Sorry if the original got through, but my server keeps telling me it was
unable to send this, so I've sent it again.  My apologies if it was
originally delivered

-Haydyn

-------- Original Message --------
Subject: rhombicity confusion
Date: Fri, 25 Apr 2003 14:44:15 +1000
From: haydyn mertens <[email protected]>
To: [email protected]



Hi guys,

I'm currently implementing RDCs into my NMR structure calculations for
a small protein domain.  The confusion I've stumbled into surrounds
the definition of rhombicity.

Q. When is the rhombicity in the range 0 - 1?

Q. When is the rhombicity in the range 0 - 2/3?

I'm making the assumption that estimations of axial and rhombic terms
from a powder-pattern distribution (histogram) yield the rhombicity
"R", as defined by Clore et al., with a max. R of 2/3.  My data yields
a value of ~0.6 when obtained by this method.  However, when applying
the maximum-likelihood method (Warren & Moore, JMR, 149, 271-275
(2201)) I obtain "most-likely" R values of 0.74 +- 0.03 .

I have also successfully run the ISAC method (Sass et al., JBNMR, 21,
275-280, 2001) yielding a rhombicity  of 0.77 +- 0.02 . In this case
it is my understanding that the max. R is 1, due to the definition of
rhombicity as: R=(Axx - Ayy)/Azz , and the equation for dipolar
coupling used:

D=Azz/2*[(3cos2(theta) -1) + R*(sin2(theta)cos2(phi))

So in this case I would multiply this value of R by 2/3 to use it in a
SANI run in CNS.  I think?

In XPLOR-NIH and CNS the equation for dipolar coupling used is:

D=Da*[(3cos2(theta) -1) + 3/2*R*(sin2(theta)cos2(phi))]

with R=Dr/Da, and Da=1/3*(Dzz-(Dxx+Dyy)/2), Dr=1/3*(Dxx-Dyy).
Using Dxx+Dyy+Dzz=0, the rhombicity R is:

R= 2/3*(Dxx-Dyy)/Dzz , with a max. value of R=2/3

So, is R always (Dxx-Dyy)/Dzz ? and when quoted as "R" always between
0 and 1?
Or is R sometimes quoted as "R" with the author meaning
R=2/3*(Dxx-Dyy)/Dzz ?

If any of you guys can set me straight on this I'd really appreciate
it, as I'm spending a lot of time trying to figure this out.  In
particular, if anyone has used the maximum liklihood method to
estimate the value of R I'd love to know whether values greater than
0.66 are reasonable.

Thanks in advance,

Haydyn

-- 
-------------------------------------------
Haydyn Mertens
PhD student
Department of Biochemistry
University of Melbourne
Australia

Reply via email to