Dear Phenixers, Prompted by Randy Read, Kevin and I have agreed that Coot should be able to auto-open MTZ files with columns labelled groupname.F_phi.F and groupname.F_phi.phi and groupname.DF_phi.DF, groupname.DF_phi.phi for difference maps for any value of groupname, of course. (Any weights should be applied to Fs before writing the file). We were wondering what the Phenix community/developers thought about that idea (there's not much point in adding this feature if you don't agree). Thanks, Paul.
Hi Paul,
Prompted by Randy Read, Kevin and I have agreed that Coot should be able to auto-open MTZ files with columns labelled
groupname.F_phi.F and groupname.F_phi.phi
and
groupname.DF_phi.DF, groupname.DF_phi.phi for difference maps
for any value of groupname, of course.
(Any weights should be applied to Fs before writing the file).
We were wondering what the Phenix community/developers thought about that idea (there's not much point in adding this feature if you don't agree).
I would be great to have this feature, but I'm not familiar with the groupname.DF_phi.DF syntax. Our mtz tools work via "make a new crytal", "make a new dataset", "make a new column", "add data to the column". Can this be mapped to your dotted syntax? Do you have an example mtz file that I could follow? Ralf
On Jan 23, 2010, at 2:55 PM, Paul Emsley wrote:
Dear Phenixers,
Prompted by Randy Read, Kevin and I have agreed that Coot should be able to auto-open MTZ files with columns labelled
groupname.F_phi.F and groupname.F_phi.phi
and
groupname.DF_phi.DF, groupname.DF_phi.phi for difference maps
for any value of groupname, of course.
I'm not sure I understand - what do the fields represent? From the user perspective, I think it's only absolutely necessary for phenix.refine map coefficients (2FOFCWT,PH2FOFCWT and FOFCWT,PHFOFCWT.) to be automatically recognized. I just tried this with a new MTZ file and it already works in 0.6. Phaser MR maps (FWT,PHFWT and DELFWT,PHDELWT) definitely work too, but SAD phased maps from AutoSol do not. Probably RESOLVE maps (FP, PHIM, FOMM) would be the next most useful, since those are used by many programs. (RESOLVE uses these for difference maps too, unfortunately, but it is at least predictable when this happens.) From *my* perspective, it doesn't really matter what the column labels are, only the other metadata in the MTZ file. If this could be made consistent across all of our programs, more auto-open functionality in Coot would be nice but not essential for the Phenix GUI. -------------------- Nathaniel Echols Lawrence Berkeley Lab 510-486-5136 [email protected]
Nathaniel Echols wrote:
On Jan 23, 2010, at 2:55 PM, Paul Emsley wrote:
Dear Phenixers,
Prompted by Randy Read, Kevin and I have agreed that Coot should be able to auto-open MTZ files with columns labelled
groupname.F_phi.F and groupname.F_phi.phi
and
groupname.DF_phi.DF, groupname.DF_phi.phi for difference maps
for any value of groupname, of course.
I'm not sure I understand - what do the fields represent?
Sorry for not being clear. They are column labels, groupname should be something informative, indicating the provenance of the data. Perhaps an example would clear things up: 2FOFCWT,PH2FOFCWT would be replace by phenix.refine.F_phi.F and phenix.refine.F_phi.F or perhaps phaser_sol_2.F_phi.F and phaser_sol_2.F_phi.phi and FP, PHIM, FOMM would be perhaps RESOLVE_weighted_Fs.F_phi.F and RESOLVE_weighted_Fs.F_phi.phi. There should be no need (for the user) to consider the weights column, we have seen too many otherwise competent people use double-weighted or unweighted maps (I'm sure you must have experienced that too). These columns can be generated in addition to the current ones of course.
From the user perspective, I think it's only absolutely necessary for phenix.refine map coefficients (2FOFCWT,PH2FOFCWT and FOFCWT,PHFOFCWT.) to be automatically recognized.
Yes, this scheme would be in addition to the already-known pairs.
From *my* perspective, it doesn't really matter what the column labels are, only the other metadata in the MTZ file. If this could be made consistent across all of our programs, more auto-open functionality in Coot would be nice but not essential for the Phenix GUI.
Sure thing, if you are in charge of the invoking process you know or have control of the column labels. This "de-facto" standard might be useful for other viewers, so that when users that click on the mtz file on the desktop, and the prefered viewer is invoked and auto-opens all the maps the the generator wanted the user to look at. Paul.
Sorry for not being clear. They are column labels, groupname should be something informative, indicating the provenance of the data.
Perhaps an example would clear things up:
2FOFCWT,PH2FOFCWT would be replace by phenix.refine.F_phi.F and phenix.refine.F_phi.F or perhaps phaser_sol_2.F_phi.F and phaser_sol_2.F_phi.phi and FP, PHIM, FOMM would be perhaps RESOLVE_weighted_Fs.F_phi.F and RESOLVE_weighted_Fs.F_phi.phi. There should be no need (for the user) to consider the weights column, we have seen too many otherwise competent people use double-weighted or unweighted maps (I'm sure you must have experienced that too). These columns can be generated in addition to the current ones of course.
Probably "F_phi.F" in "xxx.F_phi.F" above also should be something informative. For example, it should reflect the map type: 2mFo-DFc 2Fo-Fc 2mFo-DFc (computed with missing Fobs filled with DFc) 2Fo-Fc (computed with missing Fobs filled with DFc) Kick maps Anomalous difference map etc... in some obvious way. Currently, phenix.refine has to output 2mFo-DFc map with labels (2FOFCWT,PH2FOFCWT) simply because this is what Coot will automatically recognize. Looking at (2FOFCWT,PH2FOFCWT) it is absolutely not obvious which map it is (unless you coded it therefore know). However, something like (2mFoDFc, PH2mFoDFc) has the same length as (2FOFCWT,PH2FOFCWT) but much more informative. For kick or missing-Fobs-filled maps, one can simply add a suffix to corresponding map name, for example: (2mFoDFc_fill, PH2mFoDFc_fill) (2mFoDFc_kick, PH2mFoDFc_kick) Pavel.
Another idea: we could use the history lines to communicate in a flexible way what we want the browser to auto-open. E.g. auto_open = 2FOFCWT PH2FOFCWT auto_open = 2FOFCWT_fill PH2FOFCWT_fill would ask the viewer to open these two maps when the mtz file is read. Then we could even provide a mechanism for users to select themselves what maps are automatically opened, e.g.: phenix.refine.electron_density_maps { map { mtz_label_amplitudes = my_map.f mtz_label_phases = my_map.phi likelihood_weighted = True obs_factor = 3 calc_factor = 2 kicked = True fill_missing_f_obs_with_weighted_f_model = True viewer_auto_open = True } } This could be taken further: auto_open = important_refine_003.pdb auto_open = 2FOFCWT PH2FOFCWT auto_open = 2FOFCWT_fill PH2FOFCWT_fill Maybe with extra settings: auto_open = 2FOFCWT PH2FOFCWT contour_level=3 rgb=0,0,1 auto_open = 2FOFCWT_fill PH2FOFCWT_fill contour_level=3 rgb=0,1,0 Ralf
Pavel Afonine wrote:
Probably "F_phi.F" in "xxx.F_phi.F" above also should be something informative.
It is already informative. It tells Coot that this is an "auto-open-able" column.
For example, it should reflect the map type: 2mFo-DFc 2Fo-Fc 2mFo-DFc (computed with missing Fobs filled with DFc) 2Fo-Fc (computed with missing Fobs filled with DFc) Kick maps Anomalous difference map etc...
in some obvious way.
Those descriptions should go in the groupname (difference maps should have the .DF_phi.DF suffix)
Currently, phenix.refine has to output 2mFo-DFc map with labels (2FOFCWT,PH2FOFCWT) simply because this is what Coot will automatically recognize. Looking at (2FOFCWT,PH2FOFCWT) it is absolutely not obvious which map it is (unless you coded it therefore know).
I agree.
However, something like (2mFoDFc, PH2mFoDFc) has the same length as (2FOFCWT,PH2FOFCWT) but much more informative.
Fine.
For kick or missing-Fobs-filled maps, one can simply add a suffix to corresponding map name, for example:
(2mFoDFc_fill, PH2mFoDFc_fill) (2mFoDFc_kick, PH2mFoDFc_kick)
OK, so do we agree that _xxx it should go in the groupname? 2mFoDFc_fill.F_phi.F,2mFoDFc_fill.F_phi.phi Paul.
There is no problem with that on this site. How "groupname" will be generated? We have some computers (laptops) that are moving from offices (static connection to internet) to homes (dynamic ADSL connection to internet). I hope it will not affect a new approach. I hope that it is to help us and not to check on us. Dr Felix Frolow Professor of Structural Biology and Biotechnology Department of Molecular Microbiology and Biotechnology Tel Aviv University 69978, Israel Acta Crystallographica D, co-editor e-mail: [email protected] Tel: ++972 3640 8723 Fax: ++972 3640 9407 Cellular: ++972 547 459 608 On Jan 24, 2010, at 24:55 , Paul Emsley wrote:
Dear Phenixers,
Prompted by Randy Read, Kevin and I have agreed that Coot should be able to auto-open MTZ files with columns labelled
groupname.F_phi.F and groupname.F_phi.phi
and
groupname.DF_phi.DF, groupname.DF_phi.phi for difference maps
for any value of groupname, of course.
(Any weights should be applied to Fs before writing the file).
We were wondering what the Phenix community/developers thought about that idea (there's not much point in adding this feature if you don't agree).
Thanks, Paul.
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
participants (5)
-
Felix Frolow -
Nathaniel Echols -
Paul Emsley -
Pavel Afonine -
Ralf W. Grosse-Kunstleve