Dear Ligand Hunters,
Chris Weichenberger and I deposit a yearly update of ligand metrics, authors, journals, etc. of all ligands in the PDB. There is a huge spreadsheet you can transform into a Table and sort as desired, and also a basic applet to immediately view them in Coot. http://www.ruppweb.org/twilight/default.htm Before one embarks on a vigilante hunt for poor metrics, a few serious caveats: 1) The compilation is based on rote data mining. It is unaware of any context or circumstances. One needs to read the paper to evaluate how the ligand’s affairs affect the claims of the publication. The PDB now calls this ‘Focus of Research’. We have red and annotated a subset for internal validation purposes but it is not currently done (AI to the rescue?). An example for poor metrics but not invalid claims: the metrics are evaluated for the entire (enthusiastically modelled) ligand, but only a (still useful and informative) part may be visible. 2) One might consult the PDB_redo version of an interesting entry for confirmation. Sometimes odd things happened with older depositions, and new maps from PDB_redo might clarify the fit. 3) The list is to a degree overly optimistic. The real space correlation measures fail when both calculated density and observed density are weak. Multiple overlapping ligand conformations with low partial occupancies are one example. Or: Positive difference density without model (missing ligand parts or different density shape) is not evaluated. For those who are interested, I plan to present some of these situations at the ECM in Vienna. 4) The major achievements of your well-intended and valid scientific critique – should you accept the mission – will likely be frustration but you will be rewarded with captivating insights into cognitive dissonances and the complexities of human behavior. Topped off with a scoop of postmodernism. Mahlzeit. Cheers, BR From: CCP4 bulletin board <[email protected]> On Behalf Of Rhys Grinter Sent: Wednesday, July 24, 2019 4:37 AM To: [email protected] Subject: [ccp4bb] Questionable Ligand Density - Part 2 Hi All, Thanks for all the helpful comments and discussion surrounding my last post. I've been doing a little more investigation into this issue and wanted to see if people were able to provide me with some additional opinions/insights. I investigated the PDB entries for the lead deposition author of the 6MO0-1-2 series: Lin, Y.-L. (http://www.rcsb.org/pdb/results/results.do?tabtoshow=Current <http://www.rcsb.org/pdb/results/results.do?tabtoshow=Current&qrid=50E454B> &qrid=50E454B). With some slightly alarming results. A number of additional entries appear to have ligands associated with questionable density.: 6VB0,1,2,3,4 and 6BUY – Density does not support amino acids modelled and there are questionable interactions with protein and solvent (Published Scientific Reports 2017) 4NZ8, 4NAQ – Questionable density for peptide modelled, bad/unrealistic interactions with protein (Published PNAS 2012) 4HOM – Questionable density for peptide modelled. Fo-Fc map density on refinement seems more interconnected than would be indicated by water, however density does not fit peptide modelled. Bad/unrealistic interactions with protein observed, specifically Phe7 of peptide (chain B) (Published PNAS 2012, same paper as 5NZ8 and 4NAQ) 4KXD et. al. – Questionable density for amino acids in both EDS maps and re-refined with SF and model minus ligand. Poor R-factors obtained for 4KXD on refinement (W 0.27,F 0.33) (Published JBC 2013) These constitute many of entries associated with this author (some other not mentioned appear are okay) and it makes it seem much less likely to me that modelling of the structures from my previous post are a result of lack of experience or poor supervision. It is in my opinion a huge problem for the field and science in general, if structures like this are making it into publications in well respected journals. Best Wishes, Rhys -- Dr Rhys Grinter NHMRC Postdoctoral Researcher Monash University +61 (0)3 9902 9213 +61 (0)403 896 767 _____ To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB <https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1> &A=1 ######################################################################## To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1
