NMA and TIP4PQ/2005

User discussions
1

GROMACS version: 2020-doubleprecision
GROMACS modification: Yes/No
Here post your question

Dear all,

I don’t have much experience with MD or Gromacs and I am experiencing a few issues with the normal mode analysis tool. I want to get the IR spectrum of water ice on silica. I used the structure from a previous equilibration and performed a few run to energy minimize to less then 0.0001 kJ/mol/nm. In a second calculation I run the nm analysis reading the full double precision trajectory from the previous energy minimization. Everything works fine (apparently) till the hessian diagonalization, there are a lot of negative eigenvalues (more than 300) with very high values I cannot really understand. In order to figure out what s going on, I performed a few tests with a reduced number of water molecules and a different force field, TIP3P. Everything works fine with TIP3P while with TIP4PQ/2005 (same for TIP4P) I can make it work only if my system has one water molecule. Anyone has any guess about what is happening with my system or what is wrong in my procedure?

Cheers,

Carmen

2

The only thing I can think of that is different is that TIP4P has virtual interaction sites. But those should be handled properly by NM.

I your Hessian indeed smaller than the number of atoms in the system by the number of water molecules (= the number of virtual sites)?

3

Thanks for your reply.

the size is exactly that expected counting only real atoms (3Nx3N with N number of real atoms).
Yes, I spent a few days going through previous queries and what I understood is that virtual sites should be no problem but still something is wrong with my system.