Hi Kyle, I like James' suggestion.
A lightweight program to tell waters from ions used to be "xpand" by Gerard Kleywegt. I just compiled this program for you. The zipfile with a "xpand" Linux binary, and the HTML manual "xpand_man.html" is at https://cloud.uni-konstanz.de/index.php/s/iEMCFF3rjXPd3Ap It is a commandline program. When you run it, just specify "W" (for water-scrutinizer), and at the next prompt, the name of your PDB file. Tried it with 6HT2.pdb and it suggested # ? F- (or other small anion) : ( 3) out of the 122 water molecules. Maybe this can identify the ions among your waters. Hope this is useful, Kay On Wed, 14 Jan 2026 11:19:22 +0000, Kyle Gregory <[email protected]> wrote: >Hi James, > >Thank you for the suggestion. I'm getting our IT department to install phenix >soon so should be able to try this. > >Kind regards, >Kyle >________________________________ >From: CCP4 bulletin board <[email protected]> on behalf of James Holton ><[email protected]> >Sent: 13 January 2026 17:48 >To: [email protected] <[email protected]> >Subject: Re: [ccp4bb] Unusually low B-factors for 'waters' > >You don't often get email from [email protected]. >Learn why this is important<https://aka.ms/LearnAboutSenderIdentification> > >CAUTION: This email came from outside of the University. To keep your account >safe, only click on links and open attachments if you know the person who sent >the email, or you expected to receive this communication. > > > >I think if you are seeing clear "waters" at 2.6 A then they are probably not >waters. > >Try turning them into Chlorine and refine their occupancy? Multiply the >resulting occs by 8 and see how many electrons you get. > >Also, last I checked refmac5 does not support occupancies > 1, but phenix does >if you use main.occupancy_max=2. Then you don't need to change the atom names. > >Just a suggestion, > >-James Holton >MAD Scientist > > >On 1/12/2026 5:33 AM, Eleanor Dodson wrote: >No - you can check any peak over the S to get some idea if expected signal but >maybe the data is not good enough to show anom peaks.. > >On Mon, 12 Jan 2026 at 12:32, Kyle Gregory ><[email protected]<mailto:[email protected]>> wrote: >Dear Eleanor, > >There is no visible signal in the weighted anomalous difference map at 3 >sigma. Are there any other maps I should calculcate to check? > >Kind regards, >Kyle >________________________________ >From: Eleanor Dodson ><[email protected]<mailto:[email protected]>> >Sent: 12 January 2026 12:12 > >To: Kyle Gregory <[email protected]<mailto:[email protected]>> >Cc: [email protected]<mailto:[email protected]> ><[email protected]<mailto:[email protected]>> >Subject: Re: [ccp4bb] Unusually low B-factors for 'waters' > > >CAUTION: This email came from outside of the University. To keep your account >safe, only click on links and open attachments if you know the person who sent >the email, or you expected to receive this communication. > > > >You dont say if you have checked the anomalous Fourier - want to be sure they >ARE waters and not CA say! >Eleanor > >On Mon, 12 Jan 2026 at 12:02, Kyle Gregory ><[email protected]<mailto:[email protected]>> > wrote: >Dear all, > >Thank you everyone for your suggestions I will try to address everyone's >questions below. > >Indexing/integration was performed in DIALS followed by data reduction and >scaling in AIMLESS. MR with Phaser. Refinement was performed in Refmac all as >part of ccp4i2. I used automatic local NCS restraints and TLS restraints. I've >tried switching those off and the waters are still lower than protein. >Although I may try as Kay suggested (removing everything he said from PDB and >running basic Refmac settings). I've deleted the waters and re-added them, the >'problem' persists. I've ensured that the B-factors are reset at the start of >refinement, they are not automatically set to 16. 16 was an example. Some are >higher/lower but they are nearly always lower than the protein atoms they >coordinate. I truncated the data at 2.6 with a cc1/2 in outermost shell of >0.459 and i/sigi of 0.9. 100% completeness, with overall Rpim of 0.124, the >Rpim of the outermost shell is 0.872. I have ~ 350*10 residues. I have another >dataset (soaked with ligand) which behaves the same. The average B-factors of >all protein chains is 42. (33.1-49.2 range). The average B-factors for the >waters is 28.9. > >The wilson b-factor is 33.93. Some anisotropy is detected. >[cid:[email protected]] > >I am not too suspicious of 10 mols as the protein forms a biological decamer. >The current Rwork/Rfree is 0.18/0.23. Space group is C 2. I ran Zanuda because >of this query by Omid. Zanuda concludes the original space group assignment >'seems to be correct'. The unit cell dimensions are 183.444 114.965 189.441 >90.000 108.626 90.000. > >I will consider trying phenix/buster but I need to request their installation >by the IT department. > >Kind regards, >Kyle > > > > > >________________________________ >From: Ian Tickle <[email protected]<mailto:[email protected]>> >Sent: 07 January 2026 17:29 >To: Kyle Gregory <[email protected]<mailto:[email protected]>> >Cc: [email protected]<mailto:[email protected]> ><[email protected]<mailto:[email protected]>> >Subject: Re: [ccp4bb] Unusually low B-factors for 'waters' > > >CAUTION: This email came from outside of the University. To keep your account >safe, only click on links and open attachments if you know the person who sent >the email, or you expected to receive this communication. > > > >Hi Kyle > >You don't say what protocol you're using for the 10-fold NCS restraints (if >any). If the NCS restraints are too tight compared with the differences >between NCS copies this will push up the protein B factors (and the R values) >to compensate. So the problem may not be that the water B factors are too >low, rather the protein B factors are too high due to the overly tight NCS >restraints. Also as Rafael asked how does the Wilson B factor compare with >the averages for the protein and waters - that might give a clue. > >Cheers > >-- Ian > > >On Tue, 6 Jan 2026 at 14:48, Kyle Gregory ><[email protected]<mailto:[email protected]>> > wrote: >Dear all, > >I have a 2.6 angstrom structure, which has 10 molecules in the ASU. I have >modelled approx 200 water molecules. They fit spherical density nicely and >make suitable contacts with the protein. The only problem is nearly every >single water molecule has a lower B-factor than the protein atoms they >coordinate. E.g protein atoms could be 30/40 while water is 16 angstrom^2. > >Does anyone have a suggestion as to why, and whether there are some things I >should be trouble shooting? > >Kind regards, >Kyle > >________________________________ > >To unsubscribe from the CCP4BB list, click the following link: >https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > >________________________________ > >To unsubscribe from the CCP4BB list, click the following link: >https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > >________________________________ > >To unsubscribe from the CCP4BB list, click the following link: >https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > > >________________________________ > >To unsubscribe from the CCP4BB list, click the following link: >https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > >######################################################################## > >To unsubscribe from the CCP4BB list, click the following link: >https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 > >This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing >list hosted by www.jiscmail.ac.uk, terms & conditions are available at >https://www.jiscmail.ac.uk/policyandsecurity/ > ######################################################################## To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
