NESG Wiki - User contributions [en]

Protein purification

2013-03-08T17:02:21Z

Bwu:

== Samples for NMR ==

For each protein, we usually make the following samples:

*[''U''-15N, 13C]-labeled sample, for resonance assignment and NOE interpretation.
*

A U-15N and fractional 5%-13C-labeled sample for stereospecific assignment of Val and Leu isopropyl methyl groups based on the method of Neri et. al. (Biochemistry 28, 7510-7516,1989) that we call the NC5 sample.  This sample is prepared using 5% ''U''[1H,13C]-D-glucose and 95% unlabelled glucose in the ''E. coli'' minimal  growth media.

 The NC5 sample is not [''U''-5%-13C,''U''-15N] since it is not uniformly 5% -13C labeled.  The incorporation of 13C in the methyl groups is biosynthetically directed by the ''E. coli''. 

This sample can be called: 

           ''U''-15N, 5%
biosynthetically directed 13C (NC5) sample or

            ''U''-15N, fractional 5%-13C-labeled (NC5) sample  

and give the Neri reference.

 For RDC measurement:

*A secondary NC5 sample for RDC measurement. 

(Alternatively, for the H-N RDC measurement, a [U-15N]-labeled sample can be used. Rutger's uses a second NC5 sample, since this is all ready being prepared for the stereospecific assignments, and a 13C CHSQC can be compared with the [U-15N, 13C]-labeled sample, to check if the samples are the same.)

 For each protein that is a dimer in solution an extra sample may be required in addition to the samples above:

*1:1 mixture of natural abundance and [U-15N, 13C]-labeled samples, for intermolecular NOE interpretation.

 

== Typical Rutgers University NMR Buffers ==

The protein production facility at Rutgers University uses these three NMR buffers for the initial protein screening: <ref>Snyder, D, et.al. (2005). “Comparisons of NMR spectral quality and success in crystallization demonstrate that NMR and X-ray crystallography are complementary methods for small protein structure determination.” ''JACS'', '''127:''' 16505-16511. [http://www.ncbi.nlm.nih.gov/pubmed/16305237 pmid = 16305237] </ref>: 

 

*pH 4.5 NMR buffer:  20 mM NH4OAc, 100 mM NaCl, 10 mM DTT, 5 mM CaCl2, 0.02% NaN3, 5% D2O
*pH 5.5 NMR buffer:  20 mM NH4OAc, 100 mM NaCl, 10 mM DTT, 5 mM CaCl2, 0.02% NaN3, 5% D2O
*pH 6.5 NMR buffer:  20 mM MES, 100 mM NaCl, 5 mM CaCl2, 10 mM DTT, 0.02% NaN3, 5% D2O

 

== Typical Rutgers University Protein Purification Protocol ==

''E.coli'' BL21(DE3) are fermented in MJ9 medium <ref>Jansson M, Li YC, Jendeberg L, Anderson S, Montelione GT, Nilsson B (1996) High-level production of uniformly 15N- and 13C-enriched fusion proteins in ''Escherichia coli''. J Biomol NMR 7: 131-141</ref>. Cell pellets are stored at -20º C.

#add 30 ml Lysis buffer  to the a frozen cell pellet and thaw.
#sonicate in ice bath
#centrifuge to remove insoluble part
#supernatant is added to an AkTAxpressTM system with a His TrapHP column followed by HiLoad16/60 Superdex 75 gel filtration chromatography.
#exchange buffer to screening buffer by concentrating, diluting with new buffer, reconcentrating to 0.3 - 1.0 mM with Amicon ultrafiltration concentrator (Millipore). 

== Typical University of Toronto (Arrowsmith proteomics NMR lab) NMR Buffers ==

'''15N-labelled Glucose-based autoinduction media''' : 

This is for 15N-labelling of samples for screening so we can take advantage of autoinduction during the screening phase when we are dealing with a large number of samples, but be able to switch to IPTG-induction with 13C-glucose as the sole carbon source for the smaller number of targets prepared for full NMR data collection, with high reproducibility. 

For each sample, dissolve 6.8 g of Na2HPO4, 3g of KH2PO4, 0.5g of NaCl, and 0.6g of 15N-labelled NH4Cl, in 500 mL water and autoclave the media.  

In a separate small beaker or falcon tube, mix the following: 1 mL of 1M MgSO4 stock solution; 5 mg of biotin; 5 mg of thiamine.HCl; 10 uL of 1M ZnSO4 stock solution; 100 uL of 1M CaCl2 stock solution; 2.5 g of glucose; 1 g of lactose and enough water to make 10 mL. Stir or shake until the sugar dissolves. Filter sterilize through a 0.2 micron syringe filter into the autoclaved 15N-labelled solution.

Follow the same fermentation protocol as with IPTG-induction method, except there is no need to monitor the cell density and omit the addition of IPTG. Grow the cells at 37C throughout. 

 

Standard screening NMR buffers are:

 

{| cellspacing="1" cellpadding="1" border="1" style="width: 548px; height: 238px;"
|-
| name
| buffer
| NaCl concentration
| pH
|-
| A5.0n300zd
| Sodium acetate
| 300 mM
| 5.0
|-
| Bt6.0n450zd
| bis-tris
| 450 mM
| 6.0
|-
| Bt6.5n450zd
| bis-tris
| 450 mM
| 6.5
|-
| M6.5n450zd
| MOPS
| 450 mM
| 6.5
|-
| P6.5n450zd
| Sodium phosphate
| 450 mM
| 6.5
|-
| P7.0n450zd
| Sodium phosphate
| 450 mM
| 7.0
|-
| H7.0n450zd
| HEPES
| 450 mM
| 7.0
|-
| H7.5n450zd
| HEPES
| 450 mM
| 7.5
|-
| T7.7n500zd
| tris
| 500 mM
| 7.7
|}

 

All standard screening buffers contain 10 mM of the buffer, 10 uM ZnSO4, 10 mM DTT, 0.01% NaN3, 1 mM benzamidine, 1x inhibitor cocktail, 5% D2O. 

 

The final NMR buffer for [U-15N, 13C]-labeled sample depends on the protein of interest. all NMR buffers always contain : 0.01 % NaN3, 1 mM benzamidine, 1x inhibitor cocktail.

(1) If the protein has no cysteine in the sequence, do not bother to add ZnSO4 and DTT (Zn ion will just be a nuisance and deuterated DTT is expensive).

== Typical University of Toronto Protein Purification Protocol ==

Required buffers:

i015t8.5n500z : 15 mM imidazole, 10 mM tris, pH 8.5, 500 mM NaCl, 10 uM ZnSO4

i030t8.5n500z : 30 mM imidazole, 10 mM tris, pH 8.5, 500 mM NaCl, 10 uM ZnSO4

i500t8.5n500z : 500 mM imidazole, 10 mM tris, pH 8.5, 500 mM NaCl, 10 uM ZnSO4

1M DTT

1M benzamidine 

#add 25 mL of i015t8.5n500z into a frozen cell pellet and thaw. 
#sonicate in ice bath
#spin down cell pellet
#transfer supernatant into new falcon tube and add 3 mL of nickel beads
#rock the tube for at least 30 minutes in cold.
#spin down the beads and discard the supernatant
#wash the beads with i015t8.5n500z twice and with i030t8.5n500z twice 
#in the final i030t8.5n500z wash,  pour the beads unto gravity filter column 
#elute the protein with i500t8.5n500z
#add benzamidine, and add DTT 
#exchange buffer by concentrating, diluting with new buffer, reconcentrating in a vivaspin concentrator. 

if it is [U-15N, 13C]-labeled, add step

(9a) put sample in dialysis bag with protease to cut his-tag and dialyse  against  cleavage buffer

(9b) pass the sample through nickel beads again, then follow step (10) above. 

== References ==

<references />

Protein purification

2013-01-09T22:21:30Z

Bwu:

== Samples for NMR ==

For each protein, we usually make the following samples:

*[''U''-15N, 13C]-labeled sample, for resonance assignment and NOE interpretation.
*

A U-15N and fractional 5%-13C-labeled sample for stereospecific assignment of Val and Leu isopropyl methyl groups based on the method of Neri et. al. (Biochemistry 28, 7510-7516,1989) that we call the NC5 sample.  This sample is prepared using 5% ''U''[1H,13C]-D-glucose and 95% unlabelled glucose in the ''E. coli'' minimal  growth media.

 The NC5 sample is not [''U''-5%-13C,''U''-15N] since it is not uniformly 5% -13C labeled.  The incorporation of 13C in the methyl groups is biosynthetically directed by the ''E. coli''. 

This sample can be called: 

           ''U''-15N, 5%
biosynthetically directed 13C (NC5) sample or

            ''U''-15N, fractional 5%-13C-labeled (NC5) sample  

and give the Neri reference.

 For RDC measurement:

*A secondary NC5 sample for RDC measurement. 

(Alternatively, for the H-N RDC measurement, a [U-15N]-labeled sample can be used. Rutger's uses a second NC5 sample, since this is all ready being prepared for the stereospecific assignments, and a 13C CHSQC can be compared with the [U-15N, 13C]-labeled sample, to check if the samples are the same.)

 For each protein that is a dimer in solution an extra sample may be required in addition to the samples above:

*1:1 mixture of natural abundance and [U-15N, 13C]-labeled samples, for intermolecular NOE interpretation.

 

== Typical Rutgers University NMR Buffers ==

The protein production facility at Rutgers University uses these three NMR buffers for the initial protein screening: <ref>Snyder, D, et.al. (2005). “Comparisons of NMR spectral quality and success in crystallization demonstrate that NMR and X-ray crystallography are complementary methods for small protein structure determination.” ''JACS'', '''127:''' 16505-16511. [http://www.ncbi.nlm.nih.gov/pubmed/16305237 pmid = 16305237] </ref>: 

 

*pH 4.5 NMR buffer:  20 mM NH4OAc, 100 mM NaCl, 10 mM DTT, 5 mM CaCl2, 0.02% NaN3, 5% D2O
*pH 5.5 NMR buffer:  20 mM NH4OAc, 100 mM NaCl, 10 mM DTT, 5 mM CaCl2, 0.02% NaN3, 5% D2O
*pH 6.5 NMR buffer:  20 mM MES, 100 mM NaCl, 5 mM CaCl2, 10 mM DTT, 0.02% NaN3, 5% D2O

 

== Typical Rutgers University Protein Purification Protocol ==

''E.coli'' BL21(DE3) are fermented in MJ9 medium <ref>Jansson M, Li YC, Jendeberg L, Anderson S, Montelione GT, Nilsson B (1996) High-level production of uniformly 15N- and 13C-enriched fusion proteins in ''Escherichia coli''. J Biomol NMR 7: 131-141</ref>. Cell pellets are stored at -20º C.

#add 30 ml Lysis buffer  to the a frozen cell pellet and thaw.
#sonicate in ice bath
#centrifuge to remove insoluble part
#supernatant is added to an AkTAxpressTM system with a His TrapHP column followed by HiLoad16/60 Superdex 75 gel filtration chromatography.
#exchange buffer to screening buffer by concentrating, diluting with new buffer, reconcentrating to 0.3 - 1.0 mM with Amicon ultrafiltration concentrator (Millipore). 

== Typical University of Toronto (Arrowsmith proteomics NMR lab) NMR Buffers ==

<meta content="text/html; charset=utf-8" http-equiv="CONTENT-TYPE"><title></title> <meta content="OpenOffice.org 3.1 (Linux)" name="GENERATOR"><style type="text/css"> </style> <meta content="text/html; charset=utf-8" http-equiv="CONTENT-TYPE"><title></title> <meta content="OpenOffice.org 3.1 (Linux)" name="GENERATOR"><style type="text/css"> </style>

'''15N-labelled Glucose-based autoinduction media''' : </meta></meta> This is for 15N-labelling of samples for screening so we can take advantage of autoinduction during the screening phase when we are dealing with a large number of samples, but be able to switch to IPTG-induction with 13C-glucose as the sole carbon source for the smaller number of targets prepared for full NMR data collection, with high reproducibility. <meta content="text/html; charset=utf-8" http-equiv="CONTENT-TYPE"><title></title> <meta content="OpenOffice.org 3.1 (Linux)" name="GENERATOR"><style type="text/css"> </style>

 For each sample, dissolve 6.8 g of Na2HPO4, 3g of KH2PO4, 0.5g of NaCl, and 0.6g of 15N-labelled NH4Cl, in 500 mL water and autoclave the media.  <meta content="text/html; charset=utf-8" http-equiv="CONTENT-TYPE"><title></title> <meta content="OpenOffice.org 3.1 (Linux)" name="GENERATOR"><style type="text/css">In a separate small beaker or falcon tube, mix the following: 1 mL of 1M MgSO4 stock solution; 5 mg of biotin; 5 mg of thiamine.HCl; 10 uL of 1M ZnSO4 stock solution; 100 uL of 1M CaCl2 stock solution; 2.5 g of glucose; 1 g of lactose and enough water to make 10 mL. Stir or shake until the sugar dissolves. Filter sterilize through a 0.2 micron syringe filter into the autoclaved 15N-labelled solution.</style></meta></meta> <meta content="text/html; charset=utf-8" http-equiv="CONTENT-TYPE"><title></title> <meta content="OpenOffice.org 3.1 (Linux)" name="GENERATOR"><style type="text/css">n a separate small beaker or falcon tube, mix the following: 1 mL of 1M MgSO4 stock solution; 5 mg of biotin; 5 mg of thiamine.HCl; 10 uL of 1M ZnSO4 stock solution; 100 uL of 1M CaCl2 stock solution; 2.5 g of glucose; 1 g of lactose and enough water to make 10 mL. Stir or shake until the sugar dissolves. Filter sterilize through a 0.2 micron syringe filter into the autoclaved 15N-labelled solution.</style></meta></meta></meta></meta></meta></meta> <meta content="text/html; charset=utf-8" http-equiv="CONTENT-TYPE"><title></title> <meta content="OpenOffice.org 3.1 (Linux)" name="GENERATOR"><style type="text/css"> </style> In a separate small beaker or falcon tube, mix the following: 1 mL of 1M MgSO4 stock solution; 5 mg of biotin; 5 mg of thiamine.HCl; 10 uL of 1M ZnSO4 stock solution; 100 uL of 1M CaCl2 stock solution; 2.5 g of glucose; 1 g of lactose and enough water to make 10 mL. Stir or shake until the sugar dissolves. Filter sterilize through a 0.2 micron syringe filter into the autoclaved 15N-labelled solution.

 

Follow the same fermentation protocol as with IPTG-induction method, except there is no need to monitor the cell density and omit the addition of IPTG. Grow the cells at 37C throughout. </meta></meta>

 

Standard screening NMR buffers are:

 

{| cellspacing="1" cellpadding="1" border="1" style="width: 548px; height: 238px;"
|-
| name
| buffer
| NaCl concentration
| pH
|-
| A5.0n300zd
| Sodium acetate
| 300 mM
| 5.0
|-
| Bt6.0n450zd
| bis-tris
| 450 mM
| 6.0
|-
| Bt6.5n450zd
| bis-tris
| 450 mM
| 6.5
|-
| M6.5n450zd
| MOPS
| 450 mM
| 6.5
|-
| P6.5n450zd
| Sodium phosphate
| 450 mM
| 6.5
|-
| P7.0n450zd
| Sodium phosphate
| 450 mM
| 7.0
|-
| H7.0n450zd
| HEPES
| 450 mM
| 7.0
|-
| H7.5n450zd
| HEPES
| 450 mM
| 7.5
|-
| T7.7n500zd
| tris
| 500 mM
| 7.7
|}

 

All standard screening buffers contain 10 mM of the buffer, 10 uM ZnSO4, 10 mM DTT, 0.01% NaN3, 1 mM benzamidine, 1x inhibitor cocktail, 5% D2O. 

 

The final NMR buffer for [U-15N, 13C]-labeled sample depends on the protein of interest. all NMR buffers always contain : 0.01 % NaN3, 1 mM benzamidine, 1x inhibitor cocktail.

(1) If the protein has no cysteine in the sequence, do not bother to add ZnSO4 and DTT (Zn ion will just be a nuisance and deuterated DTT is expensive).

== Typical University of Toronto Protein Purification Protocol ==

Required buffers:

i015t8.5n500z : 15 mM imidazole, 10 mM tris, pH 8.5, 500 mM NaCl, 10 uM ZnSO4

i030t8.5n500z : 30 mM imidazole, 10 mM tris, pH 8.5, 500 mM NaCl, 10 uM ZnSO4

i500t8.5n500z : 500 mM imidazole, 10 mM tris, pH 8.5, 500 mM NaCl, 10 uM ZnSO4

1M DTT

1M benzamidine 

#add 25 mL of i015t8.5n500z into a frozen cell pellet and thaw. 
#sonicate in ice bath
#spin down cell pellet
#transfer supernatant into new falcon tube and add 3 mL of nickel beads
#rock the tube for at least 30 minutes in cold.
#spin down the beads and discard the supernatant
#wash the beads with i015t8.5n500z twice and with i030t8.5n500z twice 
#in the final i030t8.5n500z wash,  pour the beads unto gravity filter column 
#elute the protein with i500t8.5n500z
#add benzamidine, and add DTT 
#exchange buffer by concentrating, diluting with new buffer, reconcentrating in a vivaspin concentrator. 

if it is [U-15N, 13C]-labeled, add step

(9a) put sample in dialysis bag with protease to cut his-tag and dialyse  against  cleavage buffer

(9b) pass the sample through nickel beads again, then follow step (10) above. 

== References ==

<references />

Protein purification

2011-10-05T20:48:48Z

Bwu:

File:Nmrbuffer nesgwiki aayee 111005.doc

2011-10-05T20:39:09Z

Bwu:

Structure Refinement Using CNS Energy Minimization With Explicit Water

2011-03-28T16:09:15Z

Bwu:

== '''Introduction''' ==

Crystallography & NMR System (CNS) is a program has been designed to provide a flexible multi-level hierachical approach for the most commonly used algorithms in macromolecular structure determination [1]. Functions include heavy atom searching, experimental phasing (including MAD and MIR), density modification, crystallographic refinement with maximum likelihood targets, and NMR structure calculation using NOEs, J-coupling, chemical shift, and dipolar coupling data.

The latest version is 1.21. Please check the website [http://cns-online.org/ http://cns-online.org] for additional information.

== '''CNS Refinement Protocol''' ==

The final structure from CYANA should be refined using CNS energy minimization with explicit water before PDB deposition.

The common input files are:

*PDB coordinates
*NOE constraints
*Dihedral angle constraints
*Hydrogen bond constraints

==== '''Converting input files from CYANA to XPLOR/CNS''' ====

If the final structure calculation was performed with CYANA 2.1, then the constraint and coordinate files must be first converted to XPLOR/CNS format.

==== '''Using cyana2cns.cya script''' ====

This procedure can be used to prepare input files to use with the WaterRefCNS script (see below).

#Download [[Media:Cyana2cns.cya|cyana2cns.cya]] into directory with the final structure calculation.
#Modify the <tt>read pdb</tt>, <tt>read upl</tt> and <tt>read aco</tt> lines if necessary.
#Start CYANA 2.1 and run <tt>cyana2cns.cya</tt>.

*All conformers will be stored in a single PDB file <tt>KKK.pdb</tt>. If you plan run water bath refinement manually, use the <tt>p2X</tt> program instead.
*The resulting NOE constraint file <tt>KKK_noe.tbl</tt> will have all lower limits set to 0. If want them to be set according to VdW radii, use the <tt>r2X</tt> program instead. Experience shows that there is no significant effect on geometry and quality scores as reported by PSVS.

==== '''Using programs p2X and r2X and d2X''' ====

These programs were written by Alex Lemak at the University of Toronto. They are typically used to prepare input files for manual CNS water bath refinement.

Go to <tt>~1_projects/targetID/structure/cns/</tt> directory and <tt>conversion_scripts</tt> sub directory, convert required input files to CNS format from CYANA format by using macro <tt>mkfil</tt>: 
<pre> p2X cyana2.pdb pref
r2X found-c.upl noeupl.tbl
r2X gr-hbonds.upl hbonds.tbl
cat noeupl.tbl hbonds.tbl > noe.tbl
d2X found-c.aco aco.tbl
</pre>
*<tt>p2X splits conformers from a pdb file generated by CYANA 2.1 into individual pdb files (one file per conformer) and converts atom names to CNS format. The second argument "pref"</tt> specifies a common prefix name for the output pdb files. It should be no more than 4 characters long.
*<tt>r2X</tt> converts distance constraints (noe/hbond) from CYANA 2.1 format to CNS format. The lower limits in the resulting constraint file are set according to VdW radii.
**Distance constraints obtained from CYANA macro <tt>caliba</tt> or <tt>calibration</tt> can be used as input files.
**Distance constraints obtained from CYANA macro <tt>peak calibrate</tt> should have the pseudo-atom correction run before used as input files, run following commands under CYANA: <pre>read upl final.upl
</pre>

distance correct write upl final_corrected.upl This, however, will not add corrections for multiplicity!

*<tt>d2X</tt>, to run: <tt>d2X cyana2.aco cns.tbl</tt> This will convert angle constraints from CYANA format to CNS format. If you don't have any dihedral angle constrains, create an empty file.

Both p2X and r2X require a translation table file [[Media:AtomtransC.tbl|atomtransC.tbl]].

If have used simplified pseudoatom names (H* instead of Q*) with option pseudo=2 in CYANA 2.1, you may first want to them before running p2X and r2X: 
<pre>pseudo=0
read upl noe.upl
read pdb cyana2.pdb
write upl out.upl
write pdb out.pdb all
</pre>
Note that pseudo=0 should be set '''before''' loading PDB and constraint files.

==== '''Conversion from DYANA''' ====

Use CYANA macro <tt>MigrateFromDyanaCyana1.cya</tt> to convert of data files from standard CYANA1.x or DYANA nomenclature to the standard IUPAC nomenclature used by CYANA, then do then same thing as above. 
<pre> translate dyana # use Cyana 1.x/Dyana nomenclature
read seq demo-dyana.seq # read sequence from Cyana 1.x/Dyana
read upl demo-dyana.upl unknown=warn # read upper distance limits from Cyana 1.x/Dyana
read aco demo-dyana.aco unknown=warn # read angle restraints from Cyana 1.x/Dyana
translate off # return to standard (IUPAC) nomenclature
write demo-cyana.seq # save sequence in standard Cyana format
write upl demo-cyana.upl # save upper distance limits in standard Cyana format
write aco demo-cyana.aco # save angle restraints in standard Cyana format
</pre>
 

=== '''Running CNS using WaterRefCNS script at CABM''' ===

The WaterRefCNS script, developed at Rutgers University by Roberto Tejero, performs many of the above CNS water refinement protocols in an easy to use script with many user controled options.

To see a list of options type the command:
<pre> WaterRefCNS
</pre>
You will see a list of options and how the program is used.
<pre> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
WaterRefCN -- a tool to launch structure refinement with water
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Usage: /farm/software/WaterRefinement_cns/WaterRefCNS -na NamProt -que [PBS|No] -pro [xpl|cns]

** Full set of options [not all are needed for a perfect run, use defaults]

-na NamProt Name of the protein NOTE: this is mandatory, no default value
-qu [PBS|NO] Que system to use, one of PBS or NOQUE def NOQUE
-pr [cns|xpl] Protocol to use, X-POR, CNS def cns
-av [center|sum] average for the distance restraints (for CNS, XPLOR) def sum
-ci n1,n2 CIS pep info (PRO res num) in comma separated list, i.e 56,89
-ss n1-n2,n3-n4 S-S bridge info in comma separated, dash-separated pairs, i.e. 2-24,30-40
-he Display this help message
-hisd n1,n2 HISD pep info (HIS res num) in comma separated list, i.e 12,43
-hise n1,n2 HISE pep info (HIS res num) in comma separated list, i.e 25,67
-heat N number of cycles in heating stage, default is 200
-hot N number of cycles in HOT stage, default is 1000
-cool N number of cycles for cooling stage, default is 100
-sc N Scale all terms (noe, dihe, hbond) by xN times def 1
-par string Choice of nonbonded params, one of OPLSX|PARAM19|PARMALLH6|PROLSQ|CONTACT
default is OPLSX
-seed <int> Seed for random number generator, default 31415

Examples:
/farm/software/WaterRefinement_cns/WaterRefCNS -na SR358 -que no -pr cns -ci 22

</pre>
Given a protein name KKK, here are the input files required to run WaterRefCNS:

*  pdb coordinates (in a single file) prepared in Xplor format using PdbStat.  The name of the file should be:  <tt>KKK.pdb</tt>
*  an noe.tbl file in CNS format.  You can convert CYANA a final.upl file using PdbStat.  We typically add 10-15% to the upper limit and make the lower limit vdw.  The name of this file should be:  <tt>KKK_noe.tbl</tt>
*  a dihe.tbl file in CNS format.  Again, you can convert a CYANA .aco file using PdbStat.  The name of this file should be:  <tt>KKK_dihe.tbl</tt>
*  an hbond.tbl file in CNS format (optional).  The name of this file should be:  <tt>KKK_bhond.tbl</tt>

 

In order to run WaterRefCNS on the CABM-NMR cluster:

#Log into hummer or master3
#go to the directory with the above 4 files
#type:
<pre> /farm/software/WaterRefinement_cns/WaterRefCNS -na KKK -que PBS</pre>
and that's it.  If you have some CIS residue you add <tt>-ci</tt> NUMRES to the options above. If you want to have better vdw violation, you can add <tt>-par PARAM19</tt> to the options above.

'''Future work:'''

Currently, the WaterRefCNS script does not support refinement with RDC's. 

 

=== '''WaterRefCNS at SUNY Buffalo''' ===

The CABM water refinement package was customized to run on computers at UB:

*Fixed a bug in <tt>WaterRefCNS</tt> so that the assembled refined PDB structure file has a <tt>.pdb</tt> extension.
*PBS queue submission code in <tt>WaterRefCNS</tt> modified for use on U2 cluster
*Added <tt>-hisd</tt> and <tt>-hise</tt> options to use neutral histidine isoprotomers. This required modification of <tt>WaterRefCNS</tt>, <tt>cns_refine_h2o.inp</tt> and <tt>generate_h2o.inp</tt>.
*Fixed a bug in the protein topology file <tt>topallhdg5.3.pro</tt>. The line <tt>atom CD1 type=CR1E charge=0.130 end</tt> in the HISE patch section was removed.
*Fixed a bug that caused the order of conformers to be changed in the output PDB file by replacing the line <tt>$WATREFLIB/Agrupa *.pdb > All_${Name}_cns.pdb</tt> with <tt>$WATREFLIB/Agrupa `ls resa_?.pdb ; ls resa_??.pdb` > All_${Name}_cns.pdb</tt> in the <tt>WaterRefCNS</tt> script.

The modified package is installed in <tt>/nsm/chem/cen2/HTP2/3_src/WaterRefinement_cns</tt> on the local server and in <tt>/san/projects1/szypersk/src/WaterRefinement_cns</tt> on the U2 Linux cluster. Make sure that this directories is included in your path on <tt>spins*</tt> workstations and U2, respectively.

For reference, the download location is here: [[Media:WaterRefinement_UB.tar.Z|WaterRefinement_UB.tar.Z]]

The required files are:

*PDB coordinates (with atom names in XPLOR/CNS format)
*NOE constraints (in XPLOR/CNS format)
*Dihedral angle constraints (in XPLOR/CNS format)
*Hydrogen bond constraints (in XPLOR/CNS format)

Assuming that your coordinate file is called <tt>KKK.pdb</tt>, the constraint files should be named <tt>KKK_noe.tbl</tt>, <tt>KKK_dihe.tbl</tt> and <tt>KKK_hbond.tbl</tt>. If you don't have dihedral angle and/or hydrogen bond constraints, then create the corresponding empty files.

You may need to add a line like <tt>nassign=1000</tt> at the top of the constraint files. The value should be equal to or greater than the number of constraints in the file.

WaterRefCNS assumes "sum of r-6" averaging by default. When using "center"-averaged UPLs (e.g. from <tt>caliba</tt>) add <tt>-av center</tt> when starting WaterRefCNS. For information on averaging conventions for calibration in CYANA see [[NOE Calibration Using CYANA|NOE Calibration in CYANA]]

It is recommended to run it on U2 Linux cluster. Examples:

#<pre>WaterRefCNS -na KKK -que PBS</pre>
#<pre>WaterRefCNS -na KKK -que PBS -ci 21,49</pre>residues 21 and 49 are cis-Pro
#<pre>WaterRefCNS -na KKK -que PBS -hise 40,82 -hisd 32,65</pre>residues 40 and 82 are ε-protonated neutral His residues 32 and 65 are δ-protonated neutral His
#<pre>WaterRefCNS -na KKK -que PBS -av center </pre>for use with center-averaged calibration

To run on <tt>spins*</tt> Linux workstations without queue system use <tt>-que NO</tt> option instead of <tt>-que PBS</tt> option. The default is no queue system, so you can omit the <tt>-que</tt> option altogether.

==== '''Running WaterRefCNS''' ====

#Convert the coordinate and constraint files as described above.
#Copy the converted files into a CNS working directory on the U2 cluster or a workstation (e.g. <tt>structure/cns/calc1</tt>).
#Run the <tt>WaterRefCNS</tt> with the proper arguments.
#If the calculation was successful, the refined coordinate file will be stored in the <tt>refinedPDB</tt> subdirectory. Check the <tt>BeSureToREADME</tt> for details.

=== '''Running CNS "manually"''' ===

Go to <tt>~1_projects/targetID/structure/cns/</tt> directory and <tt>cns_scripts</tt> sub directory, follow the procedures as describe below.

'''1.  Generating MTF topology file from a PDB file. '''     modify the header of the input file <tt>generate_mtf.inp</tt> and then run <tt>cns < generate_mtf.inp > mtf.log</tt>.

*Modify input files path and name base on individual target as follows:
<pre> ...
{in} pdb_file="/san/user/gliu2/u2/cns/cns/convertion_scripts/ufc1_1.pdb";
{in} param_file="/san/user/gliu2/u2/cns/cns/cns_scripts/parallhdg5.3C.pro";
{in} topol_file="/san/user/gliu2/u2/cns/cns/cns_scripts/topallhdg5.3.pro";
{in} plink_file="/san/user/gliu2/u2/cns/cns/cns_scripts/protein.link";
{out} struct_file="ufc1_1.mtf";
...
</pre>
*Requires these files:
*[[Media:Parallhdg5.3C.pro|parallhdg5.3C.pro]]: CNS parameter file
*[[Media:Topallhdg5.3.pro|topallhdg5.3.pro]]: CNS topology file
*[[Media:Protein.link|protein.link]]: CNS peptide linkage definition file
*For <tt>cis-Proline</tt>, define the residue Id of the residue prior to <tt>cis-Proline</tt> by additional lines in <tt>generate_mtf.inp</tt> before the WRITE command
<pre> patch cisp
reference=nil=( resid 90 )
end
</pre>
*If you get an error message on the atom type, you also can try 
<pre> patch cipp
reference=nil=( resid 90 )
end </pre>
*For neutral Nε- and Nδ-protonated histidines add these lines in <tt>generate_mtf.inp</tt> before the WRITE command
<pre> patch hise
reference=nil=( resid 40 )
end
patch hisd
reference=nil=( resid 65 )
end </pre>
*For dimer refinement, add a line containing only <tt>TER</tt> between two monomer unit coordinates in the PDB conformer that is defined in <tt>generate_mtf_cis.inp</tt> as input file.

''' '''

'''2.  Rebuilding hydrogen atom positions for each structure. '''     Modify input file <tt>rebuild.inp</tt> and then run <tt>cns < rebuild.inp > rebuild.log</tt>

*Modify input files path and name base on induvidual target as follows:      
<pre> ...
evaluate ($mtf_file="ufc1_1.mtf")
evaluate ($pdbname_in="../convertion_scripts/ufc1")
evaluate ($pdbname_out="ufc1_rb")
evaluate ($number_of_struct= 20 )
...
evaluate ($topol_p_file="/san/user/gliu2/u2/cns/cns/cns_scripts/topallhdg5.3.pro")
evaluate ($param_p_file="/san/user/gliu2/u2/cns/cns/cns_scripts/parallhdg5.3C.pro")
...
</pre>
*Requires these files:
*[[Media:Parallhdg5.3C.pro|parallhdg5.3C.pro]]: CNS parameter file
*[[Media:Topallhdg5.3.pro|topallhdg5.3.pro]]: CNS topology file

'''3. Refining structures with explicit water.'''

*Running <tt>CNS</tt> on workstation with single processor, modify input file <tt>re_h2oc.inp</tt> and then run <tt>cns < re_h2oc.inp > h2o.log</tt>
*Modify input files path and name base on induvidual target as follows:
<pre> ...
evaluate ($mtf_file="ufc1_1.mtf")
evaluate ($noe_file="../noe.tbl")
evaluate ($dihe_file="../aco.tbl")
evaluate ($hb_file="../hbond.tbl")
evaluate ($pdbname_in="ufc1_rb")
evaluate ($pdbname_out="ufc1_ref")
evaluate ($number_of_struct= 20 )
...</pre>
*Modify the weight of energy contribution part if necessary, e.g. use 2x to 10x of the following scales:
<pre> scale ambi 50
scale dist 50
scale hbond 50
end
restraints dihedral
scale=200
end</pre>
*Requires
**[[Media:Parallhdg5.3C.pro|parallhdg5.3C.pro]]: CNS parameter file
**[[Media:Topallhdg5.3.pro|topallhdg5.3.pro]]: CNS topology file
**[[Media:Topallhdg5.3.sol|topallhdg5.3.sol]]: CNS solvent topology file
**[[Media:Parallhdg5.3.sol|parallhdg5.3.sol]]: CNS solvent parameter file
**[[Media:Boxtyp20.pdb|boxtyp20.pdb]]: Water box
**[[Media:Generate_water.cns|generate_water.cns]]: CNS protein solvation script
**[[Media:Print_coorheader.cns|print_coorheader.cns]]: CNS script
*Run CNS on workstation with multiple processors. CNS will refine each conformer independently and in parallel, it will be much faster if 20 conformer were refined in parallal at u2.ccr.buffalo.edu.

    1.  Run cns_convertion as descibed above        

    2.  Go to cns_scripts directory, modify input file <tt>generate_mtf_cis.inp</tt> and then run =cns < generate_mtf_cis.inp = as described above    

    3.  Run macro <tt>getfil</tt>
<pre> cp ../convertion_scripts/noe.tbl .
cp ../convertion_scripts/hbonds.tbl .
cp ../convertion_scripts/aco.tbl .
cp ufc1_1.mtf com/.</pre>
    4.  Modify input file <tt>rebuild.inp</tt> and then run <tt>cns < rebuild.inp</tt> as described above.

    5.  Modify input file <tt>re_h2oc.inp</tt> as described above, remember to set <tt>$number_of_struct = 20</tt>.

    6.  Run macro <tt>subaba</tt>

<code></code>
<pre> #bash
for i in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
do
mkdir ref$i
cd ref$i
cp ../com/* .
cp ../jan24/ufc1_rb_$i.pdb ufc1_rb_1.pdb
qsub cns.sc
cd ..

donedone</pre>
              where <tt>cns.sc</tt> is a PBS macro used to submit CNS job by using PBS:
<pre> #!/bin/csh
#PBS -m e
#PBS -q short_c
#PBS -l nodes=1:GM:ppn=1
#PBS -l walltime=00:56:00
#PBS -o out1
#PBS -j oe
#PBS -N clean1
cd $PBS_O_WORKDIR
echo "working directory = "$PBS_O_WORKDIR
set NN = `cat $PBS_NODEFILE | wc -l`
echo "NN = "$NN
# Run Job
cns < re_h2oc.inp
echo "ALL Done!"</pre>
    7.  Run <tt>getpdb</tt> to collect refined structures.
<pre> #bash
for i in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
do
cd ref$i
cp ufc1_ref_1.pdb ../pdb/ufc1_ref_$i.pdb
cp ufc1_ref_1.vio ../pdb/ufc1_ref_$i.vio
cd ..
done</pre>
    8. Run script [[Media:Agrupa.txt|Agrupa]] to put all 20 conformers into a single PDB file.<tt></tt>
<pre> Agrupa ufc1_ref*.pdb > ufc1all.pdb</pre>
=== '''Running CNS for Proteins with Metal Ions''' ===

The CNS refinement for proteins with metal ions can also be performed with the new WaterrefCNS script, using the appropriate metal ion nomenclature in both the pdb coordinates and distance constraints.

For manual CNS refinement please do the refinement as following:

1.  Set environment for CNS1.1 by run source /farm/users/gliu/alias.cns
<pre> alias cns1 /farm/software/cns/cns_solve_1.1/intel-i686-linux_g77/bin/cns
setenv CNS_TOPPAR /farm/data/gliu/cns1/</pre>
In addition to the topology and parameter files, the metal ion parameter file <tt>ion.top</tt> is required. An example can be found in "/farm/users/gliu/projects/cns_cuttha_cis" with all required input files. 2.  Prepared required files as described above (final.tbl, final cns format PDB files and put in xplorPDB dir with name as sa_#.pdb) except the PDB file should include the metal ions with format according to [http://www.nsm.buffalo.edu/Chem/HTP_twiki43/bin/view/NESG/CNS CNS] library <tt>ion.top</tt>. cp sa_1.pdb as  template.pdb, input files for creating mtf file Note that alignment is important. eg:
<pre>ATOM 1249 OT2 ALA 83 69.296 13.232 5.744 1.00 0.00
ATOM 1250 ZN+2 ZN2 150 63.086 13.789 -10.407 1.00 0.00 zinc</pre>
3.  Run generate_h2o.inp once to create temp_h2o.pdb and temp_h2o.mtf. The extra proton atom in the ligand residues, eg. HIS HD1 or CYS S, are removed by editing the generate_h2o.inp; cis proline is also defined here in the generate_h2o.inp (resid is the residue number prior the proline).
<pre>{* any special prosthetic group patches can be applied here *}
{===>}
delete select (name hg and resname cys and resid 61) end
delete select (name hg and resname cys and resid 85) end
delete select (name hd1 and resname his and resid 46) end
delete select (name he2 and resname his and resid 83) end
patch cisp
reference=1=( resid 13 )
end</pre>
4.  Edit generate_1.inp to remove the extra proton as did above.

5.  Run generate_20.com, this will run generate_#.inp 20 times, updating each pdb number and this creates cnsPDB/sa_cns_#.pdb

6.  Edit and run re_h2o_cu.inp, the refined pdb is kept in refinedPDB, or

7.  Use subcns to submit cns refinement by using PBS: eg, type "sh subcns". Before run subcns , make a folder " com" contains the following file. Type getpdb to get refined pdb files in refinedPDB after it finished.

*cns.sc: PBS submission
*cutc_h2o.mtf: mtf file created as descrive above
*topology and parameter files: parallhdg5.3C.pro, parallhdg5.3.pro, topallhdg5.3.pro
*re_h2o_cu.inp: input file for cns refinement 

 

=== '''Running CNS with RDC Constraints''' ===

Under construction

=== '''Running CNS refinement on Homodimers''' ===

Under construction

 

== '''CNS Refinement with RDC constraints, metal ions using FMCGUI at Arrowsmith lab, UofT''' ==

'''Step 1. Input files.'''

 



In order to perform water refinement using FMCGUI the following files should be prepared/available:

 

*structure file

*distance constraints file

*dihedral angles constraints file

 

User can include in the refinement the following additional constrains/data as well:

 

*hydrogen bond constraints file

*RDC data collected in different media; (up to 3 media are allowed)

*Zinc ions

 

 

#Structure file and constraints file format. 

It is expected that the structural, distance constraints, and dihedral angle constraints files are available as output of CYANA2.1 /CYANA3.0 structure calculations. The format of these files correspond to the format of'final.pdb', 'final.upl', and 'final.aco'ouput files. At the same time, these files could be loaded in FMCGUI in CNS/XPLOR format as well. It should be noted that structural file should contain key word MODEL that separate lines corresponding to different structural models.

''Example.''PDB file in CNS format:

 

[[Image:Wrf20.jpeg]]

  

  2. RDC data format 

The RDC data can be used in CYANA or PALES formats. One can use RDC data collected in up to 3 media. All data from the same media should be included in the same file. When data are prepared in PALES format than a file should contain data only from one media. When using CYANA3 format, data from different media can be included in one file(See Examples r1,and r2).

 

''Example r1.''RDC data in CYANA3 format.

 

This example present file with HN and NC rdc's measured in 2 different media.

 

[[Image:Wrf5.jpg]] 

 

 

''Example r2''. RDC data in PALES format.

This example present file with HN, CCA, and NC rdc's measured in 1 media.

[[Image:Wrf6.jpg]]

 

 3.  Zinc ligands file format. 

The Zinc-ligands file contains list of residues that coordinate each of Zinc ions present in the structure. There could be 3 or 4 Zinc ligands for one Zinc ion. Possible types of residues that coordinate Zinc ion are CYS, HIS, ASP, and GLU.  (See Example z1) 

 

''Example z1''. Zinc- ligands file.



[[Image:Wrf7.jpg]]



This file indicate that there are 2 Zinc ions, one Zinc is coordinated by 4 residues 21, 26, 31, and 34 while the other Zinc is coordinated by 3 residues 48, 52, and 62

This Zinc-file could be created using FMCGUI command ['''Structure>Add ZN ligands''']:

 

[[Image:Wrf8.jpg]] 

 

User have to type IDs of residues that ligate zinc ion into the popped up entry window “ZN ligands” (see Figure). If there are a few zinc ions, information for each ion should be provided in a separate row.

 

 

'''Step 2. Setting up water refinement calculations using FMCGUI.'''

 

2.1
Open new project ['''Project>new''']
or load the old one ['''Project>load''']

 

 

2.2   Select
command
['''Structure>Water Refinement>calculate''']





[[Image:Wrf14.jpg]]

 

 

The popped up “Water Refinement” window where user is asked to select a number of files with coordinates and constraints (see Figure). The entry fields marked in read (PDB file, Distance constraints, and Dihedral angle constraints) should be specified while the others could be left blank. In addition to coordinates and constraints, user can indicate cisProline residues (if there are any) and to specify singly protonated HIS residues that are double protonated by default.

 

[[Image:Wrf18.jpeg]] 

 

By pressing OK button in “water Refinement” window a new “Parameters” window

will pop up. User have to specify

*name of the water refinement directory

*flag for distance constraints transformation (could be 1 or 2). If the flag for distance constraints transformation is set to 2 then no change in the upper distance limit is performed. If the flag is set to 1 (default value), then the upper distance limits are modified by classifying each constraint in one of the 4 classes – strong, medium, weak and very weak

*force constants for RDC constraints (if there are any) on the final stage of the refinement protocol. 













 [[Image:Wrf10.jpg]] 

 

Pressing OK button starts the setup of water refinement calculations using provided information the specified directory.

 

[[Image:Wrf16.jpg]] 

 

 

The set up consists of the following operations:

*Splitting structure file in a number of separate files for each model. The atom names are changed from CYANA to CNS nomenclature.

*Transforming dihedral angle constraints from CYANA to CNS format. (If constraints are provided in CNS format then no transformation is performed).

*Transforming distance NOE constraints from CYANA format to CNS format. This transformation includes changing atom names from CYANA to CNS nomenclature. If the flag for distance constraints transformation is set to 2 then no change in the upper distance limit is performed. If the flag is set to 1 (default value), then the upper distance limits are modified by classifying each constraint in one of the 4 classes – strong, medium, weak and very weak.

*Transforming Hbond constraints (if any) from CYANA format to CNS format

*Generating mtf-file for CNS calculations

*"Repairing" each input structure in accord with the mtf-file 









*Generating RDC constraints, a separate file for each media

*Estimating Da and R of alignment tensor for each media. This is performed by fitting RDC data to each model structure using program PALES, with subsequent averaging of the fitting results.

*Generating CNS script "re_h2o*_tmpl.inp" for the refinement calculations


































 

 '''Step 3.  Water refinement calculations using CNS.'''

 

In a separate terminal, go to the refinement directory and run the nely generated python script
“run_refine.py”.

This script will start N jobs in parallel, where N is the number of structural models to refine. The number of structures and the number of processor to use in the calculations can be specified by making modification of the top lines in script “run_refine.py”.
If user want to reset the values for some refinement parameters, for
example the values of Da and R, it could be done by modifying
corresponding lines in the header of the CNS script
“re_h2o*_tmpl.inp”.


 

'''Step 4. Summarizing results of refinement using FMCGUI.''' 

Open
new project ['''Project>new''']
or load the old one ['''Project>load''']

 

Select
command ['''Structure>Water Refinement>summary''']                                                                                                          
 

[[Image:Wrf14.jpg]]







In the popped up “Water refinement summary” window user will be asked to specify residues used for structure superposition and to select the water refinement directory “WATDIR” were the refinement calculations were made.





 

[[Image:Wrf19.jpg]]





The refined structural models will be superimpose and combined in one file. The created summary reports values of different energy components and constraint violation statistic for each structural model. A new directory “WATDIR_results” will be created, where WATDIR is the name of selected water refinement directory. The new directory contains final superimposed coordinates, distance and dihedral angle constrains in a format suitable for PDB deposition, summary and constraint violations reports, files“final_cns_summary”
and “final_violations”
respectively. (See examples S1 and S2)

 

''Example'' S1. File "final_cns_summary"  



The file shows the total energy, energy components, and number of constraints violations for each refined structure as well as corresponding average over all structures values.

 

[[Image:Wrf12.jpg]] 

 

''ExampleS2.''File “final_violations”

The file contains a list of all constraints violated in at least one structural model. For each constraint the the number of structural models that violate the  constraint is shown.

 

[[Image:Wrf11.jpg]] 

 

 

 



== '''Files for Download''' ==

 

#[[Media:AtomtransC.tbl|atomtransC.tbl]]: atom name translation table

 

#[[Media:Agrupa.txt|Agrupa]]: script for concatenating pdb coordinate files into one file

 

#[[Media:Parallhdg5.3C.pro|parallhdg5.3C.pro]]: CNS parameter file

 

#[[Media:Topallhdg5.3.pro|topallhdg5.3.pro]]: CNS topology file

 

#[[Media:Topallhdg5.3.sol|topallhdg5.3.sol]]: CNS solvent topology file

 

#[[Media:Parallhdg5.3.sol|parallhdg5.3.sol]]: CNS solvent parameter file

 

#[[Media:Boxtyp20.pdb|boxtyp20.pdb]]: Water box

 

#[[Media:Protein.link|protein.link]]: CNS peptide linkage definition file

 

#[[Media:Generate_water.cns|generate_water.cns]]: CNS protein solvation script

 

#[[Media:Print_coorheader.cns|print_coorheader.cns]]: CNS script

 

#[[Media:Cns_cabm.tar.gz|cns_cabm.tar.gz]]: CABM CNS refinement protocol

 

#[[Media:WaterRefinement_UB.tar.Z|WaterRefinement_UB.tar.Z]]: CNS water refinement package - SUNY Buffalo

 

#[[Media:Cyana2cns.cya|cyana2cns.cya]]: CYANA script to convert coordinates and constrains to XPLOR/CNS format

 

#[[Media:Cu_cPro_cns_example.tar.gz|Cu_cPro_cns_example.tar.gz]]: Example of CNS refinment for protein with Cu and cis-Pro

 

#[[Media:Zn.tar|zn.tar]]: Exampleof CNS refinment for protein with zn (from Alex Lemark)

 

#[[Media:Er317_2zn_CNSw_example.tar.gz|er317_2zn_CNSw_example.tar.gz]]: Example of CNS refinment for protein with zn (from Theresa Ramelot)

== '''References''' ==

[http://www.ncbi.nlm.nih.gov/pubmed/9757107?itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_RVDocSum&ordinalpos=14 1.  Brünger, A.T., Adams, P.D., Clore, G.M., DeLano, W.L., Gros, P., Grosse-Kunstleve, R.W., Jiang, J.-S., Kuszewski, J., Nilges, M., Pannu, N.S., Read, R.J., Rice, L.M., Simonson, T. and Warren,G.L. (1998) Crystallography & NMR system: a new software suite for macromolecular structure determination. ''Acta Crystallogr. D54'', 905-921.]