A biochemist friend asked for examples of cases were a protein was
co-crystallized with or soaked in a ligand that bound in the wrong place -
say, because the ligand used wasn't quite the right one or because other
important ligands were absent.
I'm sure such examples are out there, especially w
Thanks!
Phoebe
At 05:56 PM 1/22/2007, you wrote:
Hi Pheobe,
I remember an interesting paper that described how a structure revealed a
surprising role the buffer was playing in inhibition:
The 1.20 A resolution crystal structure of the aminopeptidase from
Aeromonas proteolytica complexed with
I would take a well-ordered helix out of your
allegedly-correctly-placed model (BEFORE doing any refinement other
than rigid body), phase a 2Fo-Fc and an Fo-Fc map with it, and see if
the helix shows up. If it does, there is some reality in your
solution. If it doesn't, start over from scratc
TeV is great but it isn't always a magic bullet for producing native
N-termini: we've experimentally determined the obvious, that if you
bury part of its recognition site in secondary structure, it cleaves
very very slowly. We tried this on a protein where M1 is cleaved in
vivo and aa#2 is th
Isn't automatically included fabricated data for missing reflections
a really bad idea for anisotropic data where most reflections are
"missing" at high resolution? Shouldn't there be a big flashing red
flag alerting the user to what's been done?
Phoebe
At 01:22 PM 3/26/2007, Edward A
Dear Kumar,
One often has to try duplexes with many different
ends before getting decent crystals (e.g. 18 for
one project in my lab, even more for others).
Depending on your Kd, you might find that your
complex falls apart during gel filtration.
How long are your oligos? Gel purification
som
Every modern structure should be "above average" by procheck criteria.
For modest resolution structures, the availability of ncs restraints
should have a big effect on one's Ramachandran expectations: I've
found ncs restraints at 3-3.5A do wonders for the Ramachandran plot
even when they don't
Hi,
Did you do rigid body refinement or regular minimization? If over
half your structure is missing, anything besides rigid body
refinement will probably just bias the phases toward "nothing" in the
rest of the asymmetric unit.
I've had correct molrep solutions with R-factors that high, s
A comment from my collaborator's student suggests a partial
answer. This afternoon he happened to say "but of course the
reviewers will look at the model, I just deposited it!". He was
shocked to find that "hold for pub" means that even reviewers can't
access the data. Can that be changed?
I don't think all journals have that policy, and even so, making the
reviewers specifically request the data implies to the reviewees that
somebody out there doesn't trust them.
You shouldn't have to insult the authors in order to do a proper
reviewing job - you should just be able to download
While we're still on the subject of good model-building habits and
reviewing pitfalls, I've been shocked to download a couple of
structures recently that seem to have solvent channels chock full of
allegedly ordered water (many "layers" deep, and not exactly at 0.5A
resolution). To any new stu
Hi,
- I think someone already pointed out that you should try P6522.
- check that it isn't really a twinned P61 or P65 with 2 per asymmetric unit
- buy DNA with BrdU and some more with IdU, and/or grow your protein
in SeMet - at your resolution, the more "real" phase info the better!
Pho
I don't see a concise list of changes there?
At 07:34 AM 10/9/2007, Eleanor Dodson wrote:
Did you go to this web site?
http://remediation.wwpdb.org/downloads.html
It has the new dictionary stuff
Eleanor
James Stroud wrote:
Hello All,
I noticed some things different about the PDB today causi
Some proteases are metal-dependent, and inhibitors for those aren't
Ni-column-compatible. "We" (meaning my students) found that it helps
to (1) work very quickly and (2) put EDTA into the fraction collector
tubes before eluting from the Ni column.
At 06:22 AM 11/4/2007, Vijay Kumar wrote:
H
All statistics aside, I never believe a molecular
replacement solution until it produces an Fo-Fc
map with peaks for something that should have
been there but wasn't in the model.
If you know roughly where the DNA should bind,
you could try making a solvent mask that includes
that region (pl
We've been having a discussion in the lab about whether or not
middle-sized PEGs such as 4000 can be expected to serve as
cryoprotectants (and if not, why certain commercial kits are
formulated the way they are). Can anybody shed some light /
references on the question of the size of PEGs vs.
Many thanks to all who replied. The answers were remarkably varied -
see below.
My own two bits worth - vitrification of mother liquor doesn't always
lead to a nice, low-mosaicity, ice-free crystal freeze in our hands,
although we're not willing to sacrifice a statistically significant
number
This has probably been discussed before, so apologies in advance.
We're eyeing a protein that has a probable C4 Zn finger in the
middle. The collaborators who are nicely going to PCR it up want to
know if we'd like it with or without a His tag.
Is it a bad idea to co-mingle Zn-binders and Ni co
High R "merges" with no reasonable excuse can certainly be a useful
red flag during data processing (along with the % of observations
rejected, which I've never had a reviewer request).
Which brings up the point that one reasonable excuse is anisotropy -
high Rs for merging random observations
Rumor also has it that more than one "bad" codon in a row, particular
near the beginning, can be extra-bad. For one small, A/T rich
protein, we simply "fixed" a pair of bad Arg codons near the
N-terminus at the same time as recloning it, by using the N-terminal
cloning primer to do the mutagen
What is the resolution? Are you using ncs restraints (you probably
should, on the coordinates but not on the Bs)? How does your
Ramachandran plot look?
You might tighten the geometry even more. Aside from all theoretical
arguments about how big the rmsd's should be, if they're rather
loose, t
Rotational near-crystallographic ncs is easy to handle this way, but
what about translational pseudo-symmetry (or should that be
pseudo-translational symmetry)? In such cases one whole set of spots
is systematically weaker than the other set. Then what is the
"theoretically correct" way to cal
An added benefit of EDTA is that it inhibits some proteases - for one
of our wimpier proteins, spiking each fraction collector tube with a
little EDTA before running the Ni column really helped reduce keep
the sample in one piece.
Phoebe
At 01:18 PM 2/19/2008, Sophia Tsai wrote:
Hi,
Just beware that changing how you break the cells open can
change the average size of chromosome chunks, which can
change how DNA binding proteins behave in the lysate.
Original message
>Date: Mon, 3 Mar 2008 15:21:15 +
>From: Mads Gabrielsen <[EMAIL PROTECTED]>
>Subject: [ccp4b
Didn't that trick very successfully lower the
R-factors of the completely wrong models that led
to the Great Pentaretraction? Unless you have
stunningly high resolution, beware.
Phoebe
At 10:13 AM 3/28/2008, you wrote:
Some time ago I've heard about the idea of proposing
an ensemble of m
In the end, we're solving all these structures because we believe (or
at least hope) that they'll be useful for understanding
biology. That means that biologists should be able to understand
what we deposit.
When I've tried to teach undergraduates "what to make of" structural
models, I find I
Crystallization and structural biology of soluble and membrane
proteins using microfluidics. This NIH-funded project at the
University of Chicago aims to develop technologies for
crystallization and structural analysis of proteins in
nanoliter volumes (see PNAS 2006 103: 19243-19248). See
http://is
Dear all:
I got a dataset at 2.8 angstron. I have tried several ways such as phaser,
MRBUMP,BALBES,but still can't solve the
data,which means I have to edit my model. Maybe I'd better cut it off or delete
the water or loop. I really have no
idea,as it is my first time to do such things, I alwa
28 matches
Mail list logo
