[gmx-developers] Conformation representation and energy function

[Mark Abraham]

> And if you still want to get your hands dirty inside GROMACS, I think you should model your approach on the functionality of mdrun -rerun. You want GROMACS to find the energy of a potentially infinite set of coordinates that appear magically, and that is what mdrun -rerun does. The magic source is a file in its case, but that is easy to rework for your context. Get out a debugger and step through the code doing a rerun of a small box of water :-) Once you've got a feel for what's going on, now you can pay attention to the data structures of importance (basically, "state->x").

And if you're still keen even after reading Anthony's comments, you  should reconsider your choice of MD software. The prime virtues of  GROMACS are in its speed and range of recent algorithms, which are both  irrelevant to you. Other MD codes exist that are more  developer-friendly, and thus better suited for experimenting with new  algorithms. IIRC TINKER might be one.

Mark 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20101006/3e1abe88/attachment.html>