Print

Print
Hi Kristof,

Direct methods/charge flipping won’t work if your data quality is poor below ~1Å. Take a look at your Rmerge, as Jeffrey mentioned. Small molecule crystals have more stringent standards for data quality. If your Rmerge is 50% or more, it’s probably just noise and not useful for direct methods.

If SHELXT doesn’t work on the full data set, try cutting the data down to 0.9 Å or even 1 Å. I have had limited success doing that on some particularly bad data sets.  Alternatively, run SHELXD at full resolution indefinitely (NTRY 0) until it finds a solution. It could take days, but you may get something.

As a last resort, if you think this hexapeptide may have any secondary structure, you may want to try ab initio MR with ARCIMBOLDO. 

Aaron

------------------------------------------
Aaron Finke
Staff Scientist, MacCHESS
Cornell University
e-mail: [log in to unmask]

On Aug 2, 2018, at 8:53 AM, Kristof Van Hecke <[log in to unmask]> wrote:

Dear all,

I’m trying to solve a structure of a (modified) hexapeptide:
- inhouse (very decent) data up to 0.8 Angstrom
- average redundancy = 10
- according to the Matthews coefficient of 1.88 with 34.77 %solvent, there should be 3 Nmol/asym
- ‘large’ unit cell of about a=54, b=54, c=12
- SG = P3(1)12 or P3(2)12

As there’s (presumably) only C, H, N and O in the structure, I’m not able to solve this via Direct Methods, Charge Flipping etc,.
Trying MR (with Phaser) doesn’t give any results either, as there’s hardly any homologous models


Has anyone encountered a similar problem please, and could provide any possible solutions?
(building in heavy atoms isn’t my first option at the moment,. )


Thank you very much

Regards

Kristof
########################################################################

To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1


To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1