mailRe: r22872 - /trunk/user_functions/relax_disp.py


Others Months | Index by Date | Thread Index
>>   [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Header


Content

Posted by Edward d'Auvergne on April 30, 2014 - 10:24:
I missed that line from their paper:

"The initial R20 rate was estimated as the lowest R2eff rate of each
relaxation dispersion dataset."

They however do not address the R20 << min(R2eff) issue causing the
grid search to be far from the region with the minimum.  Correct me if
I'm wrong there - maybe it's in their supplement or one of Peter
Wright's other papers.

Regards,

Edward




On 29 April 2014 23:20, Troels Emtekær Linnet <tlinnet@xxxxxxxxxxxxx> wrote:
Hi Edward.

This feature is not totally un-known.

It is used in GLOVE as well. :-)

p. 280, J Biomol NMR (2013) 56:275–283
dx.doi.org/10.1007/s10858-013-9747-5

Best
Troels

2014-04-29 15:14 GMT+02:00 Edward d'Auvergne <edward@xxxxxxxxxxxxx>:
Hi,

I would maybe include a third paragraph saying that this is an
experimental unpublished feature of the dispersion analysis.  And that
if R20 << min(R2eff), then the grid search will be performed in a
region of the optimisation space quite distant from the true minimum.
So if this is the case, then this feature should not be used.  This is
essentially to cover ourselves if a relax user tries this option but
it causes their analysis to fail - because their analysis is different
or performed in a non-standard way that we cannot foresee.  We can
then say that we said to use this at your own risk.  The word
"unpublished" is quite important for this.

Regards,

Edward



On 29 April 2014 15:03,  <tlinnet@xxxxxxxxxxxxx> wrote:
Author: tlinnet
Date: Tue Apr 29 15:03:58 2014
New Revision: 22872

URL: http://svn.gna.org/viewcvs/relax?rev=22872&view=rev
Log:
Better formattting of text in user function.

Feature request: #3151, (https://gna.org/support/index.php?3151) - User 
function to set the R20 parameters in the default grid search using the 
minimum R2eff value.

Modified:
    trunk/user_functions/relax_disp.py

Modified: trunk/user_functions/relax_disp.py
URL: 
http://svn.gna.org/viewcvs/relax/trunk/user_functions/relax_disp.py?rev=22872&r1=22871&r2=22872&view=diff
==============================================================================
--- trunk/user_functions/relax_disp.py  (original)
+++ trunk/user_functions/relax_disp.py  Tue Apr 29 15:03:58 2014
@@ -910,18 +910,19 @@

 # The relax_disp.set_grid_r20_from_min_r2eff user function.
 uf = uf_info.add_uf('relax_disp.set_grid_r20_from_min_r2eff')
-uf.title = "Set the initial guess for R20 in the grid search, to that of 
the minimum R2eff points."
-uf.title_short = "Speed-up grid search."
+uf.title = "Set the R20 parameter values to that of the minimum R2eff 
value."
+uf.title_short = "Set R20 from the minimum R2eff."
 uf.add_keyarg(
     name = "force",
     default = True,
     py_type = "bool",
     desc_short = "force flag",
-    desc = "A flag which if set to True will cause the R20 Values to be 
overwritten if they already exist."
-)
-# Description.
-uf.desc.append(Desc_container())
-uf.desc[-1].add_paragraph("Instead of making the grid search find 
initial values of the R20 parameter, the minimum for the R2eff points are 
used instead. For a 2 field cpmg experiment with model CR72, that would 
drop the number of uniform grid search points from gridNr^5 to gridNr^3. 
For standard 21 grid Nr, it would make the grid search 441 times faster.")
+    desc = "A flag which if set to True will cause the R20 values to be 
overwritten if they already exist."
+)
+# Description.
+uf.desc.append(Desc_container())
+uf.desc[-1].add_paragraph("Set the R20 parameter values to that of the 
minimum R2eff value.  This user function will look through all R2eff 
values per magnetic field strength, and find the minimum.  This minimum 
is a very good guess for the minimisation.")
+uf.desc[-1].add_paragraph("Instead of making the grid search find 
initial values of the R20 parameter, the minimum for the R2eff points are 
used instead.  For example for a two field CPMG experiment with model 
CR72, that would drop the number of uniform grid search points from 5D to 
3D.  For standard 21 grid increments per dimension, it would make the 
grid search 441 times faster.")
 uf.backend = set_grid_r20_from_min_r2eff
 uf.menu_text = "&set_grid_r20_from_min_r2eff"
 uf.gui_icon = "relax.grid_search"


_______________________________________________
relax (http://www.nmr-relax.com)

This is the relax-commits mailing list
relax-commits@xxxxxxx

To unsubscribe from this list, get a password
reminder, or change your subscription options,
visit the list information page at
https://mail.gna.org/listinfo/relax-commits

_______________________________________________
relax (http://www.nmr-relax.com)

This is the relax-devel mailing list
relax-devel@xxxxxxx

To unsubscribe from this list, get a password
reminder, or change your subscription options,
visit the list information page at
https://mail.gna.org/listinfo/relax-devel



Related Messages


Powered by MHonArc, Updated Wed Apr 30 11:00:19 2014