Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions

[Phil Evans]

Just to be clear, the Pointless default setting (“CELL-BASED”), which has a On 10 Aug 2018, at 08:29, Kay Diederichs  
> wrote:
> 
> Tommi,
> 
> one aspect that may be confusing is that XDS and POINTLESS by default have a 
> different way to describe the setting of orthorhombic space groups with 1 or 
> 2 screw axes
> This seems relevant for this particular project, and is discussed in 
> https://strucbio.biologie.uni-konstanz.de/xdswiki/index.php/Pointless
> In short, to make the POINTLESS setting compatible with that of XDS, you want 
> to use
> SETTING SYMMETRY-BASED
> in the POINTLESS input. XDSGUI does that for you.
> 
> HTH,
> Kay
> 
> 
> 
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1



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

Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions

[Eleanor Dodson]

These problems are a pain!


My (half-baked) thoughts

0)  Yes - get rid of all junk images etc and inspect the frames  - we have
had a case where MR would not work for the "best" crystal as processed by
XDS, but when we looked at the images there was clear streaking and other
technical problems. the MR worked well for a better crystal with cleaner
images

1) The Space group indication seems pretty clear for P2 21 2 so I would be
reluctant to give up on that

2) Unless you have been very careful with assigning FreeR reflection in
orthrhombic then extending them to P21 you will certainly get a much lower
free R in the monoclinic spacegroup - an assignmemt of
FreeR reflections h k l and -h -k l are not really independent..

3)  Have you tried searching with the model dimer?

4) Can you send the REFMAC log - I wonder about the scaling..

5) Try MOLREP as well - it is not as powerful as Phaser but has a different
packing check algorithm

Eleanor




On 10 August 2018 at 08:29, Kay Diederichs 
wrote:

> Tommi,
>
> one aspect that may be confusing is that XDS and POINTLESS by default have
> a different way to describe the setting of orthorhombic space groups with 1
> or 2 screw axes
> This seems relevant for this particular project, and is discussed in
> https://strucbio.biologie.uni-konstanz.de/xdswiki/index.php/Pointless
> In short, to make the POINTLESS setting compatible with that of XDS, you
> want to use
> SETTING SYMMETRY-BASED
> in the POINTLESS input. XDSGUI does that for you.
>
> HTH,
> Kay
>
> 
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>



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

Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions

[Kay Diederichs]

Tommi,

one aspect that may be confusing is that XDS and POINTLESS by default have a 
different way to describe the setting of orthorhombic space groups with 1 or 2 
screw axes
This seems relevant for this particular project, and is discussed in 
https://strucbio.biologie.uni-konstanz.de/xdswiki/index.php/Pointless
In short, to make the POINTLESS setting compatible with that of XDS, you want 
to use
SETTING SYMMETRY-BASED
in the POINTLESS input. XDSGUI does that for you.

HTH,
Kay



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

Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions

[Eleanor Dodson]

Well that is pretty obviously P 2 21 2 - h 0 0   and l 0 0 are obviously
present..

The 0 k 0 absences could be generated by a nc translation of x, 1/2, z - do
you have that?

Twinning is unusual in P2/mmm  but possible of course - can you send the
pointless log file?
Eleanor



On 9 August 2018 at 09:11, Kajander, Tommi A 
wrote:

> Hi Tim - There are reflections in all directions and only one clearly has
> systematic absenses. Although only few seem to be present in one direction
> (But thats the h00 direction here
> + all are present though) -phaser finds with same data P22121 as the
> strongest solution (few reflections are weak in that direction from below…
> but most not).  TZ-score diffrence is not
> hure 13 vs 15.
>
> P 2 2 21
>   (   0,0,5): i/sigi =   20.8
>   (   0,0,7): i/sigi =   19.6
>   (   0,0,9): i/sigi =   21.0
>   (   0,0,   11): i/sigi =1.9
>   (   0,0,   13): i/sigi =   21.0
>   (   0,0,   15): i/sigi =6.0
>   (   0,0,   17): i/sigi =   20.8
>   (   0,0,   19): i/sigi =0.9
>   (   0,0,   21): i/sigi =9.0
>   (   0,0,   23): i/sigi =1.0
>   (   0,0,   25): i/sigi =4.9
>   (   0,0,   27): i/sigi =   19.9
>   (   0,0,   29): i/sigi =   20.9
>   (   0,0,   31): i/sigi =   20.2
>   (   0,0,   33): i/sigi =   19.1
>   (   0,0,   35): i/sigi =   18.2
>   (   0,0,   37): i/sigi =   13.1
>   (   0,0,   39): i/sigi =   20.8
>   (   0,0,   41): i/sigi =8.9
>   (   0,0,   43): i/sigi =   12.7
>   (   0,0,   45): i/sigi =2.5
>   (   0,0,   47): i/sigi =0.8
> P 21 2 2
>   (   3,0,0): i/sigi =   31.2
>   (   5,0,0): i/sigi =   25.2
>   (   7,0,0): i/sigi =   21.0
>   (   9,0,0): i/sigi =2.6
> P 2 21 2
>   (   0,5,0): i/sigi =0.7
>   (   0,7,0): i/sigi =0.7
>   (   0,9,0): i/sigi =0.7
>   (   0,   11,0): i/sigi =1.1
>   (   0,   13,0): i/sigi =1.0
>   (   0,   15,0): i/sigi =1.5
>   (   0,   17,0): i/sigi =0.7
>   (   0,   19,0): i/sigi =0.7
>   (   0,   21,0): i/sigi =0.7
>   (   0,   23,0): i/sigi =0.8
>   (   0,   25,0): i/sigi =0.7
>   (   0,   29,0): i/sigi =0.7
>   (   0,   31,0): i/sigi =0.8
>   (   0,   33,0): i/sigi =0.7
>   (   0,   35,0): i/sigi =0.7
>   (   0,   37,0): i/sigi =0.9
>   (   0,   39,0): i/sigi =0.7
>   (   0,   41,0): i/sigi =0.7
>   (   0,   43,0): i/sigi =0.7
>
>
> Neither seem to refine now so have to work on it a bit more beyond this  -
> we see the same “conflict" with two different crystals - seems complicated
> - the other one processed much
> better in P21  and seemed possibly twinned. Maybe the case here too.
>
> Tommi
>
>
>
>
>
> On Aug 9, 2018, at 10:29 AM, Tim Gruene  wrote:
>
> Dear Tommi,
>
> did you check whether you collected any reflections at all that should
> be absent for the second screw axis? If there are non - which could
> easily happen with low resolution, incomplete data - pointless and XDS
> might be conservative and not estimate the likelihood for the second
> screw-axis.
>
> MR, however, will pick up the space group with all data, and thus be
> able to tell between P2212 and P22121 whether or not you recorded
> reflections that are expected to be absent.
>
> Best,
> Tim
>
> On 08/08/2018 04:29 PM, Kajander, Tommi A wrote:
>
> Hi,
> Any clues why the followting happens: pointless (and just looking at the
> XDS output) clearly tells there is one screw axis in P-ortorhombic (P2212)
> yet phaser gives the best Z-scores in P22121. (...I suspect this may be to
> do with twinning - might be monoclinic twiined still though now processes
> very well in P222.)
>
> If i run the mtz after XDSCONV (ie F2MTZ) via pointless (instead of
> directly after XDS) it also suggests this - but i suppose i am not suppose
> to run merged data via pointless.
>
> Thanks for comments,
>
> Tommi
>
> 
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>
>
> --
> --
> Paul Scherrer Institut
> Tim Gruene
> - persoenlich -
> OSUA/204
> CH-5232 Villigen PSI
> phone: +41 (0)56 310 5297
>
> GPG Key ID = A46BEE1A
>
>
> Tommi Kajander, PhD
> Senior Scientist
> Structural Biology Program
> Institute of Biotechnology
> Biocenter 3, Viikinkaari
>  1 (PO
> Box 65)
> University of Helsinki
>
>
>
>
>
> --
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>



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

Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions

[Kajander, Tommi A]

Hi Tim - There are reflections in all directions and only one clearly has 
systematic absenses. Although only few seem to be present in one direction (But 
thats the h00 direction here
+ all are present though) -phaser finds with same data P22121 as the strongest 
solution (few reflections are weak in that direction from below… but most not). 
 TZ-score diffrence is not
hure 13 vs 15.

P 2 2 21
  (   0,0,5): i/sigi =   20.8
  (   0,0,7): i/sigi =   19.6
  (   0,0,9): i/sigi =   21.0
  (   0,0,   11): i/sigi =1.9
  (   0,0,   13): i/sigi =   21.0
  (   0,0,   15): i/sigi =6.0
  (   0,0,   17): i/sigi =   20.8
  (   0,0,   19): i/sigi =0.9
  (   0,0,   21): i/sigi =9.0
  (   0,0,   23): i/sigi =1.0
  (   0,0,   25): i/sigi =4.9
  (   0,0,   27): i/sigi =   19.9
  (   0,0,   29): i/sigi =   20.9
  (   0,0,   31): i/sigi =   20.2
  (   0,0,   33): i/sigi =   19.1
  (   0,0,   35): i/sigi =   18.2
  (   0,0,   37): i/sigi =   13.1
  (   0,0,   39): i/sigi =   20.8
  (   0,0,   41): i/sigi =8.9
  (   0,0,   43): i/sigi =   12.7
  (   0,0,   45): i/sigi =2.5
  (   0,0,   47): i/sigi =0.8
P 21 2 2
  (   3,0,0): i/sigi =   31.2
  (   5,0,0): i/sigi =   25.2
  (   7,0,0): i/sigi =   21.0
  (   9,0,0): i/sigi =2.6
P 2 21 2
  (   0,5,0): i/sigi =0.7
  (   0,7,0): i/sigi =0.7
  (   0,9,0): i/sigi =0.7
  (   0,   11,0): i/sigi =1.1
  (   0,   13,0): i/sigi =1.0
  (   0,   15,0): i/sigi =1.5
  (   0,   17,0): i/sigi =0.7
  (   0,   19,0): i/sigi =0.7
  (   0,   21,0): i/sigi =0.7
  (   0,   23,0): i/sigi =0.8
  (   0,   25,0): i/sigi =0.7
  (   0,   29,0): i/sigi =0.7
  (   0,   31,0): i/sigi =0.8
  (   0,   33,0): i/sigi =0.7
  (   0,   35,0): i/sigi =0.7
  (   0,   37,0): i/sigi =0.9
  (   0,   39,0): i/sigi =0.7
  (   0,   41,0): i/sigi =0.7
  (   0,   43,0): i/sigi =0.7


Neither seem to refine now so have to work on it a bit more beyond this  - we 
see the same “conflict" with two different crystals - seems complicated - the 
other one processed much
better in P21  and seemed possibly twinned. Maybe the case here too.

Tommi





On Aug 9, 2018, at 10:29 AM, Tim Gruene 
mailto:tim.gru...@psi.ch>> wrote:

Dear Tommi,

did you check whether you collected any reflections at all that should
be absent for the second screw axis? If there are non - which could
easily happen with low resolution, incomplete data - pointless and XDS
might be conservative and not estimate the likelihood for the second
screw-axis.

MR, however, will pick up the space group with all data, and thus be
able to tell between P2212 and P22121 whether or not you recorded
reflections that are expected to be absent.

Best,
Tim

On 08/08/2018 04:29 PM, Kajander, Tommi A wrote:
Hi,
Any clues why the followting happens: pointless (and just looking at the XDS 
output) clearly tells there is one screw axis in P-ortorhombic (P2212)
yet phaser gives the best Z-scores in P22121. (...I suspect this may be to do 
with twinning - might be monoclinic twiined still though now processes very 
well in P222.)

If i run the mtz after XDSCONV (ie F2MTZ) via pointless (instead of directly 
after XDS) it also suggests this - but i suppose i am not suppose to run merged 
data via pointless.

Thanks for comments,

Tommi



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


--
--
Paul Scherrer Institut
Tim Gruene
- persoenlich -
OSUA/204
CH-5232 Villigen PSI
phone: +41 (0)56 310 5297

GPG Key ID = A46BEE1A


Tommi Kajander, PhD
Senior Scientist
Structural Biology Program
Institute of Biotechnology
Biocenter 3, Viikinkaari 1 (PO Box 65)
University of Helsinki







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

Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions

[Tim Gruene]

Dear Tommi,

did you check whether you collected any reflections at all that should
be absent for the second screw axis? If there are non - which could
easily happen with low resolution, incomplete data - pointless and XDS
might be conservative and not estimate the likelihood for the second
screw-axis.

MR, however, will pick up the space group with all data, and thus be
able to tell between P2212 and P22121 whether or not you recorded
reflections that are expected to be absent.

Best,
Tim

On 08/08/2018 04:29 PM, Kajander, Tommi A wrote:
> Hi,
> Any clues why the followting happens: pointless (and just looking at the XDS 
> output) clearly tells there is one screw axis in P-ortorhombic (P2212)
> yet phaser gives the best Z-scores in P22121. (...I suspect this may be to do 
> with twinning - might be monoclinic twiined still though now processes very 
> well in P222.)
> 
> If i run the mtz after XDSCONV (ie F2MTZ) via pointless (instead of directly 
> after XDS) it also suggests this - but i suppose i am not suppose to run 
> merged data via pointless.
> 
> Thanks for comments,
> 
> Tommi 
> 
> 
> 
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
> 

-- 
--
Paul Scherrer Institut
Tim Gruene
- persoenlich -
OSUA/204
CH-5232 Villigen PSI
phone: +41 (0)56 310 5297

GPG Key ID = A46BEE1A




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


signature.asc
Description: OpenPGP digital signature

Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions

[Eleanor Dodson]

What is the tNCS vector?

Eleanor

On 8 August 2018 at 22:06, Kajander, Tommi A 
wrote:

> Dear Eleanor, no tNCS is present here..(2 mols). I think its probably the
> somewhat incomplete low resolution (90% ish) that causes problems
> somewhere... which again follows from bit poir crystal quality in some
> directions - overall data is not great...
>
> Best,
> Tommi
>
>
> Kohteesta: Eleanor Dodson
> Lähetetty: keskiviikko 8. elokuuta klo 17.46
> Aihe: Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions
> Vastaanottaja: Kajander, Tommi A
> Kopio: CCP4BB@JISCMAIL.AC.UK
>
>
> How many molecules in the asymmetric unit? and is there a
> non-crystallographic translaton?
>
> Eleanor
>
> On 8 August 2018 at 15:29, Kajander, Tommi A 
> wrote:
>
> Hi,
> Any clues why the followting happens: pointless (and just looking at the
> XDS output) clearly tells there is one screw axis in P-ortorhombic (P2212)
> yet phaser gives the best Z-scores in P22121. (...I suspect this may be to
> do with twinning - might be monoclinic twiined still though now processes
> very well in P222.)
>
> If i run the mtz after XDSCONV (ie F2MTZ) via pointless (instead of
> directly after XDS) it also suggests this - but i suppose i am not suppose
> to run merged data via pointless.
>
> Thanks for comments,
>
> Tommi
>
> 
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>
>
>
>
>



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

Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions

[Kajander, Tommi A]

Dear Eleanor, no tNCS is present here..(2 mols). I think its probably the 
somewhat incomplete low resolution (90% ish) that causes problems somewhere... 
which again follows from bit poir crystal quality in some directions - overall 
data is not great...

Best,
Tommi


Kohteesta: Eleanor Dodson
Lähetetty: keskiviikko 8. elokuuta klo 17.46
Aihe: Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions
Vastaanottaja: Kajander, Tommi A
Kopio: CCP4BB@JISCMAIL.AC.UK


How many molecules in the asymmetric unit? and is there a non-crystallographic 
translaton?

Eleanor

On 8 August 2018 at 15:29, Kajander, Tommi A 
mailto:tommi.kajan...@helsinki.fi>> wrote:
Hi,
Any clues why the followting happens: pointless (and just looking at the XDS 
output) clearly tells there is one screw axis in P-ortorhombic (P2212)
yet phaser gives the best Z-scores in P22121. (...I suspect this may be to do 
with twinning - might be monoclinic twiined still though now processes very 
well in P222.)

If i run the mtz after XDSCONV (ie F2MTZ) via pointless (instead of directly 
after XDS) it also suggests this - but i suppose i am not suppose to run merged 
data via pointless.

Thanks for comments,

Tommi



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






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

Re: [ccp4bb] screw axes /system. absenses and phaser/MR solutions

[Eleanor Dodson]

How many molecules in the asymmetric unit? and is there a
non-crystallographic translaton?

Eleanor

On 8 August 2018 at 15:29, Kajander, Tommi A 
wrote:

> Hi,
> Any clues why the followting happens: pointless (and just looking at the
> XDS output) clearly tells there is one screw axis in P-ortorhombic (P2212)
> yet phaser gives the best Z-scores in P22121. (...I suspect this may be to
> do with twinning - might be monoclinic twiined still though now processes
> very well in P222.)
>
> If i run the mtz after XDSCONV (ie F2MTZ) via pointless (instead of
> directly after XDS) it also suggests this - but i suppose i am not suppose
> to run merged data via pointless.
>
> Thanks for comments,
>
> Tommi
>
> 
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>



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

[ccp4bb] screw axes /system. absenses and phaser/MR solutions

[Kajander, Tommi A]

Hi,
Any clues why the followting happens: pointless (and just looking at the XDS 
output) clearly tells there is one screw axis in P-ortorhombic (P2212)
yet phaser gives the best Z-scores in P22121. (...I suspect this may be to do 
with twinning - might be monoclinic twiined still though now processes very 
well in P222.)

If i run the mtz after XDSCONV (ie F2MTZ) via pointless (instead of directly 
after XDS) it also suggests this - but i suppose i am not suppose to run merged 
data via pointless.

Thanks for comments,

Tommi 



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