On Mon, May 17, 2010 at 10:53 AM, xaravich ivan <span dir="ltr">&lt;<a href="mailto:xaravich.ivan@gmail.com" target="_blank">xaravich.ivan@gmail.com</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">

So I have quite a few regions build by the SA_omit map of phenix. But How do I go about looking at these maps. What am I looking for here. Should I look at the regions where the occupancy was set zero in all these outputs and find if I could build some thing from one of the maps? I guess I am asking, How do I use this information to move forward in refinement? There are 24 regions map and .pdb output in the /OMIT directory.<br>

</blockquote><div><br></div><div>Ignore these - only the final map is useful.  If you are generating a composite omit map, you want to make sure that what you&#39;ve built still matches the density reasonably well.  Obvious things to look at:</div>

<div><br></div><div>- flexible or missing loops</div><div>- ligands</div><div>- anywhere the mFo-DFc map shows large blobs at +/- 3sigma</div><div>- any outliers flagged by validation (both geometry and real-space correlation)</div>

<div><br></div><div>I don&#39;t usually find it productive to go through the structure residue-by-residue, but this may depend on the data and model quality.  What is your resolution and current R-free?</div><div><br></div>

<div>(If you are just omitting part of the structure, then you only need to look at the map around those atoms - the rest won&#39;t be much different.)</div><div><br></div><div>-Nat</div></div>