There are 2 issues raised here and the answers are:

1) It is absolutely normal that the same protein crystallizes in two or
more space groups. It may even happen under the same crystallization
conditions, due to the randomness of nucleation. It happens more often
when crystallization conditions are changed, e.g. when a ligand is added.
So the fact that you got different unit cell and space group cannot be
considered an indexing problem. Since your second crystal is monoclinic
with beta angle very different from 90 (as indicated by the distortion
index), enforcing orthorhombic (P222) lattice in Denzo will mispredict
reflections. Integration in monoclinic lattice is obviously correct.

2) A non-obvious thing happens when you enter a wrong unit cell in
Scalepack. An incorrect unit cell affects only the resolution calculations
of reflections. It will mostly impact the effective resolution and
completeness of data, as the resolution limit is defined with respect to
the unit cell. If you double all the unit cell parameters, the resolution
limit, and quadruple the B-factor restraint, the calculated scaled and
merged file will not be affected and, even more, it will have a correct
unit cell in the output if post-refinement is enabled. The starting input
for post-refinement is taken from Denzo output, so it is not affected by
an incorrect unit cell value entered into Scalepack, and the scaled and
merged file will have the unit cell from post-refinement. However, the
completeness and resolution ranges are calculated using unit cell values
provided externally to Scalepack from macro command. Increasing the
parameters of the unit cell will calculate merging statistics for lower
resolution ranges than printed in the output, so they may look better.
However, it will also reduce the number of reflections in the output due
to, effectively different, resolution limit.

Apart from this, one can always scale and merge data in lower symmetry
Laue group than the correct one; in fact, it has recently become almost a
disease, but it's a separate issue. Scaling in such a case will typically
not be much affected, since there'll be a sufficient number of symmetry
relationships to define it. However, merging statistics will be affected.
They will be (potentially)less complete, and will have lower signal/noise
ratio due to lower multiplicity of observations (redundancy). R-merges may
improve, but they are mostly meaningless statistics, so you shouldn't
bother about them anyway. Merging data in a higher Laue symmetry group
than the true one creates obvious disasters. Your e-mail is inconsistent
here; one may construct it to mean that you did merge data in higher
symmetry, while the statistical conclusions are inconsistent with it.


> Dear All,
>
> Recently I collected one data set for the protein having SG P222 with a=
> 36.8, b= 44.7, c= 78.4(reported with compound). I crystallized the protein
> with same kind of other compound. The diffraction was up to 2.3A. But I am
> facing problem during indexing. I tried with SGP222 (as reported) but the
> cell dimension was showing a= 41.5, b= 96.3, c= 112.7 and distortion index
> around 9.83% whereas for SGP2 its showing a= 96.3, b= 41.5, c= 112.7 with
> distortion index 0.03%. All spots are taken nicely (Denzo). But with P222
> predicted spots are completely different than the real one. So I processed
> it in P2. Surprisingly when I put the reported SGP222 and cell dimensions
> during scaling (scalepack), for the same data integrated in P2 (just for
> curiosity), it is showing better data statistics than P2. Anybody
> experienced this situation? Am I doing any mistake during indexing? Asking
> for the suggestions.
>
> Thanks in advance.
>
> Regards,
> Dipankar
>
>
>
> Aurigene Discovery Technologies Limited,
> #39-40 KIADB Industrial Area, Electronic City, Phase II,
> Hosur Road, Bangalore- 560 100, India
> Cell: +91-9538631469  | Office Ph  : +91 80-66204422 (Extn: 398)  | Email
> ID: [log in to unmask]<mailto:[log in to unmask]>
>
>
> ________________________________
>
> This e-mail and any files transmitted with it are for the sole use of the
> intended recipient(s) and may contain confidential and privileged
> information.If you are not the intended recipient, please contact the
> sender by reply e-mail and destroy all copies of the original message.Any
> unauthorized review, use, disclosure, dissemination, forwarding,printing
> or copying of this email or any action taken in reliance on this e-mail is
> strictly prohibited and may be unlawful.
>
> Visit us at http://www.aurigene.com
>


Zbyszek Otwinowski
UT Southwestern Medical Center at Dallas
5323 Harry Hines Blvd.
Dallas, TX 75390-8816
Tel. 214-645-6385
Fax. 214-645-6353