[phenixbb] "FOM" or HL coeffs from model?

Nathaniel Echols nechols at lbl.gov
Mon Mar 21 16:40:51 PDT 2011


On Mon, Mar 21, 2011 at 2:23 PM, Phoebe Rice <price at uchicago.edu> wrote:
> How can I get phenix to spit out not just PhiModel but an estimated FOM or fake HL coefficients (without experimental phases - just based on how badly my model agrees with FOBS)?  And I'd like the weights in their own personal mtz column(s), not just precombined in a "weighted phi" column.

It wouldn't surprise me if there is some command that already does
this that I don't know about.  However, it looks like we have almost
everything we need to do this in CCTBX, and we're only missing a
command to actually output the desired data.  I threw together a
script that appears to do the job, but I'm not 100% confident that the
output is correct.  I'll add it as "phenix.fmodel_phases" as soon as I
can verify that there aren't any stupid coding errors.  (You're not
the first person to ask for this, but it wasn't obvious to me before
how to write it.)  If there is an equivalent command in CCP4 that does
this, I'd recommend using that for now.

(I'm pretty sure that Resolve can do this too, but I don't know the
input keywords to use for running it without density modification.)

-Nat



More information about the phenixbb mailing list