I am confused, I am building a molecule and one refinement was done the gap of R free and Rmeas is 3%, so to reduce this huge gap if I join the broken chains having lesser density or side chains are not fitted into that, then if I refine with this modified or built model. Can I expect my model would be completed if I built in this way only and repeat the refinement again and again. My protein is a multimer and data is collected at 2.7 Angstrom. Any suggestions would be appreciated.
Hi, first of all please can you give a more information, such as which kind of protein, which x-ray sources you use (it cannot be 2.7 Angstrom), in which space grupe you made data collection and which software you use for crystal structure refinement and very important how you determined the phase? Did you use first isotropic refinement etc...
I mean to say that the resolution of the data is of 2.7 Angstrom. XDS was used for data processing and the space group was found to be P1. for refinement I have used Refmac (CCP4) and refinement parameter was with isotropic refinement.
Add only glycine residues if you see main chain electron density
First you have to be quite sure abour your space group. If it's not correct no amount of refinement will converge to best possible statistics. First of all you may like to delete all the parts of the model which you have feel are doubtful, then try to get is back with difference Fourier. How many molecules are there is your asymmetric unis? If you are seeing more than one molecule, can you see if there is any symmetry between/among them. You may also align the structures together and try to see if the there are differences in the corfomations of the molecule. If they are same then check if you think ther may be some bound ion(s) or other bound molecules that may be breaking the symmetry of the system for it to go to P1 space group. If you cannot find any reason for breaking of the symmetry then the space group could be one of the suspect for the problems.
What are your Rvalues currently? For a 2.7A resolution structure your R-value should be in the neighborhood of 25%.
Our recent paper Refining the macromolecular model - achieving the best agreement with the data from X-ray diffraction experiment has covered this aspect
Article Refining the macromolecular model – achieving the best agree...
- Badges
- Science topic
- Similar topics
- Engineering
- Materials Engineering
- Coating