riding model and deposition

Alex Theodossis

18 Feb 2010 18 Feb '10

1:44 a.m.

Dear phenixbb, I have refined a 1.3A structure using individual anisotropic B-factors (Rfactor, 15.5%; Rfree, 17.7%). Additional use of the riding model during refinement has improvement the R values (Rf, 14.5%; Rfree, 16.6%), although bond length and angle outliers did increase (anyone know why that would be?). I have decided to deposit the model obtained using the riding model, but am unsure whether to include the hydrogen coordinates in the deposition. The Phenix.refine documentation advises that they are left in, as they are needed to reproduce the refinement statistics. However, reviewing the high resolution structures in the PDB I found that: 1. In the case of structures refined using REFMAC or SHELX, the riding model does not seem to be included, although its use during refinement is mentioned in the pdb header. 2. In the case of structures refined using PHENIX, use of the riding model is sometimes indicated in the header, but hydrogen coordinates only appear to be included in the pdb in the case of structures refined to a resolution better than 1.3A. As far as I can understand the purpose of the riding model is to provide additional restraints for refinement, so as long as its use is made evident in the pdb header (and maybe the methods section of the publication) that should be enough. Inclusion of hydrogens in the model, when you can't see them seems inappropriate. Nevertheless, I would appreciate a more authoritative perspective on this issue. Thank you in advance, Alex Theodossis

Show replies by date

Pavel Afonine

18 Feb 18 Feb

2:06 a.m.

Hi Alex,

...

I have refined a 1.3A structure using individual anisotropic B-factors (Rfactor, 15.5%; Rfree, 17.7%). Additional use of the riding model during refinement has improvement the R values (Rf, 14.5%; Rfree, 16.6%),

this is not surprising at all -- hydrogens do contribute to the scattering.

...

although bond length and angle outliers did increase (anyone know why that would be?).

Different model - slightly different Xray/Restraints weights - slightly different statistics.

...

I have decided to deposit the model obtained using the riding model, but am unsure whether to include the hydrogen coordinates in the deposition.

If H atoms were used in refinement and final statistics calculation (such as R-factors, etc), then the H atoms have to be included into depositing model. Otherwise your statistics will not be reproducible.

...

The Phenix.refine documentation advises that they are left in, as they are needed to reproduce the refinement statistics. However, reviewing the high resolution structures in the PDB I found that:

Reviewing statistics in the PDB you can find many interesting things, such as Fobs labeled as Iobs (and Iobs labeled as Fobs), atoms with unknown scattering types labeled X (which makes R-factor impossible to compute), O-H distances in water longer than 10A, etc. one can go on for a long list.

...

1. In the case of structures refined using REFMAC or SHELX, the riding model does not seem to be included, although its use during refinement is mentioned in the pdb header.

Which is, in my opinion, not good.

...

2. In the case of structures refined using PHENIX, use of the riding model is sometimes indicated in the header, but hydrogen coordinates only appear to be included in the pdb in the case of structures refined to a resolution better than 1.3A.

This most likely indicates a manipulation with a file contained refined model, which is, again, in my opinion, not good.

...

As far as I can understand the purpose of the riding model is to provide additional restraints for refinement, so as long as its use is made evident in the pdb header (and maybe the methods section of the publication) that should be enough. Inclusion of hydrogens in the model, when you can't see them seems inappropriate.

Christina Bourne

2:55 p.m.

Alex, Pavel, In my experience when using a riding hydrogen model they are not included in the output pdb file. It is only when they are added explicitly that they are given output coordinates, which should of course be kept unchanged. Can you please verify this?? -Christina ________________________________ From: Pavel Afonine To: PHENIX user mailing list Sent: Wed, February 17, 2010 8:06:40 PM Subject: Re: [phenixbb] riding model and deposition Hi Alex,

...

I have refined a 1.3A structure using individual anisotropic B-factors (Rfactor, 15.5%; Rfree, 17.7%). Additional use of the riding model during refinement has improvement the R values (Rf, 14.5%; Rfree, 16.6%),

this is not surprising at all -- hydrogens do contribute to the scattering.

...

although bond length and angle outliers did increase (anyone know why that would be?).

Different model - slightly different Xray/Restraints weights - slightly different statistics.

...

I have decided to deposit the model obtained using the riding model, but am unsure whether to include the hydrogen coordinates in the deposition.

...

The Phenix.refine documentation advises that they are left in, as they are needed to reproduce the refinement statistics. However, reviewing the high resolution structures in the PDB I found that:

...

1. In the case of structures refined using REFMAC or SHELX, the riding model does not seem to be included, although its use during refinement is mentioned in the pdb header.

Which is, in my opinion, not good.

...

2. In the case of structures refined using PHENIX, use of the riding model is sometimes indicated in the header, but hydrogen coordinates only appear to be included in the pdb in the case of structures refined to a resolution better than 1.3A.

This most likely indicates a manipulation with a file contained refined model, which is, again, in my opinion, not good.

...

As far as I can understand the purpose of the riding model is to provide additional restraints for refinement, so as long as its use is made evident in the pdb header (and maybe the methods section of the publication) that should be enough. Inclusion of hydrogens in the model, when you can't see them seems inappropriate.

Different programs use different libraries of ideal X-H parameters. Depending on the program, the B-factors of H atoms can be inherited from atoms they are bonded to or can be multiplied by a scale from 1 to 1.5. These all makes adding H program-dependent, and therefore statistics from such models may vary. I believe that any manipulation on the refined structure is bad. Pavel. _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb

Pavel Afonine

5:36 p.m.

Hi Christina, if H atoms were used in refinement then they should not be removed from the model you are going to PDB deposit. Post-refinement model manipulation is bad. This makes the reported statistics irreproducible. The R-factor computed for a model with and without H atoms is different. Different programs use different libraries of ideal X-H parameters (some put them at longer "neutron" X-H distances, and some place at the position where the electron cloud is expected - shorter X-H distance). The B-factors of H atoms can be inherited from atoms they are bonded to or can be multiplied by a scale from 1 to 1.5. There is a class of X-H geometries where this position of H atoms is not unique: for example, C-O-H group in Tyrosine residue, where H can rotate around C-O axis. Some programs optimize the position of H atom to accommodate potential hydrogen bond and some doesn't. Destroying is always easier than building: removing existing hydrogens is a matter of running one command : phenix.reduce -trim model.pdb > model_noH.pdb and takes 1 or 2 seconds. However, to build the H atoms back the *exact* same way they were before is not that straightforward. It's just counterintuitive why you need to manipulate the refined model for which you have obtained all the statistics numbers? If you remove hydrogen atoms then you need to update the statistics. By stripping off H atoms you will get somewhat smaller file, but does the file size matter these days? Pavel. On 2/18/10 6:55 AM, Christina Bourne wrote:

...

Alex, Pavel, In my experience when using a riding hydrogen model they are not included in the output pdb file. It is only when they are added explicitly that they are given output coordinates, which should of course be kept unchanged. Can you please verify this?? -Christina

------------------------------------------------------------------------ *From:* Pavel Afonine *To:* PHENIX user mailing list *Sent:* Wed, February 17, 2010 8:06:40 PM *Subject:* Re: [phenixbb] riding model and deposition

Hi Alex,

...
I have refined a 1.3A structure using individual anisotropic B-factors (Rfactor, 15.5%; Rfree, 17.7%). Additional use of the riding model during refinement has improvement the R values (Rf, 14.5%; Rfree, 16.6%),

this is not surprising at all -- hydrogens do contribute to the scattering.

...
although bond length and angle outliers did increase (anyone know why that would be?).

Different model - slightly different Xray/Restraints weights - slightly different statistics.

...
I have decided to deposit the model obtained using the riding model, but am unsure whether to include the hydrogen coordinates in the deposition.

If H atoms were used in refinement and final statistics calculation (such as R-factors, etc), then the H atoms have to be included into depositing model. Otherwise your statistics will not be reproducible.

...
The Phenix.refine documentation advises that they are left in, as they are needed to reproduce the refinement statistics. However, reviewing the high resolution structures in the PDB I found that:

Reviewing statistics in the PDB you can find many interesting things, such as Fobs labeled as Iobs (and Iobs labeled as Fobs), atoms with unknown scattering types labeled X (which makes R-factor impossible to compute), O-H distances in water longer than 10A, etc. one can go on for a long list.

...
1. In the case of structures refined using REFMAC or SHELX, the riding model does not seem to be included, although its use during refinement is mentioned in the pdb header.

Which is, in my opinion, not good.

...
2. In the case of structures refined using PHENIX, use of the riding model is sometimes indicated in the header, but hydrogen coordinates only appear to be included in the pdb in the case of structures refined to a resolution better than 1.3A.

This most likely indicates a manipulation with a file contained refined model, which is, again, in my opinion, not good.

...
As far as I can understand the purpose of the riding model is to provide additional restraints for refinement, so as long as its use is made evident in the pdb header (and maybe the methods section of the publication) that should be enough. Inclusion of hydrogens in the model, when you can't see them seems inappropriate.

Different programs use different libraries of ideal X-H parameters. Depending on the program, the B-factors of H atoms can be inherited from atoms they are bonded to or can be multiplied by a scale from 1 to 1.5. These all makes adding H program-dependent, and therefore statistics from such models may vary.

I believe that any manipulation on the refined structure is bad.

Pavel.

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

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

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

Schubert, Carsten [PRDUS]

7 p.m.

Hi Pavel, if I understand Christina's question correctly, she was stating that when phenix.refine runs with riding hydrogens enabled using a pdb file, which does not contain hydrogens, the final hydrogens are not written out. I presume this is the case, since a phenix.reduce run is required to add hydrogens prior to refinement, phenix.refine does not add hydrogens, correct? Cheers, Carsten From: [email protected] [mailto:[email protected]] On Behalf Of Pavel Afonine Sent: Thursday, February 18, 2010 12:36 PM To: PHENIX user mailing list Subject: Re: [phenixbb] riding model and deposition Hi Christina, if H atoms were used in refinement then they should not be removed from the model you are going to PDB deposit. Post-refinement model manipulation is bad. This makes the reported statistics irreproducible. The R-factor computed for a model with and without H atoms is different. Different programs use different libraries of ideal X-H parameters (some put them at longer "neutron" X-H distances, and some place at the position where the electron cloud is expected - shorter X-H distance). The B-factors of H atoms can be inherited from atoms they are bonded to or can be multiplied by a scale from 1 to 1.5. There is a class of X-H geometries where this position of H atoms is not unique: for example, C-O-H group in Tyrosine residue, where H can rotate around C-O axis. Some programs optimize the position of H atom to accommodate potential hydrogen bond and some doesn't. Destroying is always easier than building: removing existing hydrogens is a matter of running one command : phenix.reduce -trim model.pdb > model_noH.pdb and takes 1 or 2 seconds. However, to build the H atoms back the *exact* same way they were before is not that straightforward. It's just counterintuitive why you need to manipulate the refined model for which you have obtained all the statistics numbers? If you remove hydrogen atoms then you need to update the statistics. By stripping off H atoms you will get somewhat smaller file, but does the file size matter these days? Pavel. On 2/18/10 6:55 AM, Christina Bourne wrote: Alex, Pavel, In my experience when using a riding hydrogen model they are not included in the output pdb file. It is only when they are added explicitly that they are given output coordinates, which should of course be kept unchanged. Can you please verify this?? -Christina ________________________________ From: Pavel Afonine mailto:[email protected] To: PHENIX user mailing list mailto:[email protected] Sent: Wed, February 17, 2010 8:06:40 PM Subject: Re: [phenixbb] riding model and deposition Hi Alex,

...

I have refined a 1.3A structure using individual anisotropic B-factors (Rfactor, 15.5%; Rfree, 17.7%). Additional use of the riding model during refinement has improvement the R values (Rf, 14.5%; Rfree, 16.6%),

...

although bond length and angle outliers did increase (anyone know why

this is not surprising at all -- hydrogens do contribute to the scattering. that would be?).

...

Different model - slightly different Xray/Restraints weights - slightly different statistics.

...

I have decided to deposit the model obtained using the riding model, but am unsure whether to include the hydrogen coordinates in the deposition.

...

The Phenix.refine documentation advises that they are left in, as they are needed to reproduce the refinement statistics. However, reviewing

...

1. In the case of structures refined using REFMAC or SHELX, the riding model does not seem to be included, although its use during refinement is mentioned in the pdb header.

Which is, in my opinion, not good.

...

2. In the case of structures refined using PHENIX, use of the riding model is sometimes indicated in the header, but hydrogen coordinates only appear to be included in the pdb in the case of structures refined to a resolution better than 1.3A.

...

As far as I can understand the purpose of the riding model is to

This most likely indicates a manipulation with a file contained refined model, which is, again, in my opinion, not good. provide additional restraints for refinement, so as long as its use is made evident in the pdb header (and maybe the methods section of the publication) that should be enough. Inclusion of hydrogens in the model, when you can't see them seems inappropriate.

...

Pavel Afonine

7:05 p.m.

Hi Carsten, phenix.refine does not add hydrogens. If you want to use H atoms in refinement then you need to add them first, before going into refinement. You can add hydrogens using phenix.reduce or phenix.ready_set. Pavel. On 2/18/10 11:00 AM, Schubert, Carsten [PRDUS] wrote:

...

Hi Pavel,

if I understand Christina's question correctly, she was stating that when phenix.refine runs with riding hydrogens enabled using a pdb file, which does not contain hydrogens, the final hydrogens are not written out. I presume this is the case, since a phenix.reduce run is required to add hydrogens prior to refinement, phenix.refine does not add hydrogens, correct?

Cheers,

Carsten

*From:* [email protected] [mailto:[email protected]] *On Behalf Of *Pavel Afonine *Sent:* Thursday, February 18, 2010 12:36 PM *To:* PHENIX user mailing list *Subject:* Re: [phenixbb] riding model and deposition

Hi Christina,

if H atoms were used in refinement then they should not be removed from the model you are going to PDB deposit. Post-refinement model manipulation is bad. This makes the reported statistics irreproducible. The R-factor computed for a model with and without H atoms is different.

Different programs use different libraries of ideal X-H parameters (some put them at longer "neutron" X-H distances, and some place at the position where the electron cloud is expected - shorter X-H distance). The B-factors of H atoms can be inherited from atoms they are bonded to or can be multiplied by a scale from 1 to 1.5. There is a class of X-H geometries where this position of H atoms is not unique: for example, C-O-H group in Tyrosine residue, where H can rotate around C-O axis. Some programs optimize the position of H atom to accommodate potential hydrogen bond and some doesn't.

Destroying is always easier than building: removing existing hydrogens is a matter of running one command :

phenix.reduce -trim model.pdb > model_noH.pdb

and takes 1 or 2 seconds. However, to build the H atoms back the *exact* same way they were before is not that straightforward.

It's just counterintuitive why you need to manipulate the refined model for which you have obtained all the statistics numbers? If you remove hydrogen atoms then you need to update the statistics. By stripping off H atoms you will get somewhat smaller file, but does the file size matter these days?

Pavel.

On 2/18/10 6:55 AM, Christina Bourne wrote:

Alex, Pavel, In my experience when using a riding hydrogen model they are not included in the output pdb file. It is only when they are added explicitly that they are given output coordinates, which should of course be kept unchanged. Can you please verify this?? -Christina

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

*From:* Pavel Afonine mailto:[email protected] *To:* PHENIX user mailing list mailto:[email protected] *Sent:* Wed, February 17, 2010 8:06:40 PM *Subject:* Re: [phenixbb] riding model and deposition

Hi Alex,

...
I have refined a 1.3A structure using individual anisotropic B-factors (Rfactor, 15.5%; Rfree, 17.7%). Additional use of the riding model during refinement has improvement the R values (Rf, 14.5%; Rfree, 16.6%),

this is not surprising at all -- hydrogens do contribute to the scattering.

...
although bond length and angle outliers did increase (anyone know why that would be?).

Different model - slightly different Xray/Restraints weights - slightly different statistics.

...
I have decided to deposit the model obtained using the riding model, but am unsure whether to include the hydrogen coordinates in the deposition.

If H atoms were used in refinement and final statistics calculation (such as R-factors, etc), then the H atoms have to be included into depositing model. Otherwise your statistics will not be reproducible.

...
The Phenix.refine documentation advises that they are left in, as they are needed to reproduce the refinement statistics. However, reviewing the high resolution structures in the PDB I found that:

Reviewing statistics in the PDB you can find many interesting things, such as Fobs labeled as Iobs (and Iobs labeled as Fobs), atoms with unknown scattering types labeled X (which makes R-factor impossible to compute), O-H distances in water longer than 10A, etc. one can go on for a long list.

...
1. In the case of structures refined using REFMAC or SHELX, the riding model does not seem to be included, although its use during refinement is mentioned in the pdb header.

Which is, in my opinion, not good.

...
2. In the case of structures refined using PHENIX, use of the riding model is sometimes indicated in the header, but hydrogen coordinates only appear to be included in the pdb in the case of structures refined to a resolution better than 1.3A.

This most likely indicates a manipulation with a file contained refined model, which is, again, in my opinion, not good.

...
As far as I can understand the purpose of the riding model is to provide additional restraints for refinement, so as long as its use is made evident in the pdb header (and maybe the methods section of the publication) that should be enough. Inclusion of hydrogens in the model, when you can't see them seems inappropriate.

Different programs use different libraries of ideal X-H parameters. Depending on the program, the B-factors of H atoms can be inherited from atoms they are bonded to or can be multiplied by a scale from 1 to 1.5. These all makes adding H program-dependent, and therefore statistics from such models may vary.

I believe that any manipulation on the refined structure is bad.

Pavel.

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

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

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

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

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

Pavel Afonine

7:07 p.m.

PS: I was talking about command line version of phenix.refine. If you use the GUI then there is an option to add H. Pavel. On 2/18/10 11:05 AM, Pavel Afonine wrote:

...

Hi Carsten,

phenix.refine does not add hydrogens. If you want to use H atoms in refinement then you need to add them first, before going into refinement.

You can add hydrogens using phenix.reduce or phenix.ready_set.

Pavel.

On 2/18/10 11:00 AM, Schubert, Carsten [PRDUS] wrote:

...
Hi Pavel,

if I understand Christina's question correctly, she was stating that when phenix.refine runs with riding hydrogens enabled using a pdb file, which does not contain hydrogens, the final hydrogens are not written out. I presume this is the case, since a phenix.reduce run is required to add hydrogens prior to refinement, phenix.refine does not add hydrogens, correct?

Cheers,

Carsten

*From:* [email protected] [mailto:[email protected]] *On Behalf Of *Pavel Afonine *Sent:* Thursday, February 18, 2010 12:36 PM *To:* PHENIX user mailing list *Subject:* Re: [phenixbb] riding model and deposition

Hi Christina,

if H atoms were used in refinement then they should not be removed from the model you are going to PDB deposit. Post-refinement model manipulation is bad. This makes the reported statistics irreproducible. The R-factor computed for a model with and without H atoms is different.

Different programs use different libraries of ideal X-H parameters (some put them at longer "neutron" X-H distances, and some place at the position where the electron cloud is expected - shorter X-H distance). The B-factors of H atoms can be inherited from atoms they are bonded to or can be multiplied by a scale from 1 to 1.5. There is a class of X-H geometries where this position of H atoms is not unique: for example, C-O-H group in Tyrosine residue, where H can rotate around C-O axis. Some programs optimize the position of H atom to accommodate potential hydrogen bond and some doesn't.

Destroying is always easier than building: removing existing hydrogens is a matter of running one command :

phenix.reduce -trim model.pdb > model_noH.pdb

and takes 1 or 2 seconds. However, to build the H atoms back the *exact* same way they were before is not that straightforward.

It's just counterintuitive why you need to manipulate the refined model for which you have obtained all the statistics numbers? If you remove hydrogen atoms then you need to update the statistics. By stripping off H atoms you will get somewhat smaller file, but does the file size matter these days?

Pavel.

On 2/18/10 6:55 AM, Christina Bourne wrote:

Alex, Pavel, In my experience when using a riding hydrogen model they are not included in the output pdb file. It is only when they are added explicitly that they are given output coordinates, which should of course be kept unchanged. Can you please verify this?? -Christina

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

*From:* Pavel Afonine mailto:[email protected] *To:* PHENIX user mailing list mailto:[email protected] *Sent:* Wed, February 17, 2010 8:06:40 PM *Subject:* Re: [phenixbb] riding model and deposition

Hi Alex,

...
I have refined a 1.3A structure using individual anisotropic B-factors (Rfactor, 15.5%; Rfree, 17.7%). Additional use of the riding model during refinement has improvement the R values (Rf, 14.5%; Rfree, 16.6%),

this is not surprising at all -- hydrogens do contribute to the scattering.

...
although bond length and angle outliers did increase (anyone know why that would be?).

Different model - slightly different Xray/Restraints weights - slightly different statistics.

...
I have decided to deposit the model obtained using the riding model, but am unsure whether to include the hydrogen coordinates in the deposition.

If H atoms were used in refinement and final statistics calculation (such as R-factors, etc), then the H atoms have to be included into depositing model. Otherwise your statistics will not be reproducible.

...
The Phenix.refine documentation advises that they are left in, as they are needed to reproduce the refinement statistics. However, reviewing the high resolution structures in the PDB I found that:

Reviewing statistics in the PDB you can find many interesting things, such as Fobs labeled as Iobs (and Iobs labeled as Fobs), atoms with unknown scattering types labeled X (which makes R-factor impossible to compute), O-H distances in water longer than 10A, etc. one can go on for a long list.

...
1. In the case of structures refined using REFMAC or SHELX, the riding model does not seem to be included, although its use during refinement is mentioned in the pdb header.

Which is, in my opinion, not good.

...
2. In the case of structures refined using PHENIX, use of the riding model is sometimes indicated in the header, but hydrogen coordinates only appear to be included in the pdb in the case of structures refined to a resolution better than 1.3A.

This most likely indicates a manipulation with a file contained refined model, which is, again, in my opinion, not good.

...
As far as I can understand the purpose of the riding model is to provide additional restraints for refinement, so as long as its use is made evident in the pdb header (and maybe the methods section of the publication) that should be enough. Inclusion of hydrogens in the model, when you can't see them seems inappropriate.

Different programs use different libraries of ideal X-H parameters. Depending on the program, the B-factors of H atoms can be inherited from atoms they are bonded to or can be multiplied by a scale from 1 to 1.5. These all makes adding H program-dependent, and therefore statistics from such models may vary.

I believe that any manipulation on the refined structure is bad.

Pavel.

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

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

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

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

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

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

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

Nathaniel Echols

7:32 p.m.

On Feb 18, 2010, at 11:07 AM, Pavel Afonine wrote:

...

PS: I was talking about command line version of phenix.refine. If you use the GUI then there is an option to add H.

Technically, the GUI just runs ready_set, then passes that file to phenix.refine. The only program that generates and uses "implicit hydrogens" is Refmac, and even in that case the internal model has explicit hydrogens, they just aren't printed out at the end. I considered adding an option to the GUI to strip out hydrogens at the end, but Pavel has convinced me that this is a bad idea - it is generally okay to do this during refinement (where different programs have very different attitudes towards hydrogens), but not for deposition. Nat ------------------- Nathaniel Echols Lawrence Berkeley Lab 510-486-5136 [email protected]

5262

Age (days ago)

5262

Last active (days ago)

List overview

Download

7 comments

5 participants

participants (5)

Alex Theodossis
Christina Bourne
Nathaniel Echols
Pavel Afonine
Schubert, Carsten [PRDUS]