Re: [ccp4bb] Very strange LINK cards appearing in recent structures

2018-11-12 Thread John Berrisford
Dear Frank We are working together with the mmCIF working group on exactly this. mmCIF’s from phenix and refmac have parameters of the dictionaries used in refinement as additional data blocks. We are currently making sure that the output is identical from both packages so we can use it lat

Re: [ccp4bb] Asn/Gln - pi-stacking prevalence

2018-11-12 Thread Michael Jarva
Thank you to Tomas, Bill, and Andrew for their tips and insights! It was really all I hoped for and more. The eLife paper with its downloadable text file of mined pi-pi stacking is really useful and is easily adapted into fast visual inspection with pymol. Problem now is having more examples

Re: [ccp4bb] Very strange LINK cards appearing in recent structures

2018-11-12 Thread Frank von Delft
Sorry, that was menat to be:  "... force us _to deposit_ /full details/" On 12/11/2018 09:49, Frank von Delft wrote: While you're at it, can you force us /full details/ of restraints used, whether it's the particular CCP4 library release, the CIF of the ligand, or the non-standard links?

Re: [ccp4bb] Very strange LINK cards appearing in recent structures

2018-11-12 Thread Frank von Delft
While you're at it, can you force us /full details/ of restraints used, whether it's the particular CCP4 library release, the CIF of the ligand, or the non-standard links? As far as I understand refinement, they're as least as important a determinant of the final structure as the observed data

Re: [ccp4bb] Very strange LINK cards appearing in recent structures

2018-11-12 Thread John Berrisford
Dear all We (the wwPDB) are actively looking into and correcting link records in the PDB. A lot of the recent re-releases were (and will continue to be) to correct links in existing PDB entries. I removed over 10,000 incorrect links in one sitting! We are also working to ensure that incorrect lin