You are here

NC atom type [Solved]

4 posts / 0 new
Last post
NC atom type [Solved]
#1

Dear all,

Apologies if this is a well-established feature or if answers to this or similar questions have already been posted in the forum; a search for 'atom type' did not yield a precise answer.

I want to use NC atom types for the pyridine ring nitrogen atom but this particular atom type (along with a few others) are systematically commented out in all Rosetta database files. My pyridine ring is actually functionalised so it is not exactly a bare-and-bones pyridine ring.

For example, in ${ROSETTA_DATABASE}/chemical/mm_atom_type_sets/fa_standard/mm_atom_properties.txt file:

N -0.200000 1.850000 -0.000100 1.850000 # defines 3 bond
#NC -0.600000 1.850000 -0.600000 1.850000 # from toppar_all22_prot_pyridines.str
NC2 -0.200000 1.850000 -0.200000 1.850000

This atom type has the appropriate charge and properties for my purpose but appears to be beyond reach without modifying each and every line where NC is listed across several files.

Before I go on and remove the comment signs from every one of those lines, I would like to ask whether there is any fundamental reason why I should not do it -- are those NC lines commented out for a good reason? Only convenience? Or has NC been superseded by other atom types I am unaware of (which could very well be the case)?

Thanks a lot in advance,

Fran

Post Situation: 
Tue, 2012-02-07 16:05
franfdez

I have forwarded this along to the only person who is likely to know (Doug Renfrew). He'll hopefully reply in a day or two.

Wed, 2012-02-08 08:03
smlewis

Hi franfdev,

It should be fine to uncomment the NC line in the mm_atom_properties.txt file and the lines that contain NC in mm_torsion.txt

The reason it is commented out is that none of the amino acids used it and each additional atom type adds a little to the memory foot print of Rosetta because we pre-compute some energies and store them as large multidimensional arrays in the etable.

Doug

Wed, 2012-02-08 08:25
renfrew

Hi smlewis & renfrew,

Thanks so much for the prompt replies and the explanation about commented out atom types and the effect of uncommenting them on memory usage.

Fran

Wed, 2012-02-08 11:02
franfdez