I agree with Petr. For a rigid body displacement the RMSD is not much use as a descriptor. Such
a displacement can be described by a translation along and rotation around a direction, or
rotation-translation matrices. These will allow someone else to reproduce the displacement,
but the RMSD will not. I liked the FIT program of Guoguang Lu for this type of calculation and
still use some C code based on it. Not sure where the original went though.
On Jun 20, 2012, at 9:31 AM, James Holton wrote:
> Because sometimes it is important to know how much the structure moved
> relative to the unit cell, such as before and after a round of
> refinement.
>
> -James Holton
> MAD Scientist
>
> On Tue, Jun 19, 2012 at 3:34 PM, Petr Leiman <[log in to unmask]> wrote:
>> Would anyone be kind enough to explain what kind of information does the
>> "RMSD" for two not-yet-superimposed structures transmit?
>>
>> If two structures are indeed identical but are apart spatially, then it is
>> more appropriate to calculate the COM and the inertia tensor for both
>> structures and report the displacement and the rotation (moleman or moleman2
>> do just that, right?). Reporting "RMSD" in this case does not seem to make
>> sense because the same RMSD describes an infinite number of spatial
>> configurations of the two structures.
>>
>> If two structures are not identical, one HAS to superimpose them, i.e. to
>> move all (or selected) atoms to be as close in space as possible and only
>> then calculate the RMSD for the superimposed or all atoms.
>>
>> Thank you in advance,
>>
>> Petr
>>
>> --------------------
>> Petr Leiman
>> EPFL
>> BPS-415
>> CH-1015 Lausanne
>> Suisse
>>
>> On Jun 19, 2012, at 5:04 PM, Claudia Millán Nebot wrote:
>>
>> Hello everyone :)
>>
>> I would like to know if it exist some tool that allows to calculate RMSD
>> between 2 pdbs that are identic, but just displaced in space. It should not
>> make a superposition, beause if this is the case it will just say that RMSD
>> is 0 .
>> I know is not such a difficult problem in terms of scripting, but i was
>> wondering if there are tools already.
>>
>> Claudia Millán ([log in to unmask])
>>
>> Institut de Biologia Molecular de Barcelona (IBMB-CSIC)
>>
>> Barcelona, Spain
>>
>>
>>
|