Refmac 5.6 and Twin operator

From: Kiran Kulkarni
Date: 9 November 2011 13:03

Hi All,

 Is it possible to define twin operator and twin fraction in the latest version of Refmac (i.e., in version 5.6.0117) ?

I tried to define these two keywords in Buccaneer but it fails with an error message "Refmac_5.6.0117:  Problem in some of the instructions".

Please help me to fix this.

Many thanks,

-Kiran

----------
From: Boaz Shaanan

Hi,

Isn't it done by the program (refmac) ?

       Boaz

 

 

Boaz Shaanan, Ph.D.                                        
Dept. of Life Sciences                                     
Ben-Gurion University of the Negev                         
Beer-Sheva 84105                                           
Israel                                                     
                                                           

 

 

                

----------
From: Kiran Kulkarni

Hi,

Thank you for your response.

 Yes! Indeed, Refmac (in Buccaneer) automatically calculates twin operators and fractions.

But these values are slightly different from those calculated from Xtriage  (Though twin operators are equivalent twin fractions is different).

These difference shows up in R and Rfree when I refine the structure with Phenix and Refmac. With phenix refinement is converging where as with refmac it is not.

Furthermore, when I try to build the model using Buccaneer with an existing partially (nearly 75%) built model the program distorts structure with diverging R and Rfree.

Hence to keep the consistency I tried to key in the twin parameters from xtriage into refmac and then refine which resulted in this problem.

Hope this further clarifies my problem.

 

-Kiran

 

P.S: At present I am using Buccaneer model building option in phenix autobuild, which is progressing well. But just for the sake of comparison I tried to use Buccaneer in CCP4 with twin refinement option and encountered this problem.

----------
From: Pavel Afonine


Hi Kiran,

phenix.xtriage estimates the twin fraction, while phenix.refine (and I'm pretty sure Refmac too) refine it. That explains the difference you observe.

Pavel