Re: [ccp4bb] software/web server to determine ligand volume

[Javier Gonzalez]

You can calculate volumes and much more, here:

http://www.molinspiration.com/cgi-bin/properties

Javier


On Wed, Aug 13, 2014 at 12:06 AM, sreetama das 
wrote:

> Dear all,
> Is there any software or web server available to calculate the volume of a
> ligand if the ligand coordinates are provided?
> Google seems to come up only with options to calculate protein cavity
> volume.
>
> Thanks in advance,
> Sreetama Das,
> phd student,
> Physics, IISc
>



-- 
Javier M. Gonzalez, PhD.
Protein Crystallography Station
Bioscience Division
Bioenergy and Biome Sciences Group (B-11)
Los Alamos National Laboratory
TA-3, Building 4200, Room 202B
Mailstop T007
Los Alamos, NM 87545
Phone: +1 (505) 667-9376
LinkedIn  Email

Re: [ccp4bb] Off topic: Cloning multi genes/multi cistronic in yeast

[Karsten Thierbach]

Hey,
in my old lab we developed a system to express two proteins from one 
plasmid using the before mentioned GAL1-10 Promoter. Giving the use of 
different plasmids with different selection markers, coexpression of 
multiple proteins is possible in yeast.  The system was used in a study 
published in Structure last year and you could contact my old lab to 
request plasmids. We used a TRP1 and a LEU2 plasmid and coexpressed 
three proteins in this study, but it is easily extendable to at least 4 
proteins.

You can find the paper here:
http://www.ncbi.nlm.nih.gov/pubmed/23954503
To receive plasmids from the study, please contact Ed Hurt, the 
corresponding author at the BZH in Heidelberg/Germany.
I stumbled across a similar system recently, which is also based on the 
GAL promoter, but don't find the reference now...

Good luck,
Karsten

Am 13.08.2014 15:31, schrieb Chris Putnam:

On 8/13/14 2:29 PM, Theresa Hsu wrote:

Dear all

One of my human membrane proteins have been described to interact 
with additional subunits for its activity. To obtain functional form 
in yeast (Saccharomyces), I can think of two approach of either 
cloning all the subunits under one promoter or reconstitute in vitro.


For the first option, what is the length of base pairs between the 
stop codon of one gene and the start of Kozak sequence for the next 
one? Is there any preference for the order so that only one subunit 
is His tagged?


Second option will need multiple purification steps and some trials 
with protein ratios. Is this better?





Natively, Saccharomyces does not have operons. And I am unaware of 
operon-like constructs (single promoter with multiple genes oriented 
in the same direction) working in Saccharomyces.  (The GAL1-10 
divergent promoter is probably the closest analog, but this involves 
transcription of the GAL1 and GAL10 genes that are encoded on opposite 
strands in opposite orientations.)


For multiprotein complexes, one strategy for in vivo complex assembly 
is to encode each protein (with its own promoter) on separate plasmid 
(with distinct selectable markers) and transform them all into the 
same strain. This avoids the problems of in vitro reconstitution of 
the complex as well as the problem of subunits that cannot be stably 
expressed alone.


Chris.


--
Karsten Thierbach, Dr. rer. nat.

California Institute of Technology
Division of Chemistry & Chemical Engineering
Hoelz laboratory

1200 E. California Blvd., M/C 147-75
Pasadena, CA 91125, U.S.A.

Re: [ccp4bb] Off topic: Cloning multi genes/multi cistronic in yeast

[Chris Putnam]

On 8/13/14 2:29 PM, Theresa Hsu wrote:

Dear all

One of my human membrane proteins have been described to interact with 
additional subunits for its activity. To obtain functional form in yeast 
(Saccharomyces), I can think of two approach of either cloning all the subunits 
under one promoter or reconstitute in vitro.

For the first option, what is the length of base pairs between the stop codon 
of one gene and the start of Kozak sequence for the next one? Is there any 
preference for the order so that only one subunit is His tagged?

Second option will need multiple purification steps and some trials with 
protein ratios. Is this better?




Natively, Saccharomyces does not have operons. And I am unaware of 
operon-like constructs (single promoter with multiple genes oriented in 
the same direction) working in Saccharomyces.  (The GAL1-10 divergent 
promoter is probably the closest analog, but this involves transcription 
of the GAL1 and GAL10 genes that are encoded on opposite strands in 
opposite orientations.)


For multiprotein complexes, one strategy for in vivo complex assembly is 
to encode each protein (with its own promoter) on separate plasmid (with 
distinct selectable markers) and transform them all into the same 
strain. This avoids the problems of in vitro reconstitution of the 
complex as well as the problem of subunits that cannot be stably 
expressed alone.


Chris.

[ccp4bb] monomer/dimer protein

[Todd Jason Green]

Hello All-

I am interested in monomer/dimer contamination when building a crystal lattice, 
ie. if you are building a crystal lattice with a monomeric species of protein, 
incorporation of dimers may yield lattice or surface defects. This species may 
be considered a macromolecular contaminant. I have read a few papers on this 
subject, a couple are listed here:

I. Yoshizaki et al. / Journal of Crystal Growth 290 (2006) 185–191

Caylor CL1, Dobrianov I, Lemay SG, Kimmer C, Kriminski S, Finkelstein KD, 
Zipfel W, Webb WW, Thomas BR, Chernov AA, Thorne RE. Proteins. 1999 Aug 
15;36(3):270-81.

In each of the studies that I have read, lysozyme is the model protein for 
these studies. I have not seen studies thus far that have been done with other 
proteins. Can anyone point me toward other studies, specifically non-lysozyme 
studies, where incorporation of two different oligomerization states has been 
shown to yield crystals with higher level of defects? Microscopy studies of 
such would be great too.

Thanks in advance-
Todd

[ccp4bb] Off topic: Cloning multi genes/multi cistronic in yeast

[Theresa Hsu]

Dear all

One of my human membrane proteins have been described to interact with 
additional subunits for its activity. To obtain functional form in yeast 
(Saccharomyces), I can think of two approach of either cloning all the subunits 
under one promoter or reconstitute in vitro.

For the first option, what is the length of base pairs between the stop codon 
of one gene and the start of Kozak sequence for the next one? Is there any 
preference for the order so that only one subunit is His tagged?

Second option will need multiple purification steps and some trials with 
protein ratios. Is this better?

Thank you.


Theresa

[ccp4bb] Computational Crystallography Newsletter - Volume 5, Number 2

[Nigel Moriarty]

 I am pleased to announce the publication of the latest issue of the
Computational Crystallography Newsletter:

   http://www.phenix-online.org/newsletter/

A listing of the articles and short communications is given below.
Please note that the newsletter accepts articles of a general nature
of interest to all crystallographers. Please send any articles to me at
nwmoria...@lbl.gov noting that there is a Word Template on the website
to streamline production.

Articles
Details of the Conformation-Dependent Library


Short Communications
Connectivity analysis tools in CCTBX

phenix.fab_elbow_angle: Fragment Antigen-Binding elbow angle calculation
tool 

Fitting tips #8 – Acetyl Groups are like Peptides: Planar and trans


Cheers

Nigel

-- 
Nigel W. Moriarty
Building 64R0246B, Physical Biosciences Division
Lawrence Berkeley National Laboratory
Berkeley, CA 94720-8235
Phone : 510-486-5709 Email : nwmoria...@lbl.gov
Fax   : 510-486-5909   Web  : CCI.LBL.gov

Re: [ccp4bb] desiring untouched ligand

[Garib Murshudov]

Hi Remie,

You can add harmonic restraints for parts you do not want to move too much. 
Instructions could be found here:
http://www2.mrc-lmb.cam.ac.uk/groups/murshudov/content/refmac/refmac_keywords.html#Harmonic


Regards
Garib


On 13 Aug 2014, at 15:44, Remie Fawaz-Touma  wrote:

> Hi everyone,
> 
> Does anyone know of a way to refine with CCP4 - Refmac5 (restrained 
> refinement is what I do) fixing a part of the the ligand?
> 
> Thank you very much for your input,
> 
> Remie

Dr Garib N Murshudov
MRC-LMB
Francis Crick Avenue
Cambridge 
CB2 0QH UK
Web http://www.mrc-lmb.cam.ac.uk, 
http://www2.mrc-lmb.cam.ac.uk/groups/murshudov/

[ccp4bb] desiring untouched ligand

[Remie Fawaz-Touma]

Hi everyone,

Does anyone know of a way to refine with CCP4 - Refmac5 (restrained refinement 
is what I do) fixing a part of the the ligand?

Thank you very much for your input,

Remie

Re: [ccp4bb] thin shell Rfree set selection

[Eleanor Dodson]

It depends to a large extent on the orientation of the Ncs operator.

If that aligns with a crystal axis then there can be bias in the Rfree
selection but if it doesn't NCS does not affect the Free R much ( in my
opinion..)
Eleanor


On 13 August 2014 09:14, Ed Pozharski  wrote:

> By all means, try it both ways and see whether the R-Rfree gap narrows
> with random vs thin shell selection. Depending on resolution and data
> quality, you may also consider imposing NCS restraints.
>
>
> Sent on a Sprint Samsung Galaxy S® III
>
>
>  Original message 
> From: Xianchi Dong
> Date:08/12/2014 4:45 PM (GMT-05:00)
> To: CCP4BB@JISCMAIL.AC.UK
> Subject: [ccp4bb] thin shell Rfree set selection
>
> Dear all,
> I have a dataset of C2 symmetry and 2 molecules in the ASU. I am wondering
> if I have to use thin shell Rfree set selection to avoid Rfree bias by NCS.
> And which Rfree selection method is better?
>
> Thanks in advance.
>
> Xianchi
>

Re: [ccp4bb] random half data sets

[Nat Echols]

On Tue, Aug 12, 2014 at 10:28 PM, Keller, Jacob 
wrote:

> A somewhat similar question, with a quick answer I hope: when programs
> output CC's of 1/2 datasets, are several random halvings compared/averaged,
> and if not, does this make a difference, or are the scores so similar
> there's no point?
>

The latter, I think.  It probably only matters for data where you have a
lot of erroneous observations (like ice rings) or at the fringes where
there's almost no signal anyway.  In my hands, a dataset with mulitplicity
of 3.9 has an outer shell with a CC1/2 between 0.497 and 0.503 depending on
random seed, which isn't worth worrying about.

-Nat

Re: [ccp4bb] Twinning in space group Pc

[George Sheldrick]

Dear Kristof,

Have you tried to solve it with the new SHELXT? You can force it to 
consider only chiral (Sohnke) space groups by putting -c on the command 
line.


Best wishes, George




On 13.08.2014 11:00, Kristof Van Hecke wrote:

Dear,

I’m struggling with the following (small molecule) problem:

We are trying to solve the structure of a metal-organic framework containing a 
chiral compound.
The space group is most probably Pc, but when refining, SHELX gives the error 
“Possible racemic twin or wrong absolute structure - try TWIN refinement”.
As we know our compound is enantiopure, a racemic twin is very unlikely. In 
this regard, also a centro-symmetric space group is not possible (although 
CrysAlisPro always gives P2/c as the proper space group). As a matter of fact, 
trying different space groups is not solving the problem.

The second problem is that half of the structure is visible, but the other half 
is completely not clear. Refinement is not possible at all (R-value of 33%).
When running TwinRotMat (Platon), I get the following possible 2-fold twin axes:

2-axis (   0   1  -1 ) [  -2   5  -4 ], Angle () [] =  2.31 Deg, Freq =14
 *
(-0.992   -0.0190.015)   (h1)   (h2)   Nr Overlap =84
(-0.4300.075   -0.860) * (k1) = (k2) BASF =  0.96
( 0.459   -1.146   -0.083)   (l1)   (l2)DEL-R =-0.064
  
2-axis (   0   1   1 ) [   2   5   4 ], Angle () [] =  2.31 Deg, Freq =15

 *
(-0.9920.0190.015)   (h1)   (h2)   Nr Overlap =   229
( 0.4300.0750.860) * (k1) = (k2) BASF =  0.94
( 0.4591.146   -0.083)   (l1)   (l2)DEL-R =-0.050
  
2-axis (   1   0  -2 ) [   3   0  -2 ], Angle () [] =  0.54 Deg, Freq =19

 *
(-0.1360.000   -0.576)   (h1)   (h2)   Nr Overlap =   992
( 0.000   -1.0000.000) * (k1) = (k2) BASF =  0.86
(-1.7030.0000.136)   (l1)   (l2)DEL-R =-0.030
  
2-axis (   1   2  -1 ) [   5   5  -1 ], Angle () [] =  0.39 Deg, Freq =13

 *
(-0.3800.620   -0.124)   (h1)   (h2)   Nr Overlap =   854
( 1.2560.256   -0.251) * (k1) = (k2) BASF =  0.88
(-0.617   -0.617   -0.877)   (l1)   (l2)DEL-R =-0.019
  
However, none of these do actually improve the refinement.



Has anyone encountered possible twinning/twin laws in Pc please?
Or any other suggestions are most welcome?


Thank you very much

Kristof




--
Prof. George M. Sheldrick FRS
Dept. Structural Chemistry,
University of Goettingen,
Tammannstr. 4,
D37077 Goettingen, Germany
Tel. +49-551-39-33021 or -33068
Fax. +49-551-39-22582

[ccp4bb] AW: [ccp4bb] off topic

[Herman . Schreuder]

Dear Careina,

even in vivo, the pka depends on the local environment, so the question boils 
down to: is the local environment in vivo different from in vitro? E.g. is your 
protein embedded or in contact with a membrane, is it part of a large 
multiprotein complex etc? Does your crystallization solution or the in vivo 
compartment of your protein contain molecules that could influence the pka?

Since I assume that you did not measure pka's experimentally, I assume that the 
remarks of the referee concern the discussion of your paper. In this case, I 
would use some arguments like above as response to the referee, depending on 
the context of course, which I do not know.

Good luck!
Herman


Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von Careina 
Edgooms
Gesendet: Mittwoch, 13. August 2014 14:45
An: CCP4BB@JISCMAIL.AC.UK
Betreff: Re: [ccp4bb] off topic

thank you, yes I am aware of  propka and I know very well that the local 
environment of the amino acid is what contributes to its pka based on things 
like whether it is buried or exposed or involved in H-bonding. I am in the 
process of responding to a reviewer who suggests but does not give references 
that  in vivo conditions can alter the ionisation of some amino acids 
differently to in vitro. I am struggling to respond to this without some sort 
of proof or study on the matter
C

On Wednesday, August 13, 2014 2:38 PM, David Briggs 
mailto:drdavidcbri...@gmail.com>> wrote:

Dear Careina,

Following on from what Herman said, if you have a structure you can use the 
propKa server

http://propka.ki.ku.dk/

to predict pKas for amino acids in the local environment as found in your 
structure.

Perhaps some of the propKa literature might also be helpful?

HTH,

Dave






David Briggs
about.me/david_briggs




On 13 August 2014 12:40, Careina Edgooms 
<02531c126adf-dmarc-requ...@jiscmail.ac.uk>
 wrote:
Sorry for off topic question, just wondering if anyone has come across a study 
that shows the residue pka of certain amino acids is different in vitro 
compared to in vivo?

Best
Careina

Re: [ccp4bb] thin shell Rfree set selection

[Ed Pozharski]

By all means, try it both ways and see whether the R-Rfree gap narrows with 
random vs thin shell selection. Depending on resolution and data quality, you 
may also consider imposing NCS restraints.


Sent on a Sprint Samsung Galaxy S® III

 Original message From: Xianchi Dong 
 Date:08/12/2014  4:45 PM  (GMT-05:00) 
To: CCP4BB@JISCMAIL.AC.UK Subject: [ccp4bb] thin shell 
Rfree set selection 
Dear all,
I have a dataset of C2 symmetry and 2 molecules in the ASU. I am wondering if I 
have to use thin shell Rfree set selection to avoid Rfree bias by NCS. And 
which Rfree selection method is better?

Thanks in advance.

Xianchi

Re: [ccp4bb] off topic

[Boaz Shaanan]

Hi Careina,


Regardless of what I think of this reviewer's comment, the only technique of which I'm aware that can measure in vivo (and in vitro of course) the pKa's of amino-acids is NMR. Perhaps you can find some NMR literature where in-vivo and in-vitro pKa's have been
 compared? Just a thought. 


Good luck with the reviewer.


             Boaz


 
 
Boaz Shaanan, Ph.D.

Dept. of Life Sciences  
Ben-Gurion University of the Negev  
Beer-Sheva 84105    
Israel  
    
E-mail: bshaa...@bgu.ac.il
Phone: 972-8-647-2220  Skype: boaz.shaanan  
Fax:   972-8-647-2992 or 972-8-646-1710
 
 








From: CCP4 bulletin board [CCP4BB@JISCMAIL.AC.UK] on behalf of Careina Edgooms [02531c126adf-dmarc-requ...@jiscmail.ac.uk]
Sent: Wednesday, August 13, 2014 3:45 PM
To: 
CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] off topic





thank you, yes I am aware of  propka and I know very well that the local environment of the amino acid is what contributes to its pka based on things like whether it is buried or exposed or involved in H-bonding. I am in the process of responding
 to a reviewer who suggests but does not give references that  in vivo conditions can alter the ionisation of some amino acids differently to in vitro. I am struggling to respond to this without some sort of proof or study on the matter

C






On Wednesday, August 13, 2014 2:38 PM, David Briggs  wrote:






Dear Careina, 


Following on from what Herman said, if you have a structure you can use the propKa server


http://propka.ki.ku.dk/



to predict pKas for amino acids in the local environment as found in your structure. 


Perhaps some of the propKa literature might also be helpful? 


HTH, 


Dave








 







 


David Briggs

about.me/david_briggs




 








On 13 August 2014 12:40, Careina Edgooms 
<02531c126adf-dmarc-requ...@jiscmail.ac.uk> wrote:



Sorry for off topic question, just wondering if anyone has come across a study that shows the residue pka of certain amino acids is different in vitro compared to in vivo?



Best


Careina

[ccp4bb] PSDI 2014, Abstract and Registration Info

[Anna Gardberg]

[image: Inline image 1]

*PROTEIN STRUCTURE DETERMINATION IN INDUSTRY 2014*



The 22nd PSDI will take place at the Hotel Miragem in Cascais, Portugal, a
coastal town located 30 kilometers west of Lisbon, Portugal.



The meeting will start with registration and an opening reception on
Sunday, 2nd November.  This will be followed by two and a half days of
scientific program starting Monday morning and finishing at noon on
Wednesday, 5th November. A half day excursion to Lisbon is scheduled as a
part of the meeting.



The scientific program will include the following sessions:

·Drug Discovery Stories

·Challenging Targets

·Biophysics Supporting Structural Biology

·Binding Kinetics in Drug Discovery

·Structural Biology of Biopharmaceuticals



Registration for participants and exhibitors is open at the website:
www.psdi2014.org

*Please note the deadline for registration and payment: 31 Aug 2014*



If you are interested in speaking in any of the scientific sessions, please
contact the organizers at psdi2...@merckgroup.com with an abstract.
Abstracts for “Binding Kinetics in Drug Discovery” are particularly
encouraged.



Additional information concerning the conference and further updates can be
found at the PSDI2014 website . Vendor/Sponsor
opportunities are available.



If you have any questions, please contact the organizers at
psdi2...@merckgroup.com.



Please feel free to forward this announcement to anyone who may be
interested in participating.



We look forward to a great conference this fall.



With kind regards,



The PSDI 2014 organizing committee:

·Tiago Bandeiras, IBET, Portugal

·Anna Gardberg, EMD Serono, USA

·Theresa Johnson, EMD Serono, USA

·Beate Matlok, Merck KGaA, Germany

·Djordje Musil, Merck KGaA, Germany



psdi2...@merckgroup.com

http://www.psdi2014.org

Re: [ccp4bb] AW: [ccp4bb] RMSD between structures of homologous proteins

[George Devaniranjan]

Thanks to the mailing list I came across this:
http://www.ncbi.nlm.nih.gov/pubmed/24167157

The C code to compute is available on the link mentioned in the paper.

Best wishes,
George


On Wed, Aug 13, 2014 at 8:01 AM, Eugene Krissinel <
eugene.krissi...@stfc.ac.uk> wrote:

> It is probably Gesamt/SSM (Superpose in CCP4), or PDBeFold at EBI rather
> than PISA -- Eugene
>
> On 13 Aug 2014, at 12:33, Eleanor Dodson wrote:
>
> I think PISA does this for you - it overlaps structural features and gives
> an RMSD on those parts that fit and a useful list of matching residues.
> You can run it from CCP4I or at PDBe.
>
> If you ant more than CA RMSD then you will need to select out the spans to
> fit and use the LSQKAB overlap procedure
> FIT MAIN RESi n n+m CHAIN A
> MATCH RESI b b+m CHAIN C
>
> FIT ….
>
> etc
>
> for example - you can do that in the GUI
>
> Eleanor
>
>
> On 13 August 2014 02:48,  herman.schreu...@sanofi.com>> wrote:
> Dear Sreetama,
>
> I use an ancient superposition program which rejects all atom pairs
> deviating more than 3 sigma and repeats this procedure until convergence.
> It then reports the RMSD for all atoms and for the atoms deviating less
> than 3 sigma. I find this an excellent method to separate the core from
> variable loop regions. It also ensures a robust superposition of the cores,
> ignoring variable loops.
>
> Best,
> Herman
>
> Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK CCP4BB@JISCMAIL.AC.UK>] Im Auftrag von sreetama das
> Gesendet: Mittwoch, 13. August 2014 08:12
> An: CCP4BB@JISCMAIL.AC.UK
> Betreff: [ccp4bb] RMSD between structures of homologous proteins
>
> Dear all,
>When calculating the RMSD between structures of homologous
> proteins, where there are large changes in the loop region(s), which RMSD
> should be reported - an overall value which may be inflated due to the
> deviations in the loops, or separate values for the core and loop regions?
> What is the best way to calculate the RMSD for superposition of the cores
> - should I prepare a separate PDB file by removing the coordinates of the
> loop residues and then superpose?
>
> Thanks in advance,
> Sreetama das,
> PhD student,
> Physics, IISc
>
>
>
> --
> Scanned by iCritical.
>
>

Re: [ccp4bb] off topic

[Careina Edgooms]

thank you, yes I am aware of  propka and I know very well that the local 
environment of the amino acid is what contributes to its pka based on things 
like whether it is buried or exposed or involved in H-bonding. I am in the 
process of responding to a reviewer who suggests but does not give references 
that  in vivo conditions can alter the ionisation of some amino acids 
differently to in vitro. I am struggling to respond to this without some sort 
of proof or study on the matter
C


On Wednesday, August 13, 2014 2:38 PM, David Briggs  
wrote:
 


Dear Careina, 

Following on from what Herman said, if you have a structure you can use the 
propKa server

http://propka.ki.ku.dk/


to predict pKas for amino acids in the local environment as found in your 
structure. 

Perhaps some of the propKa literature might also be helpful? 

HTH, 

Dave


  
   David Briggs
about.me/david_briggs 
  


On 13 August 2014 12:40, Careina Edgooms 
<02531c126adf-dmarc-requ...@jiscmail.ac.uk> wrote:

Sorry for off topic question, just wondering if anyone has come across a study 
that shows the residue pka of certain amino acids is different in vitro 
compared to in vivo?
>
>
>Best
>Careina

Re: [ccp4bb] off topic

[David Briggs]

Dear Careina,

Following on from what Herman said, if you have a structure you can use the
propKa server

http://propka.ki.ku.dk/

to predict pKas for amino acids in the local environment as found in your
structure.

Perhaps some of the propKa literature might also be helpful?

HTH,

Dave


[image: David Briggs on about.me]

David Briggs
about.me/david_briggs
  


On 13 August 2014 12:40, Careina Edgooms <
02531c126adf-dmarc-requ...@jiscmail.ac.uk> wrote:

> Sorry for off topic question, just wondering if anyone has come across a
> study that shows the residue pka of certain amino acids is different in
> vitro compared to in vivo?
>
> Best
> Careina
>

[ccp4bb] AW: [ccp4bb] off topic

[Herman . Schreuder]

Dear Careina,
The pka of amino acids is not dependent on in vivo/in vitro, but on the local 
environment, e.g. free in solution vs. part of a folded protein. As part of a 
folded protein the shifts will be specific for the particular protein and I am 
not aware of any general rules.
Best,
Herman

Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von Careina 
Edgooms
Gesendet: Mittwoch, 13. August 2014 13:41
An: CCP4BB@JISCMAIL.AC.UK
Betreff: [ccp4bb] off topic

Sorry for off topic question, just wondering if anyone has come across a study 
that shows the residue pka of certain amino acids is different in vitro 
compared to in vivo?

Best
Careina

Re: [ccp4bb] software/web server to determine ligand volume

[Patrick Loll]

VOIDOO will do this


On 13 Aug 2014, at 2:06 AM, sreetama das wrote:

> Dear all,
> Is there any software or web server available to calculate the volume of a 
> ligand if the ligand coordinates are provided?
> Google seems to come up only with options to calculate protein cavity volume.
> 
> Thanks in advance,
> Sreetama Das,
> phd student,
> Physics, IISc




---
Patrick J. Loll, Ph. D.  
Professor of Biochemistry & Molecular Biology
Director, Biochemistry Graduate Program
Drexel University College of Medicine
Room 10-102 New College Building
245 N. 15th St., Mailstop 497
Philadelphia, PA  19102-1192  USA

(215) 762-7706
pjl...@gmail.com
pl...@drexelmed.edu

Re: [ccp4bb] Twinning in space group Pc

[Mark J van Raaij]

how many (quasi) equivalent positions do you think the chiral compound can 
adopt in the organo-metal framework?
Or the organo-metal framework in the crystal?
If it's two, you can probably do alternate conformations - if there are more, 
it could become impossible, even in P1.

Mark J van Raaij
Lab 20B
Dpto de Estructura de Macromoleculas
Centro Nacional de Biotecnologia - CSIC
c/Darwin 3
E-28049 Madrid, Spain
tel. (+34) 91 585 4616
http://www.cnb.csic.es/~mjvanraaij







On 13 Aug 2014, at 11:00, Kristof Van Hecke wrote:

> Dear, 
> 
> I’m struggling with the following (small molecule) problem:
> 
> We are trying to solve the structure of a metal-organic framework containing 
> a chiral compound. 
> The space group is most probably Pc, but when refining, SHELX gives the error 
> “Possible racemic twin or wrong absolute structure - try TWIN refinement”. 
> As we know our compound is enantiopure, a racemic twin is very unlikely. In 
> this regard, also a centro-symmetric space group is not possible (although 
> CrysAlisPro always gives P2/c as the proper space group). As a matter of 
> fact, trying different space groups is not solving the problem. 
> 
> The second problem is that half of the structure is visible, but the other 
> half is completely not clear. Refinement is not possible at all (R-value of 
> 33%).
> When running TwinRotMat (Platon), I get the following possible 2-fold twin 
> axes:
> 
> 2-axis (   0   1  -1 ) [  -2   5  -4 ], Angle () [] =  2.31 Deg, Freq =14
>*
> (-0.992   -0.0190.015)   (h1)   (h2)   Nr Overlap =84
> (-0.4300.075   -0.860) * (k1) = (k2) BASF =  0.96
> ( 0.459   -1.146   -0.083)   (l1)   (l2)DEL-R =-0.064
> 
> 2-axis (   0   1   1 ) [   2   5   4 ], Angle () [] =  2.31 Deg, Freq =15
>*
> (-0.9920.0190.015)   (h1)   (h2)   Nr Overlap =   229
> ( 0.4300.0750.860) * (k1) = (k2) BASF =  0.94
> ( 0.4591.146   -0.083)   (l1)   (l2)DEL-R =-0.050
> 
> 2-axis (   1   0  -2 ) [   3   0  -2 ], Angle () [] =  0.54 Deg, Freq =19
>*
> (-0.1360.000   -0.576)   (h1)   (h2)   Nr Overlap =   992
> ( 0.000   -1.0000.000) * (k1) = (k2) BASF =  0.86
> (-1.7030.0000.136)   (l1)   (l2)DEL-R =-0.030
> 
> 2-axis (   1   2  -1 ) [   5   5  -1 ], Angle () [] =  0.39 Deg, Freq =13
>*
> (-0.3800.620   -0.124)   (h1)   (h2)   Nr Overlap =   854
> ( 1.2560.256   -0.251) * (k1) = (k2) BASF =  0.88
> (-0.617   -0.617   -0.877)   (l1)   (l2)DEL-R =-0.019
> 
> However, none of these do actually improve the refinement. 
> 
> 
> Has anyone encountered possible twinning/twin laws in Pc please?
> Or any other suggestions are most welcome?
> 
> 
> Thank you very much
> 
> Kristof

Re: [ccp4bb] [ccp4bb] Twinning in space group Pc

[Kristof Van Hecke]

Dear, 

Indeed,. Pc is not compatible with chiral molecules,. my mistake. 
I’m trying to process/solve the structure in P1 now and see how far I get. 

Thank you very much for pointing things out. 

Regards

Kristof


On 13 Aug 2014, at 13:58, herman.schreu...@sanofi.com wrote:

> Dear Kristof,
>  
> Lijun is right, space group Pc is not compatible with chiral molecules. Maybe 
> diffraction of your non-chiral metal structure overwhelmed the chiral 
> contribution of your organic framework. Why not use a trick from protein 
> crystallography: process and solve your structure in P1? According to the 
> international tables there are two asymmetric units in Pc, so the 
> crystallographic problem should remain manageable.
>  
> Best,
> Herman
>  
> Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von Lijun 
> Liu
> Gesendet: Mittwoch, 13. August 2014 11:59
> An: CCP4BB@JISCMAIL.AC.UK
> Betreff: Re: [ccp4bb] Twinning in space group Pc
>  
> Hi, Pc may not be the space group for your crystal, if the molecule is 
> chiral.  Seems like the data were forced to be reduced to a mirror_related 
> SG.  Lijun
> 
> On Aug 13, 2014 2:00 AM, "Kristof Van Hecke"  
> wrote:
> Dear,
> 
> I’m struggling with the following (small molecule) problem:
> 
> We are trying to solve the structure of a metal-organic framework containing 
> a chiral compound.
> The space group is most probably Pc, but when refining, SHELX gives the error 
> “Possible racemic twin or wrong absolute structure - try TWIN refinement”.
> As we know our compound is enantiopure, a racemic twin is very unlikely. In 
> this regard, also a centro-symmetric space group is not possible (although 
> CrysAlisPro always gives P2/c as the proper space group). As a matter of 
> fact, trying different space groups is not solving the problem.
> 
> The second problem is that half of the structure is visible, but the other 
> half is completely not clear. Refinement is not possible at all (R-value of 
> 33%).
> When running TwinRotMat (Platon), I get the following possible 2-fold twin 
> axes:
> 
> 2-axis (   0   1  -1 ) [  -2   5  -4 ], Angle () [] =  2.31 Deg, Freq =14
> *
> (-0.992   -0.0190.015)   (h1)   (h2)   Nr Overlap =84
> (-0.4300.075   -0.860) * (k1) = (k2) BASF =  0.96
> ( 0.459   -1.146   -0.083)   (l1)   (l2)DEL-R =-0.064
> 
> 2-axis (   0   1   1 ) [   2   5   4 ], Angle () [] =  2.31 Deg, Freq =15
> *
> (-0.9920.0190.015)   (h1)   (h2)   Nr Overlap =   229
> ( 0.4300.0750.860) * (k1) = (k2) BASF =  0.94
> ( 0.4591.146   -0.083)   (l1)   (l2)DEL-R =-0.050
> 
> 2-axis (   1   0  -2 ) [   3   0  -2 ], Angle () [] =  0.54 Deg, Freq =19
> *
> (-0.1360.000   -0.576)   (h1)   (h2)   Nr Overlap =   992
> ( 0.000   -1.0000.000) * (k1) = (k2) BASF =  0.86
> (-1.7030.0000.136)   (l1)   (l2)DEL-R =-0.030
> 
> 2-axis (   1   2  -1 ) [   5   5  -1 ], Angle () [] =  0.39 Deg, Freq =13
> *
> (-0.3800.620   -0.124)   (h1)   (h2)   Nr Overlap =   854
> ( 1.2560.256   -0.251) * (k1) = (k2) BASF =  0.88
> (-0.617   -0.617   -0.877)   (l1)   (l2)DEL-R =-0.019
> 
> However, none of these do actually improve the refinement.
> 
> 
> Has anyone encountered possible twinning/twin laws in Pc please?
> Or any other suggestions are most welcome?
> 
> 
> Thank you very much
> 
> Kristof

Re: [ccp4bb] AW: [ccp4bb] RMSD between structures of homologous proteins

[Eugene Krissinel]

It is probably Gesamt/SSM (Superpose in CCP4), or PDBeFold at EBI rather than 
PISA -- Eugene

On 13 Aug 2014, at 12:33, Eleanor Dodson wrote:

I think PISA does this for you - it overlaps structural features and gives an 
RMSD on those parts that fit and a useful list of matching residues.
You can run it from CCP4I or at PDBe.

If you ant more than CA RMSD then you will need to select out the spans to fit 
and use the LSQKAB overlap procedure
FIT MAIN RESi n n+m CHAIN A
MATCH RESI b b+m CHAIN C

FIT ….

etc

for example - you can do that in the GUI

Eleanor


On 13 August 2014 02:48, 
mailto:herman.schreu...@sanofi.com>> wrote:
Dear Sreetama,

I use an ancient superposition program which rejects all atom pairs deviating 
more than 3 sigma and repeats this procedure until convergence. It then reports 
the RMSD for all atoms and for the atoms deviating less than 3 sigma. I find 
this an excellent method to separate the core from variable loop regions. It 
also ensures a robust superposition of the cores, ignoring variable loops.

Best,
Herman

Von: CCP4 bulletin board 
[mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von 
sreetama das
Gesendet: Mittwoch, 13. August 2014 08:12
An: CCP4BB@JISCMAIL.AC.UK
Betreff: [ccp4bb] RMSD between structures of homologous proteins

Dear all,
   When calculating the RMSD between structures of homologous proteins, 
where there are large changes in the loop region(s), which RMSD should be 
reported - an overall value which may be inflated due to the deviations in the 
loops, or separate values for the core and loop regions?
What is the best way to calculate the RMSD for superposition of the cores - 
should I prepare a separate PDB file by removing the coordinates of the loop 
residues and then superpose?

Thanks in advance,
Sreetama das,
PhD student,
Physics, IISc



-- 
Scanned by iCritical.

[ccp4bb] AW: [ccp4bb] Twinning in space group Pc

[Herman . Schreuder]

Dear Kristof,

Lijun is right, space group Pc is not compatible with chiral molecules. Maybe 
diffraction of your non-chiral metal structure overwhelmed the chiral 
contribution of your organic framework. Why not use a trick from protein 
crystallography: process and solve your structure in P1? According to the 
international tables there are two asymmetric units in Pc, so the 
crystallographic problem should remain manageable.

Best,
Herman

Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von Lijun Liu
Gesendet: Mittwoch, 13. August 2014 11:59
An: CCP4BB@JISCMAIL.AC.UK
Betreff: Re: [ccp4bb] Twinning in space group Pc


Hi, Pc may not be the space group for your crystal, if the molecule is chiral.  
Seems like the data were forced to be reduced to a mirror_related SG.  Lijun
On Aug 13, 2014 2:00 AM, "Kristof Van Hecke" 
mailto:kristofrg.vanhe...@gmail.com>> wrote:
Dear,

I’m struggling with the following (small molecule) problem:

We are trying to solve the structure of a metal-organic framework containing a 
chiral compound.
The space group is most probably Pc, but when refining, SHELX gives the error 
“Possible racemic twin or wrong absolute structure - try TWIN refinement”.
As we know our compound is enantiopure, a racemic twin is very unlikely. In 
this regard, also a centro-symmetric space group is not possible (although 
CrysAlisPro always gives P2/c as the proper space group). As a matter of fact, 
trying different space groups is not solving the problem.

The second problem is that half of the structure is visible, but the other half 
is completely not clear. Refinement is not possible at all (R-value of 33%).
When running TwinRotMat (Platon), I get the following possible 2-fold twin axes:

2-axis (   0   1  -1 ) [  -2   5  -4 ], Angle () [] =  2.31 Deg, Freq =14
*
(-0.992   -0.0190.015)   (h1)   (h2)   Nr Overlap =84
(-0.4300.075   -0.860) * (k1) = (k2) BASF =  0.96
( 0.459   -1.146   -0.083)   (l1)   (l2)DEL-R =-0.064

2-axis (   0   1   1 ) [   2   5   4 ], Angle () [] =  2.31 Deg, Freq =15
*
(-0.9920.0190.015)   (h1)   (h2)   Nr Overlap =   229
( 0.4300.0750.860) * (k1) = (k2) BASF =  0.94
( 0.4591.146   -0.083)   (l1)   (l2)DEL-R =-0.050

2-axis (   1   0  -2 ) [   3   0  -2 ], Angle () [] =  0.54 Deg, Freq =19
*
(-0.1360.000   -0.576)   (h1)   (h2)   Nr Overlap =   992
( 0.000   -1.0000.000) * (k1) = (k2) BASF =  0.86
(-1.7030.0000.136)   (l1)   (l2)DEL-R =-0.030

2-axis (   1   2  -1 ) [   5   5  -1 ], Angle () [] =  0.39 Deg, Freq =13
*
(-0.3800.620   -0.124)   (h1)   (h2)   Nr Overlap =   854
( 1.2560.256   -0.251) * (k1) = (k2) BASF =  0.88
(-0.617   -0.617   -0.877)   (l1)   (l2)DEL-R =-0.019

However, none of these do actually improve the refinement.


Has anyone encountered possible twinning/twin laws in Pc please?
Or any other suggestions are most welcome?


Thank you very much

Kristof

Re: [ccp4bb] Twinning in space group Pc

[Harry Powell]

Hi folks

Pc *must* have both enantiomers, since it's got a glide plane ( =  
mirror + translation parallel to mirror).


So the sample *cannot* be enantiopure if the space group is Pc (or P2/ 
c)...


BTW, Pc isn't a centrosymmetric space group.

Unless I'm wrong...

On 13 Aug 2014, at Wed13 Aug 12:37, Eleanor Dodson wrote:

Twin laws are possible if there are 2 ways to index your cell, and  
non-merefedral twinning is possible in any system depending on the  
cell.


I am not sure of the small molecule tools to check twinning though.
Eleanor Dodson



On 13 August 2014 05:58, Lijun Liu  wrote:
Hi, Pc may not be the space group for your crystal, if the molecule  
is chiral.  Seems like the data were forced to be reduced to a  
mirror_related SG.  Lijun


On Aug 13, 2014 2:00 AM, "Kristof Van Hecke"  
 wrote:

Dear,

I’m struggling with the following (small molecule) problem:

We are trying to solve the structure of a metal-organic framework  
containing a chiral compound.
The space group is most probably Pc, but when refining, SHELX gives  
the error “Possible racemic twin or wrong absolute structure - try  
TWIN refinement”.
As we know our compound is enantiopure, a racemic twin is very  
unlikely. In this regard, also a centro-symmetric space group is  
not possible (although CrysAlisPro always gives P2/c as the proper  
space group). As a matter of fact, trying different space groups is  
not solving the problem.


The second problem is that half of the structure is visible, but  
the other half is completely not clear. Refinement is not possible  
at all (R-value of 33%).
When running TwinRotMat (Platon), I get the following possible 2- 
fold twin axes:


2-axis (   0   1  -1 ) [  -2   5  -4 ], Angle () [] =  2.31 Deg,  
Freq =14

*
(-0.992   -0.0190.015)   (h1)   (h2)   Nr  
Overlap =84
(-0.4300.075   -0.860) * (k1) = (k2)  
BASF =  0.96
( 0.459   -1.146   -0.083)   (l1)   (l2)DEL- 
R =-0.064


2-axis (   0   1   1 ) [   2   5   4 ], Angle () [] =  2.31 Deg,  
Freq =15

*
(-0.9920.0190.015)   (h1)   (h2)   Nr  
Overlap =   229
( 0.4300.0750.860) * (k1) = (k2)  
BASF =  0.94
( 0.4591.146   -0.083)   (l1)   (l2)DEL- 
R =-0.050


2-axis (   1   0  -2 ) [   3   0  -2 ], Angle () [] =  0.54 Deg,  
Freq =19

*
(-0.1360.000   -0.576)   (h1)   (h2)   Nr  
Overlap =   992
( 0.000   -1.0000.000) * (k1) = (k2)  
BASF =  0.86
(-1.7030.0000.136)   (l1)   (l2)DEL- 
R =-0.030


2-axis (   1   2  -1 ) [   5   5  -1 ], Angle () [] =  0.39 Deg,  
Freq =13

*
(-0.3800.620   -0.124)   (h1)   (h2)   Nr  
Overlap =   854
( 1.2560.256   -0.251) * (k1) = (k2)  
BASF =  0.88
(-0.617   -0.617   -0.877)   (l1)   (l2)DEL- 
R =-0.019


However, none of these do actually improve the refinement.


Has anyone encountered possible twinning/twin laws in Pc please?
Or any other suggestions are most welcome?


Thank you very much

Kristof



Harry
--
** note change of address **
Dr Harry Powell, MRC Laboratory of Molecular Biology, Francis Crick  
Avenue, Cambridge Biomedical Campus, Cambridge, CB2 0QH
Chairman of European Crystallographic Association SIG9  
(Crystallographic Computing)

[ccp4bb] off topic

[Careina Edgooms]

Sorry for off topic question, just wondering if anyone has come across a study 
that shows the residue pka of certain amino acids is different in vitro 
compared to in vivo?

Best
Careina

Re: [ccp4bb] Twinning in space group Pc

[Eleanor Dodson]

Twin laws are possible if there are 2 ways to index your cell, and
non-merefedral twinning is possible in any system depending on the cell.

I am not sure of the small molecule tools to check twinning though.
Eleanor Dodson



On 13 August 2014 05:58, Lijun Liu  wrote:

> Hi, Pc may not be the space group for your crystal, if the molecule is
> chiral.  Seems like the data were forced to be reduced to a mirror_related
> SG.  Lijun
> On Aug 13, 2014 2:00 AM, "Kristof Van Hecke" 
> wrote:
>
>> Dear,
>>
>> I’m struggling with the following (small molecule) problem:
>>
>> We are trying to solve the structure of a metal-organic framework
>> containing a chiral compound.
>> The space group is most probably Pc, but when refining, SHELX gives the
>> error “Possible racemic twin or wrong absolute structure - try TWIN
>> refinement”.
>> As we know our compound is enantiopure, a racemic twin is very unlikely.
>> In this regard, also a centro-symmetric space group is not possible
>> (although CrysAlisPro always gives P2/c as the proper space group). As a
>> matter of fact, trying different space groups is not solving the problem.
>>
>> The second problem is that half of the structure is visible, but the
>> other half is completely not clear. Refinement is not possible at all
>> (R-value of 33%).
>> When running TwinRotMat (Platon), I get the following possible 2-fold
>> twin axes:
>>
>> 2-axis (   0   1  -1 ) [  -2   5  -4 ], Angle () [] =  2.31 Deg, Freq =
>>  14
>> *
>> (-0.992   -0.0190.015)   (h1)   (h2)   Nr Overlap =
>>  84
>> (-0.4300.075   -0.860) * (k1) = (k2) BASF =
>>  0.96
>> ( 0.459   -1.146   -0.083)   (l1)   (l2)DEL-R
>> =-0.064
>>
>> 2-axis (   0   1   1 ) [   2   5   4 ], Angle () [] =  2.31 Deg, Freq =
>>  15
>> *
>> (-0.9920.0190.015)   (h1)   (h2)   Nr Overlap =
>> 229
>> ( 0.4300.0750.860) * (k1) = (k2) BASF =
>>  0.94
>> ( 0.4591.146   -0.083)   (l1)   (l2)DEL-R
>> =-0.050
>>
>> 2-axis (   1   0  -2 ) [   3   0  -2 ], Angle () [] =  0.54 Deg, Freq =
>>  19
>> *
>> (-0.1360.000   -0.576)   (h1)   (h2)   Nr Overlap =
>> 992
>> ( 0.000   -1.0000.000) * (k1) = (k2) BASF =
>>  0.86
>> (-1.7030.0000.136)   (l1)   (l2)DEL-R
>> =-0.030
>>
>> 2-axis (   1   2  -1 ) [   5   5  -1 ], Angle () [] =  0.39 Deg, Freq =
>>  13
>> *
>> (-0.3800.620   -0.124)   (h1)   (h2)   Nr Overlap =
>> 854
>> ( 1.2560.256   -0.251) * (k1) = (k2) BASF =
>>  0.88
>> (-0.617   -0.617   -0.877)   (l1)   (l2)DEL-R
>> =-0.019
>>
>> However, none of these do actually improve the refinement.
>>
>>
>> Has anyone encountered possible twinning/twin laws in Pc please?
>> Or any other suggestions are most welcome?
>>
>>
>> Thank you very much
>>
>> Kristof
>>
>

Re: [ccp4bb] AW: [ccp4bb] RMSD between structures of homologous proteins

[Eleanor Dodson]

I think PISA does this for you - it overlaps structural features and gives
an RMSD on those parts that fit and a useful list of matching residues.
You can run it from CCP4I or at PDBe.

If you ant more than CA RMSD then you will need to select out the spans to
fit and use the LSQKAB overlap procedure
FIT MAIN RESi n n+m CHAIN A
MATCH RESI b b+m CHAIN C

FIT ….

etc

for example - you can do that in the GUI

Eleanor


On 13 August 2014 02:48,  wrote:

>  Dear Sreetama,
>
>
>
> I use an ancient superposition program which rejects all atom pairs
> deviating more than 3 sigma and repeats this procedure until convergence.
> It then reports the RMSD for all atoms and for the atoms deviating less
> than 3 sigma. I find this an excellent method to separate the core from
> variable loop regions. It also ensures a robust superposition of the cores,
> ignoring variable loops.
>
>
>
> Best,
>
> Herman
>
>
>
> *Von:* CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] *Im Auftrag von
> *sreetama das
> *Gesendet:* Mittwoch, 13. August 2014 08:12
> *An:* CCP4BB@JISCMAIL.AC.UK
> *Betreff:* [ccp4bb] RMSD between structures of homologous proteins
>
>
>
> Dear all,
>
>When calculating the RMSD between structures of homologous
> proteins, where there are large changes in the loop region(s), which RMSD
> should be reported - an overall value which may be inflated due to the
> deviations in the loops, or separate values for the core and loop regions?
>
> What is the best way to calculate the RMSD for superposition of the cores
> - should I prepare a separate PDB file by removing the coordinates of the
> loop residues and then superpose?
>
>
>
> Thanks in advance,
>
> Sreetama das,
>
> PhD student,
>
> Physics, IISc
>

Re: [ccp4bb] Twinning in space group Pc

[Lijun Liu]

Hi, Pc may not be the space group for your crystal, if the molecule is
chiral.  Seems like the data were forced to be reduced to a mirror_related
SG.  Lijun
On Aug 13, 2014 2:00 AM, "Kristof Van Hecke" 
wrote:

> Dear,
>
> I’m struggling with the following (small molecule) problem:
>
> We are trying to solve the structure of a metal-organic framework
> containing a chiral compound.
> The space group is most probably Pc, but when refining, SHELX gives the
> error “Possible racemic twin or wrong absolute structure - try TWIN
> refinement”.
> As we know our compound is enantiopure, a racemic twin is very unlikely.
> In this regard, also a centro-symmetric space group is not possible
> (although CrysAlisPro always gives P2/c as the proper space group). As a
> matter of fact, trying different space groups is not solving the problem.
>
> The second problem is that half of the structure is visible, but the other
> half is completely not clear. Refinement is not possible at all (R-value of
> 33%).
> When running TwinRotMat (Platon), I get the following possible 2-fold twin
> axes:
>
> 2-axis (   0   1  -1 ) [  -2   5  -4 ], Angle () [] =  2.31 Deg, Freq =
>  14
> *
> (-0.992   -0.0190.015)   (h1)   (h2)   Nr Overlap =
>  84
> (-0.4300.075   -0.860) * (k1) = (k2) BASF =
>  0.96
> ( 0.459   -1.146   -0.083)   (l1)   (l2)DEL-R
> =-0.064
>
> 2-axis (   0   1   1 ) [   2   5   4 ], Angle () [] =  2.31 Deg, Freq =
>  15
> *
> (-0.9920.0190.015)   (h1)   (h2)   Nr Overlap =
> 229
> ( 0.4300.0750.860) * (k1) = (k2) BASF =
>  0.94
> ( 0.4591.146   -0.083)   (l1)   (l2)DEL-R
> =-0.050
>
> 2-axis (   1   0  -2 ) [   3   0  -2 ], Angle () [] =  0.54 Deg, Freq =
>  19
> *
> (-0.1360.000   -0.576)   (h1)   (h2)   Nr Overlap =
> 992
> ( 0.000   -1.0000.000) * (k1) = (k2) BASF =
>  0.86
> (-1.7030.0000.136)   (l1)   (l2)DEL-R
> =-0.030
>
> 2-axis (   1   2  -1 ) [   5   5  -1 ], Angle () [] =  0.39 Deg, Freq =
>  13
> *
> (-0.3800.620   -0.124)   (h1)   (h2)   Nr Overlap =
> 854
> ( 1.2560.256   -0.251) * (k1) = (k2) BASF =
>  0.88
> (-0.617   -0.617   -0.877)   (l1)   (l2)DEL-R
> =-0.019
>
> However, none of these do actually improve the refinement.
>
>
> Has anyone encountered possible twinning/twin laws in Pc please?
> Or any other suggestions are most welcome?
>
>
> Thank you very much
>
> Kristof
>

[ccp4bb] Twinning in space group Pc

[Kristof Van Hecke]

Dear, 

I’m struggling with the following (small molecule) problem:

We are trying to solve the structure of a metal-organic framework containing a 
chiral compound. 
The space group is most probably Pc, but when refining, SHELX gives the error 
“Possible racemic twin or wrong absolute structure - try TWIN refinement”. 
As we know our compound is enantiopure, a racemic twin is very unlikely. In 
this regard, also a centro-symmetric space group is not possible (although 
CrysAlisPro always gives P2/c as the proper space group). As a matter of fact, 
trying different space groups is not solving the problem. 

The second problem is that half of the structure is visible, but the other half 
is completely not clear. Refinement is not possible at all (R-value of 33%).
When running TwinRotMat (Platon), I get the following possible 2-fold twin axes:

2-axis (   0   1  -1 ) [  -2   5  -4 ], Angle () [] =  2.31 Deg, Freq =14
*
(-0.992   -0.0190.015)   (h1)   (h2)   Nr Overlap =84
(-0.4300.075   -0.860) * (k1) = (k2) BASF =  0.96
( 0.459   -1.146   -0.083)   (l1)   (l2)DEL-R =-0.064
 
2-axis (   0   1   1 ) [   2   5   4 ], Angle () [] =  2.31 Deg, Freq =15
*
(-0.9920.0190.015)   (h1)   (h2)   Nr Overlap =   229
( 0.4300.0750.860) * (k1) = (k2) BASF =  0.94
( 0.4591.146   -0.083)   (l1)   (l2)DEL-R =-0.050
 
2-axis (   1   0  -2 ) [   3   0  -2 ], Angle () [] =  0.54 Deg, Freq =19
*
(-0.1360.000   -0.576)   (h1)   (h2)   Nr Overlap =   992
( 0.000   -1.0000.000) * (k1) = (k2) BASF =  0.86
(-1.7030.0000.136)   (l1)   (l2)DEL-R =-0.030
 
2-axis (   1   2  -1 ) [   5   5  -1 ], Angle () [] =  0.39 Deg, Freq =13
*
(-0.3800.620   -0.124)   (h1)   (h2)   Nr Overlap =   854
( 1.2560.256   -0.251) * (k1) = (k2) BASF =  0.88
(-0.617   -0.617   -0.877)   (l1)   (l2)DEL-R =-0.019
 
However, none of these do actually improve the refinement. 


Has anyone encountered possible twinning/twin laws in Pc please?
Or any other suggestions are most welcome?


Thank you very much

Kristof

Re: [ccp4bb] software/web server to determine ligand volume

[Harry Powell]

Hi

A rough calculation is 18Å^3 per non-hydrogen atom in the ligand - I  
use a pencil and a bit of paper for the mechanics of this!


On 13 Aug 2014, at Wed13 Aug 07:06, sreetama das wrote:


Dear all,
Is there any software or web server available to calculate the  
volume of a ligand if the ligand coordinates are provided?
Google seems to come up only with options to calculate protein  
cavity volume.


Thanks in advance,
Sreetama Das,
phd student,
Physics, IISc


Harry
--
** note change of address **
Dr Harry Powell, MRC Laboratory of Molecular Biology, Francis Crick  
Avenue, Cambridge Biomedical Campus, Cambridge, CB2 0QH
Chairman of European Crystallographic Association SIG9  
(Crystallographic Computing)

[ccp4bb] AW: [ccp4bb] Protein_Ligand Suggestion

[Herman . Schreuder]

Dear Monica,

here are some comments from my side:
-one can ask the bulletin board, or just decrease the occupancy during 
refinement and see what happens. If the statistics get better: keep it, if they 
get worse: reject the result.
-I think it is not a question of correct or incorrect structures. You have two 
structures: one with a short soak and low ligand concentration, and one with a 
longer soak and higher ligand concentration, but with 88% complete data. I 
would refine both structures independently and in both cases refine a group 
occupancy for the ligand. If this is done correctly, I would expect that both 
structures will look very similar and provide independent proof for the binding 
mode of your ligand. One question is: what to do with ambiguities in the 
low-occupancy structure? I would look in the high-occupancy structure for hints 
how to resolve these, but more puritan crystallographers may have different 
opinions. Also 88% completeness is not great, but also not that bad. If the 
electron density looks good, I would trust it.

However, if both independently refined structures show a very different binding 
mode, I would get suspicious. In this case I would not just ignore the 
low-occupancy structure but would try to find out what happened: was the ligand 
incorrectly fitted? Where the soaking conditions (precipitant, buffer, pH) 
different, could it be that the ligand was degraded/turned over during the long 
soak?

Best,
Herman

Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von Monica 
Mittal
Gesendet: Mittwoch, 13. August 2014 07:48
An: CCP4BB@JISCMAIL.AC.UK
Betreff: [ccp4bb] Protein_Ligand Suggestion

Dear all

I have solved a structure of protein with ligand at 2.7 and 2.8 A. Both have 
ligand bound in it as i soaked it.

For the first structure i soaked for lesser time and lesser concentration of 
ligand and solved it at 2.7 A resolution. Although the ligand has good density 
(2Fo-Fc at 0.9 sigma) but it is little unconnected from its middle portion. Its 
RSCC value is 0.76 and B-factor of 70.0. Overall completeness of structure is > 
91%. So If i decrease the occupancy of ligand here, will it improve the ligand 
statistics esp. the Real space R-value??

However for the second structure, i soaked for longer time and higher ligand 
concentration, so the ligand is perfect with clear 2Fo-Fc map at 1.2 sigma and 
Fo-Fc map at 2.8 sigma along with well defined simulated annealing and 
composite omit maps. The RSCC value is 0.81 with B-factor of 40.1.The only 
issue is the overall completeness of the structure is 88%. But if i get 
evrything OK at this completeness, den can i consider the second structure with 
ligand as the correct one??

Thanks in advance.
Monica