The increase in RFZ is relatively small and not entirely unexpected.
While hydrogens only contribute one electron (as opposed to carbon (6),
nitrogen(7), oxygen(8) and sulfur (16)), there are many hydrogens (in
fact, almost as many as all the other atoms combined). For instance, in
lysozyme you have (I used 3a67)
613 carbons, 193 nitrogens, 185 oxygens, 10 sulfurs and
959 hydrogens.
Electron-wise this translates into ~13% of electrons being supplied by
hydrogens. I certainly do not know how many hydrogens your model
contains, but from the snippet you present it seems that you may have
full set. Also, the "problem" will be exacerbated by the fact that you
have B-factors of the hydrogens set to zero. Assuming that your
non-hydrogen atoms have B-factors set to, say, 20, this will
approximately double the relative contribution of hydrogens at the
higher resolution end. True, contribution of hydrogens is not exactly
additive, but it's probably not surprising to see roughly 5 percent
change in RFZ. What is very interesting is that it seems to be general
*improvement*, so maybe we ought to include hydrogens when doing MR. I
don't know if PHASER generates riding hydrogens automatically.
I think that characterize hydrogens as "transparent" to x-rays is
somewhat misleading. There are plenty of examples where hydrogens are
seen in ultrahigh resolution structures. Modern refinement programs all
use riding hydrogens and it improves models. More precise statement is
that at not-ultrahigh (non-atomic) resolution (i.e. worse than 1.2A)
optical resolution is too low to distinguish two atoms separated by ~1A
(typical value of X-H covalent bond).
Cheers,
Ed.
On Tue, 2010-01-12 at 17:42 +0900, Francois Berenger wrote:
> Hello,
>
> I have 4 molecules.
> If I use Phaser's AUTO_MR with all default parameters on them,
> I get the following scores:
>
> no_modif remove_H
> molecule RFZ TFZ RFZ TFZ
> 1 6.9 9.6 7.0 9.4
> 2 5.2 5.9 5.4 6.2
> 3 4.2 4.5 4.7 5.2
> 4 4.9 6.8 5.2 6.7
>
> Hence, here are my existential questions, given that I heard
> many times "H is transparent for X-rays":
> * should I always remove H before running Phaser?
> * why?
>
> Here are a few lines of one of my PDBs containing H, in case it is
> misinterpreted by Phaser:
> ---
> ATOM 6 1H ALA A 1 18.629 -3.443 12.782 1.00 0.00
>
> ATOM 7 2H ALA A 1 18.427 -2.484 13.918 1.00 0.00
>
> ATOM 8 3H ALA A 1 17.496 -2.462 12.742 1.00 0.00
>
> ATOM 9 HA ALA A 1 19.811 -0.866 13.013 1.00 0.00
>
> ATOM 10 1HB ALA A 1 21.071 -1.469 10.982 1.00 0.00
>
> ATOM 11 2HB ALA A 1 21.038 -2.809 12.152 1.00 0.00
>
> ATOM 12 3HB ALA A 1 19.983 -2.852 10.720 1.00 0.00
>
> ---
>
> Thanks a lot,
> Francois.
--
Edwin Pozharski, PhD, Assistant Professor
University of Maryland, Baltimore
----------------------------------------------
When the Way is forgotten duty and justice appear;
Then knowledge and wisdom are born along with hypocrisy.
When harmonious relationships dissolve then respect and devotion arise;
When a nation falls to chaos then loyalty and patriotism are born.
------------------------------ / Lao Tse /
|
