Re: [phenixbb] Rigid body refinement question

Ralf W. Grosse-Kunstleve

29 Apr 2010 29 Apr '10

9:17 p.m.

Is it possible to specify individual residues of a chain to undergo rigid refinement while the rest of the chain does regular refinement?

No, sorry. Ralf

Show replies by date

Young-Jin Cho

29 Apr 29 Apr

9:40 p.m.

New subject: Rigid body refinement question

Hi Ralf, What if you do like this? "strategy=rigid_body+individual_sites" "sites.individual=chain A and resid 36" "sites.rigid_body=chain A" "sites.rigid_body=chain B" Young-Jin ----- Original Message ----- From: "Ralf W. Grosse-Kunstleve" To: "s stampfer" , [email protected] Sent: Thursday, April 29, 2010 5:17:27 PM GMT -05:00 US/Canada Eastern Subject: Re: [phenixbb] Rigid body refinement question

...

Is it possible to specify individual residues of a chain to undergo rigid refinement while the rest of the chain does regular refinement?

No, sorry. Ralf _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb -- Postdoctoral Fellow, Kern Lab Biochemistry Department/HHMI Brandeis University

Ralf W. Grosse-Kunstleve

10:49 p.m.

New subject: Rigid body refinement question

...

What if you do like this? "strategy=rigid_body+individual_sites" "sites.individual=chain A and resid 36" "sites.rigid_body=chain A" "sites.rigid_body=chain B"

During rigid-body refinement, the bonds/angles/dihedrals/nonbondeds to the non-refined sites are not restrained. This could seriously mess up the geometry. In addition, rigid-body refinement and individual refinement will only run alternatingly. I'd be surprised if the parameter combination above did much good, but you never know before you try it. Simultaneous refinement of individual sites and (small) rigid bodies with full use of restraints is on our long-term wish list. Ralf

Pavel Afonine

30 Apr 30 Apr

3:43 a.m.

New subject: Rigid body refinement question

...

During rigid-body refinement, the bonds/angles/dihedrals/nonbondeds to the non-refined sites are not restrained. This could seriously mess up the geometry.

Just to clarify: all the above parameters do not change for all atoms within the rigid group. They can get distorted at the border: say you split a chain into two rigid groups, then the bond corresponding to a hinge point where the one rigid group ends and the second one starts, can get distorted. You can do torsion angle SA which is constrained rigid body refinement: the whole molecules is split into tiny rigid bodies and they are refined without breaking the stereochemistry. Pavel.

Raja Dey

7:37 p.m.

New subject: Error in PDB Validation letter

Hi, I am getting two errors in pdb validation report as follows: I think I need to use B-DNA constraint in the refinement. This is a 2.4 A data. How can I do that in PHENIX in cammand line promt. Another thing is, I have a small molecule in the complex, which I called as BML. I think that's might be the problem. This is known as BML210. What I should call this compound? Should I also give a chain ID for this compound? See the errors given below. *** Covalent Angle Values: The overall RMS deviation for covalent angles relative to the standard dictionary is 1.4 degrees. The following table contains a list of the covalent bond angles greater than 6 times standard deviation. Deviation Residue Chain Sequence AT1 - AT2 - AT3 Bond Dictionary Standard Name ID Number Angle Value Deviation ------------------------------------------------------------------------------------------- -4.8 DA E 1 C1' - C2' - C3' 97.6 102.4 0.8 -5.8 DA E 2 C1' - C2' - C3' 96.6 102.4 0.8 3.3 DA E 2 N9 - C1' - O4' 111.6 108.3 0.3 -5.0 DC E 5 N1 - C1' - O4' 103.0 108.0 0.7 -5.1 DT E 6 C1' - C2' - C3' 97.3 102.4 0.8 -6.7 DT E 8 C1' - O4' - C4' 103.4 110.1 1.0 2.3 DT E 8 N1 - C1' - O4' 110.6 108.3 0.3 -4.5 DT E 11 N1 - C1' - O4' 103.5 108.0 0.7 -6.1 DA F 2 C1' - C2' - C3' 96.3 102.4 0.8 3.0 DA F 2 N9 - C1' - O4' 111.3 108.3 0.3 -5.3 DC F 5 N1 - C1' - O4' 102.7 108.0 0.7 -4.9 DT F 6 C1' - C2' - C3' 97.5 102.4 0.8 -4.5 DA F 10 N9 - C1' - O4' 103.5 108.0 0.7 -7.8 DA F 11 N9 - C1' - O4' 100.2 108.0 0.7 ==> The following residues have extra atoms: RES MOD#C SEQ ATOMS BML( 100) N1 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 O2 N2 C20 N3 C7 C8 C9 ERROR: dissociated residue(s): BML 100 Raja

Pavel Afonine

1 May 1 May

8:17 a.m.

New subject: Error in PDB Validation letter

Hi Raja, first of all I would have a look in .geo file that phenix.refine generates. Just run phenix.refine like this with your model: phenix.refine model.pdb data.mtz --dry-run and then inspect the .geo file for atoms you list below. Do the ideal values in phenix.refine are in close agreement with those in PDB? If not, you should tell us. If the ideal values in PDB and in .geo are close then you probably need to do one more round of refinement to achieve with more weight put on restraints. Pavel. On 4/30/10 12:37 PM, Raja Dey wrote:

...

Hi, I am getting two errors in pdb validation report as follows: I think I need to use B-DNA constraint in the refinement. This is a 2.4 A data. How can I do that in PHENIX in cammand line promt.

Another thing is, I have a small molecule in the complex, which I called as BML. I think that's might be the problem. This is known as BML210. What I should call this compound? Should I also give a chain ID for this compound? See the errors given below.

*** Covalent Angle Values:

The overall RMS deviation for covalent angles relative to the standard dictionary is 1.4 degrees.

The following table contains a list of the covalent bond angles greater than 6 times standard deviation.

Deviation Residue Chain Sequence AT1 - AT2 - AT3 Bond Dictionary Standard Name ID Number Angle Value Deviation ------------------------------------------------------------------------------------------- -4.8 DA E 1 C1' - C2' - C3' 97.6 102.4 0.8 -5.8 DA E 2 C1' - C2' - C3' 96.6 102.4 0.8 3.3 DA E 2 N9 - C1' - O4' 111.6 108.3 0.3 -5.0 DC E 5 N1 - C1' - O4' 103.0 108.0 0.7 -5.1 DT E 6 C1' - C2' - C3' 97.3 102.4 0.8 -6.7 DT E 8 C1' - O4' - C4' 103.4 110.1 1.0 2.3 DT E 8 N1 - C1' - O4' 110.6 108.3 0.3 -4.5 DT E 11 N1 - C1' - O4' 103.5 108.0 0.7 -6.1 DA F 2 C1' - C2' - C3' 96.3 102.4 0.8 3.0 DA F 2 N9 - C1' - O4' 111.3 108.3 0.3 -5.3 DC F 5 N1 - C1' - O4' 102.7 108.0 0.7 -4.9 DT F 6 C1' - C2' - C3' 97.5 102.4 0.8 -4.5 DA F 10 N9 - C1' - O4' 103.5 108.0 0.7 -7.8 DA F 11 N9 - C1' - O4' 100.2 108.0 0.7

==> The following residues have extra atoms:

RES MOD#C SEQ ATOMS

BML( 100) N1 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 O2 N2 C20 N3 C7 C8 C9

ERROR: dissociated residue(s): BML 100

Raja _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

Nigel W Moriarty

2 May 2 May

4:43 p.m.

New subject: Error in PDB Validation letter

Raja Regarding the BML compound. This code is already taken for 4-BROMOPHENOL. You can look at it using phenix.reel --chemical-components=BML You can generate a new unused code thus elbow.get_new_ligand_code and one being with B thus elbow.get_new_ligand_code B Then the PDB with not compare you compound with the one currently known as BML. Regarding the deviations. The nucleotides denoted as DA, DC, DT, ... are not using the standard codes. This may mean that you generated new restraints that are not as accurate as the standard set in the restraints library. Either way, I suggest renaming the residues to the standard codes and re-refining. This will use the standard restraints and will likely approach the values mandated by the PDB. Nigel On 4/30/10 12:37 PM, Raja Dey wrote:

...

Hi, I am getting two errors in pdb validation report as follows: I think I need to use B-DNA constraint in the refinement. This is a 2.4 A data. How can I do that in PHENIX in cammand line promt.

Another thing is, I have a small molecule in the complex, which I called as BML. I think that's might be the problem. This is known as BML210. What I should call this compound? Should I also give a chain ID for this compound? See the errors given below.

*** Covalent Angle Values:

The overall RMS deviation for covalent angles relative to the standard dictionary is 1.4 degrees.

The following table contains a list of the covalent bond angles greater than 6 times standard deviation.

Deviation Residue Chain Sequence AT1 - AT2 - AT3 Bond Dictionary Standard Name ID Number Angle Value Deviation ------------------------------------------------------------------------------------------- -4.8 DA E 1 C1' - C2' - C3' 97.6 102.4 0.8 -5.8 DA E 2 C1' - C2' - C3' 96.6 102.4 0.8 3.3 DA E 2 N9 - C1' - O4' 111.6 108.3 0.3 -5.0 DC E 5 N1 - C1' - O4' 103.0 108.0 0.7 -5.1 DT E 6 C1' - C2' - C3' 97.3 102.4 0.8 -6.7 DT E 8 C1' - O4' - C4' 103.4 110.1 1.0 2.3 DT E 8 N1 - C1' - O4' 110.6 108.3 0.3 -4.5 DT E 11 N1 - C1' - O4' 103.5 108.0 0.7 -6.1 DA F 2 C1' - C2' - C3' 96.3 102.4 0.8 3.0 DA F 2 N9 - C1' - O4' 111.3 108.3 0.3 -5.3 DC F 5 N1 - C1' - O4' 102.7 108.0 0.7 -4.9 DT F 6 C1' - C2' - C3' 97.5 102.4 0.8 -4.5 DA F 10 N9 - C1' - O4' 103.5 108.0 0.7 -7.8 DA F 11 N9 - C1' - O4' 100.2 108.0 0.7

==> The following residues have extra atoms:

RES MOD#C SEQ ATOMS

BML( 100) N1 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 O2 N2 C20 N3 C7 C8 C9

ERROR: dissociated residue(s): BML 100

Raja _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

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

Raja Dey

5:51 p.m.

New subject: Error in PDB Validation letter

Dear Nigel, Thanks for your reply. I understood the error with BML. I will try to fix that now. I am not clear what you suggested about the other error. Are you suggesting to change the nucleotides as follows: DA to ADE DC to CYT and so on... Actually I used 3 letter code for nucleotides in PHENIX refinement. Before submission to PDB I changed to DA, DC ....like this. Otherwise PDB deposition system can not recognize. Might be there is a way, so that I can avoid this hassle of changing this code for PDB deposition. Please tell me if I did anything wrong. Thanks... Raja ----- Original Message ----- From: Nigel W Moriarty Date: Sunday, May 2, 2010 9:43 am Subject: Re: [phenixbb] Error in PDB Validation letter To: PHENIX user mailing list

...

Raja

Regarding the BML compound. This code is already taken for 4- BROMOPHENOL. You can look at it using

phenix.reel --chemical-components=BML

You can generate a new unused code thus

elbow.get_new_ligand_code

and one being with B thus

elbow.get_new_ligand_code B

Then the PDB with not compare you compound with the one currently known as BML.

Regarding the deviations. The nucleotides denoted as DA, DC, DT, ... are not using the standard codes. This may mean that you generated new restraints that are not as accurate as the standard set in the restraints library. Either way, I suggest renaming the residues to the standard codes and re-refining. This will use the standard restraints and will likely approach the values mandated by the PDB.

Nigel

On 4/30/10 12:37 PM, Raja Dey wrote:

...
Hi, I am getting two errors in pdb validation report as follows: I think I need to use B-DNA constraint in the refinement. This is a 2.4 A data. How can I do that in PHENIX in cammand line promt.

Another thing is, I have a small molecule in the complex, which I called as BML. I think that's might be the problem. This is known as BML210. What I should call this compound? Should I also give a chain ID for this compound? See the errors given below.

*** Covalent Angle Values:

The overall RMS deviation for covalent angles relative to the standard dictionary is 1.4 degrees.

The following table contains a list of the covalent bond angles greater than 6 times standard deviation.

Deviation Residue Chain Sequence AT1 - AT2 - AT3 Bond Dictionary Standard Name ID Number Angle Value Deviation -------------------------------------------------------------------

...
-4.8 DA E 1 C1' - C2' - C3' 97.6 102.4 0.8 -5.8 DA E 2 C1' - C2' - C3' 96.6 102.4 0.8 3.3 DA E 2 N9 - C1' - O4' 111.6 108.3 0.3 -5.0 DC E 5 N1 - C1' - O4' 103.0 108.0 0.7 -5.1 DT E 6 C1' - C2' - C3' 97.3 102.4 0.8 -6.7 DT E 8 C1' - O4' - C4' 103.4 110.1 1.0 2.3 DT E 8 N1 - C1' - O4' 110.6 108.3 0.3 -4.5 DT E 11 N1 - C1' - O4' 103.5 108.0 0.7 -6.1 DA F 2 C1' - C2' - C3' 96.3 102.4 0.8 3.0 DA F 2 N9 - C1' - O4' 111.3 108.3 0.3 -5.3 DC F 5 N1 - C1' - O4' 102.7 108.0 0.7 -4.9 DT F 6 C1' - C2' - C3' 97.5 102.4 0.8 -4.5 DA F 10 N9 - C1' - O4' 103.5 108.0 0.7 -7.8 DA F 11 N9 - C1' - O4' 100.2 108.0 0.7

==> The following residues have extra atoms: RES MOD#C SEQ ATOMS BML( 100) N1 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 O2 N2 C20 N3 C7 C8 C9

ERROR: dissociated residue(s): BML 100

Raja _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

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

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

Nigel W Moriarty

6:50 p.m.

New subject: Error in PDB Validation letter

Raja

...

Actually I used 3 letter code for nucleotides in PHENIX refinement. Before submission to PDB I changed to DA, DC ....like this. Otherwise PDB deposition system can not recognize. Might be there is a way, so that I can avoid this hassle of changing this code for PDB deposition. Please tell me if I did anything wrong. Thanks...

This is very interesting. In my previous email, I had mixed up the codes AD, CD, GD, TD that the monomer library uses and the standard codes that PDB uses, DA, DC, ... Ralf tells me that phenix.refine recognises these latter codes so there is something else up with the refinement. You should follow Pavel's advice and check the .geo file. It certainly seems like the refinement has not converged. Nigel

...

Raja

----- Original Message ----- From: Nigel W Moriarty Date: Sunday, May 2, 2010 9:43 am Subject: Re: [phenixbb] Error in PDB Validation letter To: PHENIX user mailing list

...
Raja

Regarding the BML compound. This code is already taken for 4- BROMOPHENOL. You can look at it using

phenix.reel --chemical-components=BML

You can generate a new unused code thus

elbow.get_new_ligand_code

and one being with B thus

elbow.get_new_ligand_code B

Then the PDB with not compare you compound with the one currently known as BML.

Regarding the deviations. The nucleotides denoted as DA, DC, DT, ... are not using the standard codes. This may mean that you generated new restraints that are not as accurate as the standard set in the restraints library. Either way, I suggest renaming the residues to the standard codes and re-refining. This will use the standard restraints and will likely approach the values mandated by the PDB.

Nigel

On 4/30/10 12:37 PM, Raja Dey wrote:

...
Hi, I am getting two errors in pdb validation report as follows: I think I need to use B-DNA constraint in the refinement. This is

a 2.4 A data. How can I do that in PHENIX in cammand line promt.

...
Another thing is, I have a small molecule in the complex, which I

called as BML. I think that's might be the problem. This is known as BML210. What I should call this compound? Should I also give a chain ID for this compound? See the errors given below.

...
*** Covalent Angle Values:

The overall RMS deviation for covalent angles relative to the

standard dictionary is 1.4 degrees.

...
The following table contains a list of the covalent bond angles greater than 6 times standard deviation.

Deviation Residue Chain Sequence AT1 - AT2 - AT3 Bond

Dictionary Standard

...
Name ID Number Angle

Value Deviation

...
-------------------------------------------------------------------

------------------------

...
-4.8 DA E 1 C1' - C2' - C3' 97.6

102.4 0.8

...
-5.8 DA E 2 C1' - C2' - C3' 96.6

102.4 0.8

...
3.3 DA E 2 N9 - C1' - O4' 111.6

108.3 0.3

...
-5.0 DC E 5 N1 - C1' - O4' 103.0

108.0 0.7

...
-5.1 DT E 6 C1' - C2' - C3' 97.3

102.4 0.8

...
-6.7 DT E 8 C1' - O4' - C4' 103.4

110.1 1.0

...
2.3 DT E 8 N1 - C1' - O4' 110.6

108.3 0.3

...
-4.5 DT E 11 N1 - C1' - O4' 103.5

108.0 0.7

...
-6.1 DA F 2 C1' - C2' - C3' 96.3

102.4 0.8

...
3.0 DA F 2 N9 - C1' - O4' 111.3

108.3 0.3

...
-5.3 DC F 5 N1 - C1' - O4' 102.7

108.0 0.7

...
-4.9 DT F 6 C1' - C2' - C3' 97.5

102.4 0.8

...
-4.5 DA F 10 N9 - C1' - O4' 103.5

108.0 0.7

...
-7.8 DA F 11 N9 - C1' - O4' 100.2

108.0 0.7

...
==> The following residues have extra atoms:

RES MOD#C SEQ ATOMS

...
BML( 100) N1 C10 C11 C12 C13 C14 C15

C16 C17 C18 C19 O2 N2 C20 N3 C7 C8 C9

...
ERROR: dissociated residue(s): BML 100

Raja _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

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

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

Pavel Afonine

30 Apr 30 Apr

3:37 a.m.

New subject: Rigid body refinement question

...

What if you do like this? "strategy=rigid_body+individual_sites" "sites.individual=chain A and resid 36" "sites.rigid_body=chain A" "sites.rigid_body=chain B"

Precisely, it will do refine individual coordinates of all atoms in residue number 36 of chain A, and it will do rigid body refinement of chain A and chain B (two independent bodies). The coordinates of everything else will be fixed. Pavel.

Pavel Afonine

3:33 a.m.

New subject: Rigid body refinement question

Yes, you can: phenix.refine model.{pdb,mtz} strategy=individual_sites+rigid_body refine.sites.rigid_body="resseq 1" refine.sites.individual="not resseq 1" In the above command the residue number 1 will be refined as a rigid body while the coordinates of the rest will be refined individually. Pavel. On 4/29/10 2:17 PM, Ralf W. Grosse-Kunstleve wrote:

...

...
Is it possible to specify individual residues of a chain to undergo rigid refinement while the rest of the chain does regular refinement?

No, sorry. Ralf _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

5353

Age (days ago)

5356

Last active (days ago)

List overview

Download

10 comments

5 participants

participants (5)

Nigel W Moriarty
Pavel Afonine
Raja Dey
Ralf W. Grosse-Kunstleve
Young-Jin Cho