Also R cryst is sometimes used for the same number, I think (of course
there are historical reasons for the different terms, but...).
JPK
On Mon, Jan 9, 2012 at 8:47 AM, Ed Pozharski <[log in to unmask]> wrote:
> On Mon, 2012-01-09 at 10:28 +0000, Guillaume Gotthard wrote:
>> Is there a mean to obtain statistics about R-Sym for deposited
>> structures databases ?
>
> 1. It's actually quite easy to do on your own if you want. This
> one-liner will get you the Rsym
>
> wget http://www.rcsb.org/pdb/files/XXXX.pdb?headerOnly=YES -O - -q |
> grep 'R SYM ' | cut -d: -f 2
>
> just replace the XXXX with the uppercase pdb id. I don't know what kind
> of statistics you want, but assuming that you are after resolution
> dependence, you can do this
>
> wget http://www.rcsb.org/pdb/files/XXXX.pdb?headerOnly=YES -O - -q |
> egrep 'R SYM |RESOLUTION RANGE HIGH \(' | cut -d: -f 2 | tr -d '\n' |
> tr -s ' '; echo
>
> 2. Beware that for probably the majority of the records you will get
> NULL as the answer. Using "R MERGE" instead of "R SYM" would result in
> a better outcome.
>
> 3. In any event, this is, imho, rather pointless. Rsym/Rmerge strongly
> depend on multiplicity/resolution cutoff, thus their values across the
> PDB don't really tell you much. Maybe this is why the PDB report
> generator does not even list it as the option under "Data collection".
>
> Maybe it's time for the PDB to start asking the depositors to provide
> the Rpim instead. Well, maybe this is 10 years overdue.
>
> Cheers,
>
> Ed.
>
> --
> Oh, suddenly throwing a giraffe into a volcano to make water is crazy?
> Julian, King of Lemurs
--
*******************************************
Jacob Pearson Keller
Northwestern University
Medical Scientist Training Program
email: [log in to unmask]
*******************************************
