Basically data/parameter ratio no? Restraints have to be accounted for as either data or reduced parameters (not both). There is a more subtle caveat, or caveats, re: data quality. Not all data are created equal, but I’ll leave that to the experts.

If you’re refining the structure, divergent R and Rfree should reflect a poor data/parameter ratio. If you are reviewing somebody else’s refined structure, you only have the end values, so you have to use some judgment.  R and Rfree from structures of similar resolution in the structure database can be a useful guide.

Brent

From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of Eleanor Dodson
Sent: Friday, June 17, 2016 4:01 AM
To: [log in to unmask]
Subject: Re: [ccp4bb] over refined structure

Traditionally discrepancy between FreeR and R factor

Anecdotally - many many many "waters" with high B factors which reduce R factor but not FreeR

Eleanor

On 17 June 2016 at 11:16, chemocev marker <[log in to unmask]> wrote:

Hi all

What parameters should I look to judge if the structure is not over refined.

best

Jiri