Dear Justin,

Thanks for testing it out. As I see it, it *ought* to work so that you set 
the protonation states you want in the molecule, and the ARIA export then 
exports them correctly. Seeing that it does not, we need to think about 
changing it at some suitable time.

CCPN coleagues take note.

Yours,

Rasmus

---------------------------------------------------------------------------
Dr. Rasmus H. Fogh                  Email: [log in to unmask]
Dept. of Biochemistry, University of Cambridge,
80 Tennis Court Road, Cambridge CB2 1GA, UK.     FAX (01223)766002

On Wed, 26 Jan 2011, Justin Lecher wrote:

> On 25/01/11 13:31, Rasmus Fogh wrote:
> Dear Rasmus,
>
> So I played around a bit. Changing the protonation state in analysis
> wasn't that difficult. I already found this some time ago.
> But this doesn't influence the Aria runs. Brian was right, the only way
> to get it right in ARIA is to use the HIS patches, which works really
> good. I see the expected result, which makes much more sense ( == fits
> to the experimental results :) ).
> Currently I am trying what happens, if I also change those histidines,
> which are exchanging on both side with the water to statically single
> protonated ones in Aria. If this gives huge changes I will report it.
>
> Thanks justin
>
>
>> Dear Justin,
>>
>> Have you tried to set the protonation state correctly in Analysis and see
>> if ARIA got the correct sequence? I had a quick look but I do not have
>> reasonnable test data to use - that should be easier for you.
>>
>> Technicaly what we use is not called 'chain state' - that is something
>> else - but we do indeed treat all protonation states of His as the same
>> ChemComp with the same ccpCode, and use another attribute (the
>> 'descriptor') to distinguish. The Aria set-up ought to read the necessary
>> data from the sequence, and use it to output the correct CNS residue code.
>> It would have to be tested, though.
>>
>> The way to set it up is to go to M:Molecule:Molecules {Add Sequence} and
>> enter a sequence with 'Destination Molecule':'New, and 'Destination Mol
>> System':'<None>'. The go to the {Sequences} tab and change the 'Descriptor
>> and Stereochemistry' setting for yor Histidine(s) to have the correct
>> protonation state. After that you go to the {Chains} tab and create a
>> MolSystem using [Make Chain From Template] seting Mol System: '<New>' and
>> using the new molecule. To get your NMR data to the new moledcule, use
>> M:Assignment:Copy Assignments {Between Molecule Chains}. That ought to get
>> you to a state where you have a non-protonated His in your system, and you
>> can run Aria and see how it behaves. Hopefully you get the right result.
>>
>> Doubly protonated His is not always correct, but setting from pH would
>> also sometimes be wrong. Getting the right protonation state requires user
>> input, especially for the kind of pH you tend to do NMR on. We could maybe
>> make the process more streamlined than the above (assuming that it
>> works), but we could not really do it for you.
>>
>> Hope this helps,
>>
>> Rasmus
>>
>> ---------------------------------------------------------------------------
>> Dr. Rasmus H. Fogh                  Email: [log in to unmask]
>> Dept. of Biochemistry, University of Cambridge,
>> 80 Tennis Court Road, Cambridge CB2 1GA, UK.     FAX (01223)766002
>>
>> On Thu, 20 Jan 2011, Justin Lecher wrote:
>>
>>> On 20/01/11 10:53, Brian Smith wrote:
>>>> Just checked back through old versions of the CNS/XPLOR forcefield and it
>>>> has always been so. I guess this is partly historical - NMR structures
>>>> were typically recorded at lower pH if they could be - and partly a good
>>>> default setting as electrostatics didn't used to be used for refinement.
>>>>
>>>> Even if you set a different (partial?) protonation state in analysis, I
>>>> would be surprised if ARIA paid attention to what you'd done if the
>>>> residue code was still HIS in the sequence though. Looks as if the data
>>>> model is intended to deal with protonation state via "chain state"s rather
>>>> than alternative residue types.
>>>>
>>>> If you know that your HIS are likely to be singly protonated (experimental
>>>> evidence, pH, etc.) and can either determine or want to take a punt on
>>>> which N is protonated then you should use the HISD or HISE patches in your
>>>> ARIA setup.
>>>>
>>>> I reckon that you are right that it would be nice to be able to set your
>>>> experimental pH, per residue pKas, and have CNS/ARIA cope with that,
>>>> because as you say it will likely have an impact on the water refinement
>>>> stage. Probably not a high priority in the grand scheme of things though!
>>>>
>>>> Brian Smith
>>>>
>>>
>>> Hi Brian,
>>>
>>> it might be historical, but this doesn't make it correct.
>>> In my case I defined an hbond to a NE2 of a HIS which doesn't work if is
>>> protonated. The donor proton and HE2 just clash.
>>> Thanks for pointing to the HIS-patches in Aria. I never payed attantion
>>> to it. Am I correct, HISD means proton on ND1 and HISE means prton on HE2 ?
>>>
>>> Justin
>>>
>>> --
>>> Justin Lecher
>>> Institute of Complex Systems
>>> ICS-6 Structural Biochemistry
>>> Research Centre Juelich
>>> 52425 Juelich, Germany
>>> phone: +49 2461 61 2117
>>>
>>>
>>>
>
>
> -- 
> Justin Lecher
> Institute of Complex Systems
> ICS-6 Structural Biochemistry
> Research Centre Juelich
> 52425 Juelich, Germany
> phone: +49 2461 61 2117
>
>
>