have slowed by research/coding efforts dramatically.

I'm currently traveling, but I will make a concerted effort next week to try and get this code merged, or at the very least get some kind of minimal support for CMAP term energies. Thanks for bringing it back to my attention!

[vedranmiletic]

Was there any luck with this? Should I try to give it a go myself?

[drroe]

Sorry for the delays! I've been actively pushing to get the https://github.com/drroe/cpptraj/tree/prepareforleap.remote.buildatom branch committed (will be cpptraj V7) since that contains significant changes to Topology/parameter handling under the hood. I would be happy to have you work on CMAP support but I don't want you to waste any effort.

I'm currently in the clean up and documentation phase, so I should be able to cut a version 7 release soon (my goal is to do it before the Amber dev meeting later this month but that might be too ambitious). Would that timeline work?

[drroe]

I've got support for calculating CMAP energies with the energy action working in this branch: https://github.com/drroe/cpptraj/tree/prepareforleap.remote.buildatom

This will be merged in over the next few weeks to become CPPTRAJ version 7.0.

[vedranmiletic]

Wow, that was fast, thanks! Is it a drop-in replacement for the existing version in AmberTools? Anyhow, I will try it in our workflow before the end of the week and report back.

[drroe]

So you'll need to pull from that specific branch (prepareforleap.remote.buildatom) of my fork of cpptraj for now. I'm still working on officially releasing it and getting it into AmberTools.

It should be mostly backwards compatible with previous versions of cpptraj, but the topologies produced will be slightly different since I tried very hard to match LEaP ordering, etc, and also be smarted about how parameters are handled under the hood (trimming unused parameters after a strip and so on).

If you do try the code from that branch I'd appreciate any feedback.

[vedranmiletic]

Thanks for the tips.

On an unrelated note:

but the topologies produced will be slightly different since I tried very hard to match LEaP ordering

Can you point me to the part of the code that does this? I am curious to see how it is implemented (and compare to what we have in GROMACS).

[drroe]

(sorry for the delayed response, I'm swamped this week!)

Sure, although be aware it's a little hacky. The specific LEaP ordering depends on what LEaP routine was reading the input coords (e.g. PDB vs mol2 does within-residue ordering differently). The major difference is whether the LEaP loop structure orders things wiithin a residue front-to-back or back-to-front. The code for doing this in the upcoming cpptraj is in this file: https://github.com/drroe/cpptraj/blob/prepareforleap.remote.buildatom/src/Structure/GenerateConnectivityArrays.cpp

This took me a while to figure out and like I said the code is a bit hacky, but it seems to work. Let me know if you have any questions or if I can help in any way.

[vedranmiletic]

I tried compiling it, but I get:

CMake Error at src/CMakeLists.txt:156 (add_library):
  Cannot find source file:

    ParameterSet.cpp

  Tried extensions .c .C .c++ .cc .cpp .cxx .cu .mpp .m .M .mm .ixx .cppm
  .ccm .cxxm .c++m .h .hh .h++ .hm .hpp .hxx .in .txx .f .F .for .f77 .f90
  .f95 .f03 .hip .ispc


CMake Error at src/CMakeLists.txt:156 (add_library):
  No SOURCES given to target: cpptraj_parm


CMake Generate step failed.  Build files cannot be regenerated correctly.

I looked at CMakeLists.txt, but it looks fine to me. Do you see this yourself? Any idea how to fix it?

[drroe]

It has been a while since I tested the Cmake build on that branch - let me check and get back to you. Thanks for the report!

[drroe]

OK, it should be fixed now. However, note that installing via the configure script is recommended for standalone cpptraj install since it is usually better tested. For most systems, something like:

./configure --buildlibs -shared gnu

should be enough to configure cpptraj and set up all requirements prior to make.

Hopefully this helps!

[drroe]

Should be addressed by #1166