<html>
<head>
<meta http-equiv="Content-Type" content="text/html;
charset=windows-1252">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Hi Ruki,<br>
<br>
<blockquote type="cite" cite="mid:1510670976023.17895@mbg.au.dk">
<meta http-equiv="Content-Type" content="text/html;
charset=windows-1252">
<style type="text/css" style="display:none"><!-- p { margin-top: 0px; margin-bottom: 0px; }--></style>
<p>I have been playing around with phenix.real.space.refine as
part of a small pipeline I am making for EM structures.<br>
</p>
<p><br>
</p>
<p>Testing various deposited EM�structures and their corresponding
maps,�within�my pipeline, I sometimes run�in to the below error:<br>
</p>
<p><br>
</p>
<p>RuntimeError: Bond distance > max_reasonable_bond_distance:
79.8663 > 50:<br>
</p>
</blockquote>
<br>
I'm trying to chase this down and eliminate. So if you come across a
case that can be reproduced please send it to me and I will start
from there.<br>
<br>
For now simply set "max_reasonable_bond_distance=999" for all your
refinements that will mask but not fix the problem. However, this
should not affect the refinement results.<br>
<br>
Make sure you are using the latest Phenix version.<br>
<br>
<blockquote type="cite" cite="mid:1510670976023.17895@mbg.au.dk">
<p>
</p>
<p>Followed by a list of distances between, what I am assuming is,
atom numbers and the actual distance.<br>
</p>
<p><br>
</p>
<p>I can get past it simply�by changing
the�max_reasonable_bond_distance�for the run�to, in this case,
80.<br>
</p>
</blockquote>
<br>
This happens because the procedure optimizes weights by trying a
large sample of possible weight values. For some values restraints
may be too weak and atoms run one onto another, thus creating a huge
clash and strong repulsion that stretches the bond. This leads to
the crash.<br>
<br>
<blockquote type="cite" cite="mid:1510670976023.17895@mbg.au.dk">
<p>
</p>
<p>But I tend to see, that the final�results gets
significantly�worse when I use this bypass.<br>
</p>
<p>This is based on very similar�models going through the
pipeline, where the ones which yields the error, scores much
worse�when using the above mentioned bypass, compared to the
models that showed no max_reasonable_distance_error. <br>
</p>
</blockquote>
<br>
Again, I'm very keep to investigate concrete reproducible cases. So
if you have anything to share that would be most helpful!<br>
<br>
<blockquote type="cite" cite="mid:1510670976023.17895@mbg.au.dk">
<p>
</p>
B: Any ide what I can do, to avoid�running�into this bond
distance�problem?<br>
</blockquote>
<br>
I'd say help us to identify the issue so that we fix it and then you
don't run into it any more.<br>
<br>
All the best,<br>
Pavel<br>
<br>
</body>
</html>