On Tue, Jul 12, 2011 at 12:51 PM, <span dir="ltr"><<a href="mailto:lesser@helix.nih.gov">lesser@helix.nih.gov</a>></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;">
Also what surprises me is that when I run phenix.xtriage or phenix and then<br>
"reflection tools" xtriage, I get different results.<br></blockquote><div><br></div><div>Yeah, this is a bug - getting Xtriage output to display nicely in the GUI was non-trivial, and in a few cases I misinterpreted which data arrays were relevant. It'll be fixed in the next nightly build.</div>
<div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"> as a big fan of phenix, I still wonder if there is a way to improve the<br>output of xtriage. Even after looking at it for a number of years now, I<br>
manage to confuse myself. I for one would prefer clear statements over<br>probability scores. I never know whether a high number is good or bad (or<br>what the scale is!) because sometimes a score reflects the probability<br>
whether something is true or whether an assumption is wrong. Would it not be<br>possible to print likelihood of a space group scaled to 100%?</blockquote><div><br></div><div>We agree that it needs to be made much more intuitive - I have tried to clarify some issues in the GUI but I'm limited to those parts I actually understand. Peter and I probably need to spend a couple of weeks revising and cleaning up the output, but as usual this has to compete with about a dozen other projects...</div>
<div><br></div><div>-Nat</div></div>