Page 1 of 1

PEQM polarazibility parameters for DNA

Posted: 24 Feb 2016, 10:27
by Zeljka.San
Dear all,
I am new to using PEQM, and I am trying to make potential.inp file. Does someone here have experience with using DNA as surrounding molecule?
I could use some help, because my surrounding molecule is DNA (just a few basis pairs) and I was wondering is there some library I can use to get the parameters for @POLARIZABILITIES from?
Thank you,
Željka

Re: PEQM polarazibility parameters for DNA

Posted: 24 Feb 2016, 11:53
by magnus
Unfortunately, we do not have any libraries yet. We are currently working on a library of standard parameters for a series of solvents, proteins, lipids and DNA, but these are not ready yet. Alternatively, you could use Thole's model (see http://pubs.acs.org/doi/pdf/10.1021/jp980221f) which is also used in some force fields such as AMOEBA and Amber's ff02. Possibly you could use the polarizabilities from one of those force fields but you have to be careful with their definitions of excluded interactions. Also you most likely have to damp the interactions between the induced dipoles (can be done using the .DAMP INDUCED keyword).

Normally what we would do is to calculate multipoles and polarizabilities using a property distribution scheme such as LoProp. For large molecules we combine it with a fragmentation scheme. If you want more details concerning this approach you are welcome to contact me.

Re: PEQM polarazibility parameters for DNA

Posted: 24 Feb 2016, 13:21
by Zeljka.San
Dear Magnus,

thank you very much for your fast answer. We would appreciate very much if you could supply us with additional information how to calculate multipoles and polarizabilities using a property distribution scheme such as LoProp. This has been mentioned in your article Molecular properties through polarizable embedding (AdvQuantumChem,61,107) in A hierarchy of force fields.
Of course we don't know what is easier for us to use this proposal or to adapt polarizabilities from Amber ff02.
We would appreciate your advice and help very much.
Best regards,
Željka
(Group of Vlasta Bonacic-Koutecky)

Re: PEQM polarazibility parameters for DNA

Posted: 24 Feb 2016, 15:57
by magnus
What is most suitable/easiest depends on whether you have one structure or many, and the size of the individual molecules that you need to calculate parameters for. If you have many structures, e.g. snapshots from a MD simulation, then it can become costly to calculate the parameters for each of them, though this will give you the optimal parameters for each structure. If one or more molecules are very large then you may benefit from using a fragmentation scheme, such as the molecular fractionation with conjugate caps (see http://scitation.aip.org/content/aip/jo ... /1.1591727 and http://pubs.acs.org/doi/abs/10.1021/jp8073514 for details), to obtain the parameters. The multipoles and polarizabilities of each molecule (or fragment) can for instance be calculated using the LoProp scheme using MOLCAS, or DMA using CamCASP (see their individual documentation for details).