On Thu, Jun 16, 2011 at 2:11 PM, Xun Lu <span dir="ltr">&lt;<a href="mailto:[log in to unmask]"><a href="/cgi-bin/wa-jisc.exe?LOGON=A3%3Dind1106%26L%3DCCP4BB%26E%3Dquoted-printable%26P%3D14855737%26B%3D--e0cb4e8878911ac85b04a5dae78e%26T%3Dtext%252Fhtml%3B%2520charset%3DISO-8859-1%26pending%3D" target="_parent" >[log in to unmask]</a></a>&gt;</span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
      I have a 3.2A dataset for a protein-DNA complex. The protein is a homodimer, and the DNA is almost palindromic (except one base pair in the middle and two or three base pairs at both two ends). It is my first time solving structures, and unfortunately the resolution is low. No body in our lab has used ccp4 or phenix, so I am really frustrated as a second year student. <br>

      I mainly used ccp4. So far, the best R/Rfree I got is 0.27/0.34.  I went to the crystallography meeting, and people suggested me to rely more on geometry. I remember I got a DNA restraints file and a refmac script from someone on this mailing list, and that really helped (otherwise the DNA base pairing will be weird). Can someone tell me how to restraint the protein (helix)?  I can only see a few side chains in the helix, so it&#39;s hard to say whether the registration of the helix is correct or not. Maybe my high R-free is due to the helix in the wrong position?<br>
</blockquote><div><br></div><div>Refmac uses a syntax like this to define custom restraints (apologies to Garib if I&#39;m screwing this up):</div><div><br></div><div><div>exte dist first chain B residue  5 atom  N   second chain B residue  1 atom  O   value 2.900 sigma 0.05</div>
</div><div><br></div><div>Phenix can generate the annotations for you - either automatically if you run phenix.refine, or by doing this:</div><div><br></div><div>phenix.secondary_structure_restraints model.pdb format=refmac</div>
<div><br></div><div>The high R-free could be due to any number of problems, but the structure is obviously a little bit overfit (that is, the difference in R-work and R-free is too large); adding NCS restraints and TLS usually helps this, sometimes a lot.  You may have an easier time interpreting sidechain density if you try B-factor sharpening the maps, which both Refmac and Phenix (in addition to CNS) can now do.</div>
<div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
      People also suggested me to include NCS and TLS in the refinement, but I don&#39;t know how to. For NCS, I should define a region that are the same in both monomers? Should I use tight or loose restraints?  For TLS, I don&#39;t have a clue.<br>
</blockquote><div><br></div><div>For TLS, you could start with one group per chain.  You can get a more fine-grained division by running the TLSMD web server (<a href="http://skuld.bmsc.washington.edu/~tlsmd/">http://skuld.bmsc.washington.edu/~tlsmd/</a>) or phenix.find_tls_groups.</div>
<div><br></div><div>NCS restraints are a little trickier and others will probably have better advice for you - however, if there is NCS present, you should absolutely use these restraints at 3.2A resolution.</div><div><br>
</div><div>-Nat</div></div>