You knowe there are alternative indexing for H3 - you couldnt have solved
one as h k l and the other as k h -l?
And alternate origins - H3 is a polar spacegroup so if you redid the
molecular replacement you may finish up anywhere along the c axis..
Easiest is to do a superpose of both sets of coordinates and see what the
transformation required is - it will almost certainly be a
crystallographically acceptable fit
Eleanor


On 3 May 2013 10:02, Kavyashree Manjunath <[email protected]> wrote:

> Dear Sir,
>
>
> > There used to be a chaos with H3 and R3 settings so you first thing you
> > might want to check that the same setting is used in both cases. Easiest
> > would be to check the CRYST1 record in your pdb files to make sure that
> > the same cell is used.
>
> I check it, the unit cell dimensions are identical as I did
> not reprocess the data. I used the same processed data for both
> with change only during scaling by adding the native Rfree set.
>
> > If you did not start for the second data set from the refined coordinates
> > from the first one but reran molrep instead, your molecule might have
> > landed in a different asymmetric unit/unit cell/origin. In that case I
> > would superimpose the coordinates from the second refinement on those of
> > the first refinement and maybe run one more cycle of refinement to get
> rid
> > of rounding errors. The should solve your problem.
>
> Ok.
>
> > Herman
> >
> > PS: I like the Effortless program, especially if they would add an option
> > to write the paper as well with some buttons to select the desired
> journal
> > e.g. Nature, Science, Cell etc. The cheat button should only be available
> > for experienced users though. ;-)
>
> That would probably be the ultimate aim of CCP4!!
>
> Thank you
> Regards
> Kavya
>
> >
> > -----Original Message-----
> > From: CCP4 bulletin board [mailto:[email protected]] On Behalf Of
> > Kavyashree Manjunath
> > Sent: Friday, May 03, 2013 10:07 AM
> > To: [email protected]
> > Subject: [ccp4bb] A small clarification
> >
>
>
> > Dear users,
> >
> > I wanted a small clarification, I was solving a ligand data in H3 space
> > group with a dimer as the asymmetric unit.
> >
> > Initially, I had solved and refined this without using the same Rfree
> > reflections as that of native data. So I resolved and refined the same
> > data by considering the native Rfree reflections. In both the cases after
> > the final refinement, the R and Rfree had reasonably good values.
> >
> > The problem arose when I compared the maps of data -
> > (1) solved without same Rfree as that of native and
> > (2) solved with same Rfree as that of the native
> >
> > They did not superpose at all. How is this possible? Although it is
> > originally the same data and in each case the map traced the molecule
> very
> > well, but when both maps are opened in coot they do not superpose at all.
> >
> > I would like to mention that the data has a pseudo translation symmetry
> > with 17% peak. Is this responsible for the results I am getting?
> >
> > I hope one day there will be a program in CCP4 called "EFFORTLESS"
> > which gives a structure from a given sequence!! :)
> >
> > Thanking you
> > Regards
> > Kavya
> >
> >
> > --
> > This message has been scanned for viruses and dangerous content by
> > MailScanner, and is believed to be clean.
> >
> > --
> > This message has been scanned for viruses and
> > dangerous content by MailScanner, and is
> > believed to be clean.
> >
> >
>
>
>
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
>

Reply via email to