[phenixbb] Partial Model for Mol Rep.

[Nathaniel Echols]

On Fri, May 2, 2014 at 3:48 PM, Yuri <yuri.pompeu at ufl.edu> wrote:

> I am trying to solve a structure (to 2.15 Ang). I am pretty sure about one
> of the domains fold but it only covers residues 26-250 while the whole
> protein is 470aa.
> Could I attempt to use this small part of the protein to phase my data


probably, if it's actually reasonably close to your target protein, and not
something with 25% sequence identity.

and hopefully start building?
>

Very difficult.  The density modification in AutoBuild will certainly help,
but even with higher resolution the maps for these severely incomplete
structures are hard to interpret.  If it is helical you might have an
easier time, because you could try docking in secondary structure - fitting
individual residues is unlikely to work.  (I suspect this is somewhat
dependent on solvent content, but most of the structures I've tried this
with were all around 50% or lower.)

Also, if you have reasonably good anomalous signal for endogenous sulfur,
this may help you identify the unbuilt Met sites even if the other maps are
unclear.

ps. Selenomet protein will be expressed soon..


The good news is that if MR works, the rest of the phasing process should
be a breeze once you have (reasonably good) SeMet data.

-Nat
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20140502/0de6f05c/attachment.htm>