On Tue, Mar 27, 2012 at 9:22 AM, John Rose
I downloaded the latest PHENIX build which has the "Find difference map peaks and holes" option in the GUI.
However, when I run it and look at the anomalous peak heights the peaks do not match the density in the anomalous difference Fourier map created using "Calculate Maps".
To generate the anomalous difference Fourier map I used a SCALEPACK file containing the anomalous pairs and the corresponding refined PDB file with the sulfur atoms omitted. The data labels used were I(+), sig(+), I(-), sig (-).
The anomalous difference Fourier gave the positions of the missing sulfur as expected. However, using the same approach (and files) with the "Find difference map peaks and holes" . The resulting peaks in the GUI or PDB file produced do not correspond to any of the expected sites.
Sorry, my description of the program wasn't entirely accurate. The default settings (inherited from the existing peak search in phenix.refine) are tuned to find *unmodeled* peaks and holes, without significant overlap with existing atoms. (Water molecules are handled separately.) Setting the "Min model peak dist" to 0 will probably allow it to find peaks on top of atoms. This may be somewhat misleading - it would not be difficult to change the default settings for the program if this would make it more useful. -Nat