Dear all
this is *really* embarrassing for me. It's just been pointed out to
me (and also demonstrated...) that this script won't produce the
predictions as I suggested it would. In order to make an image with
the predictions on, you need an extra line in the script -
matrix r3.mat
symm r3
image hg_001.mar1600
xgui on
go
predict_spots
create_image type ppm prediction on binary true filename junk.ppm
return
exit
Thanks to the anonymous user (still wearing shorts, even in
February...) who pointed this out!
On 8 Dec 2010, at 16:16, Harry Powell wrote:
Hi Keitaro
perhaps we should discuss this discreetly off-board!
Yes, you can make a picture on the command-line with prediction
boxes. I do it frequently.
If I take one of our example HypF images (see http://www.ccp4.ac.uk/
autostruct/testdata/mosflm/hg_images.tar), then on the command line
I can -
matrix r3.mat
symm r3
image hg_001.mar1600
xgui on
go
create_image type ppm prediction on binary true filename junk.ppm
return
exit
where "r3.mat" is a matrix file produced from indexing with Mosflm
and contains -
-0.00443809 -0.00922337 0.02048824
-0.01923344 0.01406363 0.00304649
-0.00630639 -0.01210521 0.00058584
0.000 0.000 0.000
-0.2141751 -0.5582786 0.8015324
-0.9281747 0.3719797 0.0110743
-0.3043363 -0.7415902 -0.5978489
62.0720 62.0720 62.0720 56.3041 56.3041
56.3041
0.000 0.000 0.000
and I get an image (converted to a jpg to reduce bandwidth hogging) -
http://www.mrc-lmb.cam.ac.uk/harry/junk.jpg
On 8 Dec 2010, at 16:03, Keitaro Yamashita wrote:
Dear Harry,
Do you mean that it cannot be done in command line to make picture
with prediction boxes?
I wrote input file for ipmosflm, by which I had only to click
"Predict" to see the predictions on GUI.
(I wrote matrix, symmetry and mosaic spread information in the
input file!)
Or, how can I get the predictions on the images in command line?
Thank you very much,
K. Yamashita
2010/12/9 Harry Powell <ha...@mrc-lmb.cam.ac.uk>:
Hi Keitaro
You should get the predictions on the images using Mosflm if you
supply a matrix & spacegroup. Running the "create_image" part
after indexing in iMosflm means that the matrix, symmetry and
mosaic spread are all available, so it can be easier doing it
this way.
There's a very old script under our "FAQ" page for producing a
series of images - see http://www.mrc-lmb.cam.ac.uk/harry/mosflm/
FAQ.html#movie
On 8 Dec 2010, at 14:54, Keitaro Yamashita wrote:
Dear Boaz and Harry,
Thank you very much for your suggestions.
As pointed out by Harry, I think it is difficult to
automatically make
a picture of spots labeled with their indexes..
But I have never known that iMosflm could accept the commands of
ipmosflm.
I'm glad to hear that and it worked very fine.
But when I did the same thing in ipmosflm input file, I got the
picture *without* prediction boxes (only copy of diffraction
image).
Of course I input MATRIX file.
This is the end of output of ipmosflm:
NeoCtrl => Image will be written to file image.ppm
invert = 0, theta = 0, wide = 3072
NeoCtrl
=>*****************************************************************
***************
At the beginning of CONTROL, the arguments have these values:
FIRSTTIME: F IFIRSTPACK: 1 NEWGENF: T GENOPEN: F RPTFIRST: F
MODE: 0
CELLSTR:
*******************************************************************
*************
MOSFLM =>
*** WARNING ** NO "RUN" CARD GIVEN
*********** END OF PROCESSING *****************
I'm using Mosflm 7.0.6.
Since I'd like to automatically make pictures of several frames
with
prediction boxes, so I'm trying to do that in ipmosflm.
Thank you for advice in advance,
K. Yamashita
2010/12/8 Harry Powell <ha...@mrc-lmb.cam.ac.uk>:
Hi Keitaro
Mosflm has been able to output jpegs with predictions for many
years - it can also output ppm files (which are not compressed
in a lossy way, so probably better for zooming in or converting
to other formats). However, you'd need to add the indices to
the prediction boxes manually - I don't know a really robust
way to automate putting them in in sensible places with
sensible font sizes etc. for all possible cells.
iMosflm (i.e. the new GUI) itself doesn't have the widgets to
do this, but if you run an iMosflm job and get at least as far
as having indexed and estimated (or entered) a mosaicity do the
following;
* go to the "History" task
* click on "Log" (to bring up the light green on green log
display)
* left mouse click in the display
* <Ctrl>-k to bring up the command entry line
* enter the following commands
image <image number of the image you want to plot>
xgui on
go
create_image type ppm prediction on binary true filename
image.ppm (creates a ppm)
or
create_image prediction on binary true filename
image.jpg (creates a jpeg, default)
then
return (redirects input back to the normal Mosflm commands)
It's also possible to do this from the normal Mosflm command
line or put it in a batch file - I often do this to prepare a
set of images for movies.
On 8 Dec 2010, at 05:31, Keitaro Yamashita wrote:
Dear all,
I would like to make a picture of diffraction photograph with
(hkl) indexes.
I found it in Fig. 1 in the paper: Acta Cryst. (2009). D65,
553-559
http://dx.doi.org/10.1107/S0907444909010725
Direct link to the figure:
http://journals.iucr.org/d/issues/2009/06/00/dz5158/
dz5158fig1.html
Can LABELIT, Mosflm or other program make image file (jpg or
something) such like that?
Thank you very much in advance,
K. Yamashita
Harry
--
Dr Harry Powell, MRC Laboratory of Molecular Biology, MRC
Centre, Hills Road, Cambridge, CB2 0QH
Harry
--
Dr Harry Powell, MRC Laboratory of Molecular Biology, MRC Centre,
Hills Road, Cambridge, CB2 0QH
Harry
--
Dr Harry Powell, MRC Laboratory of Molecular Biology, MRC Centre,
Hills Road, Cambridge, CB2 0QH
Harry
--
Dr Harry Powell,
MRC Laboratory of Molecular Biology,
Hills Road,
Cambridge,
CB2 0QH
