[phenixbb] Pb LocalResolutionMap calculation

Tue Oct 31 00:59:43 PDT 2023

Hi Tom,

i dowloaded and installed the nightly builds  1.21rc1-5134 but i do not see a direct access to the n_boxes parameter in the gui of the local Resolution Map menu. Is that normal, is it in the previous nightly build (5133)?

Another question : in the gui of the local Resolution Map menu, i can see the value of the parameters by navigating the parameters menu, but i cannot modified their values. Is that normal an how to solve it ?

Thank you so much for your help.

Best regards.

Philippe.

________________________________
De : phenixbb-bounces at phenix-online.org <phenixbb-bounces at phenix-online.org> de la part de CUNIASSE Philippe
Envoyé : dimanche 29 octobre 2023 22:06:55
À : Tom Terwilliger
Cc : phenixbb at phenix-online.org
Objet : [PROVENANCE INTERNET] Re: [phenixbb] Pb LocalResolutionMap calculation

Hi Tom,

Thank you so much.

I will test some values of this n_boxes parameter.

Best regards.

Philippe.

________________________________
De : Tom Terwilliger <tterwilliger at newmexicoconsortium.org>
Envoyé : dimanche 29 octobre 2023 14:03:45
À : CUNIASSE Philippe
Cc : phenixbb at phenix-online.org; Tom Terwilliger
Objet : Re: [phenixbb] Pb LocalResolutionMap calculation

Hi Philippe,

Yes, certainly the number of boxes affects the resulting local_resolution map.  The 2400 boxes correspond to 14x14x14 boxes for a cubic map which means that you have only a gradual sampling along any axis of your map.

You can adjust the number of boxes with the parameter n_boxes=2400.  In the GUI, search under "All parameters" for "box" and you'll get the entry form.   I will move this to the front page in tomorrow's nightly build of Phenix so it is easier to set.

All the best,
Tom T

On Sun, Oct 29, 2023 at 3:42 AM CUNIASSE Philippe <Philippe.CUNIASSE at cea.fr<mailto:Philippe.CUNIASSE at cea.fr>> wrote:

Hi Tom,

thank so much you for your answer. I uploaded the nigthtly build and it worked far better (a couple of hours) and ended correctly.

I noticed that the gain was due to the fact the only about 2400 boxes were calculated (in place of more than 40000 in my slow previous calculation, likely explaining the gain of a factor 20). As i have a very large map, i wonder how this affects the quality of the LocalResolutionMap. And i also wonder if there is a mean to define this parameter to study this effect.

Thank you so much for your help.

Best regards.

Philippe.

-----------------------------------------------
Philippe Cuniasse, PhD/HDR.
Institut de Biologie Integrative de la cellule.
UMR 9198 CNRS-CEA-Univ Paris Sud
Bat 144 CE-Saclay
91191 Gif-sur-Yvette Cedex
France
Tel:      (33) 1 69 08 56 35
Fax:      (33) 1 69 08 47 12
Email: philippe.cuniasse at cea.fr<mailto:philippe.cuniasse at cea.fr>
Web: http://biodev.cea.fr/rasmot3d/
-----------------------------------------------

________________________________
De : Tom Terwilliger <tterwilliger at newmexicoconsortium.org<mailto:tterwilliger at newmexicoconsortium.org>>
Envoyé : vendredi 27 octobre 2023 11:04:52
À : CUNIASSE Philippe
Cc : phenixbb at phenix-online.org<mailto:phenixbb at phenix-online.org>; Tom Terwilliger
Objet : Re: [phenixbb] Pb LocalResolutionMap calculation

Hi Philippe,

I'm sorry for the slow calculation...the problem is the number of boxes is too high.  If you download a nightly build like this one:

https://phenix-online.org/download/phenix/nightly/?version=1.21rc1-5130

then it should be a lot faster (still not quick, but in your case should be 20x faster).
All the best,
Tom T

On Fri, Oct 27, 2023 at 2:16 AM CUNIASSE Philippe <Philippe.CUNIASSE at cea.fr<mailto:Philippe.CUNIASSE at cea.fr>> wrote:

Dear all,

i try to calculate local resolution maps with phenix since a couple of days. It turns out that the calculation seems very slow and i suspect that there is a problem because i even cannot stop the calculation with the abort button in the gui. In my last run (still running), i use 2 half-maps and the program can read them correctly, start the calculation by returning that 42875 boxes are considered. I run it on my MacBook pro  (M1 running OSX 13.4 with 32Gb RAM) on 4 cpu. The calculation actually runs on 4 CPUs. However, with this calculation (and also a smaller map with about 17000  boxes on 1CPU), the calculations run more than 24 hours without finishing. I wonder if there coud be a problem and if no, what could be the duration of such calculation.

Thanks in advance for your help.

Best regards.

Philippe Cuniasse.

-----------------------------------------------
Philippe Cuniasse, PhD/HDR.
Institute for Integrative Biology of the Cell (I2BC)
UMR 9198 CNRS-CEA-Univ Paris Sud
Bat 144 CE-Saclay
91191 Gif-sur-Yvette Cedex, France

Tel:      (33) 1 69 08 56 35
Fax:      (33) 1 69 08 47 12
Email: philippe.cuniasse at cea.fr<mailto:philippe.cuniasse at cea.fr>
Web: http://biodev.cea.fr/rasmot3d/
Web: https://www.i2bc.paris-saclay.fr
[signature_2661676233]
-----------------------------------------------

_______________________________________________
phenixbb mailing list
phenixbb at phenix-online.org<mailto:phenixbb at phenix-online.org>
http://phenix-online.org/mailman/listinfo/phenixbb
Unsubscribe: phenixbb-leave at phenix-online.org<mailto:phenixbb-leave at phenix-online.org>

--
Thomas C Terwilliger
Laboratory Fellow, Los Alamos National Laboratory
Senior Scientist, New Mexico Consortium
100 Entrada Dr, Los Alamos, NM 87544
Email: tterwilliger at newmexicoconsortium.org<mailto:tterwilliger at newmexicoconsortium.org>
Tel: 505-431-0010

--
Thomas C Terwilliger
Laboratory Fellow, Los Alamos National Laboratory
Senior Scientist, New Mexico Consortium
100 Entrada Dr, Los Alamos, NM 87544
Email: tterwilliger at newmexicoconsortium.org<mailto:tterwilliger at newmexicoconsortium.org>
Tel: 505-431-0010

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20231031/8ebdca06/attachment-0001.htm>