Can't launch decomposition of CHARMM protein-ligand system

[Ildar]

Greetings, dear developers! 

I have a protein-ligand system modeled with CHARMM27 forcefield. I'd like to use decomposition analysis, pairwise interaction between ligands and their neighbors. 

I've been following the example for MMPBSA for CHARMM systems and used provided in the example input file. I can successfully perform MMPBSA analysis without decomposition, but adding &decomp section makes it fail after trajectory preparation.

I tried to specify residues for decomposition direct and by "within", but both were unsuccessful. And there is also no clue(at least for me) in the log file what occurs the error, nor in mdout file. 

I'm sure that the ligands group in the index file is correct - I specified directly by atoms indices from the PDB file. 

Attaching the output of the command, log, index, mdout, input, and PDB files. 

Sincerely,

Ildar

[Mario Sergio Valdes]

This is a known error in v1.6.0. Please, update gmx_MMPBSA to the latest version (v1.6.1) and check if the error persists.

Let me know if this works for you.

PS: Please, note you will compute 3501 frames in only 2 CPUs using PB. This calculation can take a long time. We recommend you use a few frames (about 10 to 100) to test if the method works for your system. For example, you can set interval = 100 to get only 35 frames spaced in the trajectory. If everything looks ok for you, then perform the calculation in the entire trajectory.

Happy BFE calculation

Mario S.

[Ildar]

Yes, it worked for me. (Actually, I faced this issue earlier, but after several attempts to install gmxMMPBSA on a cluster forgot to update.)

Thanks a lot!

PS. This is the test (on my local laptop), the main launch will be on a cluster.

среда, 3 мая 2023 г. в 14:26:47 UTC+8, mariosergi...@gmail.com: