OK. I guess I was thinking of subtracting 2fo-Fc maps, which would give essentially (Fo1-Fo2) coefficients. But coming out of phenix it will be at least 2mFo-DFc, with different m's and D's for the two datasets. Makes it rather more complicated.
eab

On 05/12/2015 11:58 AM, Philip Kiser wrote:
> Good points. But also, the traditional isomorphous difference map is not as susceptible to model bias issues compared to a real space difference map (or a "vector" isomorphous difference map).
>
> On Tue, May 12, 2015 at 11:37 AM, Edward A. Berry <[log in to unmask] <mailto:[log in to unmask]>> wrote:
>
>
>
>     On 05/12/2015 09:46 AM, Philip Kiser wrote:
>
>         crystallogrphy,
>
>         If you did everything correctly, you should be able to directly calculate the difference map in Coot from the MTZ file generated by PHENIX. You would need to open it with "Open MTZ" rather than "Auto open MTZ" and then select the appropriate map coefficients. Note that what Tim described (a real space difference map) is technically different from the traditional isomorphous difference map where the SF amplitudes of the two data sets are subtracted prior to the map calculation.
>
>         Philip
>
>
>     Should be the same, due to linear property of the Fourier transform, EXCEPT for possible scaling differences and treatment of missing data.  This latter could be significant, since presumably the traditional methods omits reflections unless present in both datasets, whereas in Tim's method each map will be made with all data available in that dataset. As a result, the largest coefficients in Tim's difference map might be the reflections which are absent in one map and present in the other.  But if both have good completeness, it  may not matter.
>
>     eab
>
>
>
>         On Tue, May 12, 2015 at 9:24 AM, Tim Gruene <[log in to unmask] <mailto:[log in to unmask]> <mailto:[log in to unmask] <mailto:[log in to unmask]>>> wrote:
>
>              Dear crystallogrphy,
>
>              you can open the PDB file with Coot, then use File -> Open MTZ to read one
>              mtz-file after the other into Coot. Coot will offer to generate the map
>              coefficients from the previously loaded PDB file.
>
>              Once you have both maps, you can go to Extensions -> Maps -> Make a Difference
>              Map to create the difference map.
>
>              In order to see both positive and negative density, you choose Extensions ->
>              Maps -> Set Map is a Difference Map.
>
>              Regards,
>              Tim
>
>              On Tue, May 12, 2015 at 08:43:45PM +0800, crystallogrphy wrote:
>               > Hi,
>               >
>               > I am trying to generate an isomorphous difference map with phenix GUI. I
>               > input reflection 1 and 2 with mtz format and a PDB model, I got an MTZ
>               > format map file. Because I want to load it to coot or pymol with negative
>               > and positive density, I need to transform this mtz map to ccp4 map.
>               > However, FFT map cannot work.
>               > Does any know how can I get a ccp4 format maps with positive or negative
>               > density?
>               >
>               > Thanks!
>
>               > _______________________________________________
>               > phenixbb mailing list
>               > [log in to unmask] <mailto:[log in to unmask]> <mailto:[log in to unmask] <mailto:[log in to unmask]>>
>               > http://phenix-online.org/mailman/listinfo/phenixbb
>               > Unsubscribe: [log in to unmask] <mailto:[log in to unmask]> <mailto:[log in to unmask] <mailto:[log in to unmask]>>
>
>
>              --
>              --
>              Dr Tim Gruene
>              Institut fuer anorganische Chemie
>              Tammannstr. 4
>              D-37077 Goettingen
>              phone: +49 (0)551 39 22149 <tel:%2B49%20%280%29551%2039%2022149> <tel:%2B49%20%280%29551%2039%2022149>
>
>              GPG Key ID = A46BEE1A
>
>
>              _______________________________________________
>              phenixbb mailing list
>         [log in to unmask] <mailto:[log in to unmask]> <mailto:[log in to unmask] <mailto:[log in to unmask]>>
>         http://phenix-online.org/mailman/listinfo/phenixbb
>              Unsubscribe: [log in to unmask] <mailto:[log in to unmask]> <mailto:[log in to unmask] <mailto:[log in to unmask]>>
>
>
>
>
>         _______________________________________________
>         phenixbb mailing list
>         [log in to unmask] <mailto:[log in to unmask]>
>         http://phenix-online.org/mailman/listinfo/phenixbb
>         Unsubscribe: [log in to unmask] <mailto:[log in to unmask]>
>
>