[buster-discuss] test flag number in BUSTER

[Keitaro Yamashita]

Dear David and Claus,

Thank you for your helpful advices.

David wrote:
> Concerning the gap Rfree-R, you can get an estimation of the refinement
> behavior on the Buster
> wiki http://www.globalphasing.com/buster/wiki/index.cgi?AutoBusterExample1osgBasicNCS

When I tried refinement without NCS restraints, R-work and R-free
became 32.9% and 35.6%.
I think I should check test set selection bias.


Claus wrote:
> by default '0' is used. You can double-check that by inspecting
> the value of the FREFLG keyword in one of the CRD files eg.
> % grep FREFLG ./run00/01-BUSTER/Cycle-1/CRD
> KEYWORD    FREFLG=0 USESIG BLKBLR=50 MXLCYC=1 MSKRAD=400

Checking CRD file, I confirmed test set flag number was correctly chosen.

But I always found FREFLG=0 in CRD file though I specified
BusterFreeFlagValue= in command line argument.
Moreover, trying all flag values, I got the same R/R-free factors.

Could you tell me what to check?


Thanks in advance,

Keitaro


2011/3/30 Claus Flensburg <claus at globalphasing.com>:
> Dear Keitaro,
>
> On Sun, Mar 27, 2011 at 12:02:42AM +0900, Keitaro Yamashita wrote:
>> Dear BUSTER developers,
>>
>> I have questions about running autoBUSTER 2.8.
>> I'm very new to this software.
>>
>> How can I know which FreeR_flag number is used as test set?
>
> as David correctly said you can see column name in the main output
> from refine.
>
>> And how can I tell BUSTER the correct flag number?
>
> by default '0' is used. You can double-check that by inspecting
> the value of the FREFLG keyword in one of the CRD files eg.
> % grep FREFLG ./run00/01-BUSTER/Cycle-1/CRD
> KEYWORD    FREFLG=0 USESIG BLKBLR=50 MXLCYC=1 MSKRAD=400
>
> so in this case a value of '0' is used.
>
>> Looking at free-R value given by autoBUSTER, I noticed it is very
>> smaller than by other softwares (phenix.refine and refmac5 say 43%,
>> but BUSTER says 36%).
>> So I'm afraid that free-R flag number was wrongly chosen.
>
> well, not necessarily. To add to what David said, you can test
> it by selecting a different set.
>
> Please see a new wiki-page that shows how to set the value and
> what to expect when a 'new' set is chosen:
>
> https://www.globalphasing.com/buster/wiki/index.cgi?AutoBusterFreeRFlag
>
>
> Hope this helps,
>
> ClAuS
>