Hello, all it needs to resolve this particular problem is not use an ancient Phenix version (that has that feature). Edna's problem was solved after switching from 1.8_1069 (which I consider ancient) to a more recent (in fact latest) version. Pavel On 11/18/13 11:45 AM, Diana Tomchick wrote:

I have had this problem in the past for refinement of some low resolution datasets. The problem resolved itself when I elected to optimize both the X-ray/stereochemistry and the X-ray/ADP weights. It adds some time to the refinement cycles, but I always have lots of other things to do while I'm waiting for the job to finish.

If you are using the Phenix GUI, then look under the refinement results "Atomic properties" tab which reports ADP statistics. This should provide you a listing of atoms with problematic ADP values.

Diana

* * * * * * * * * * * * * * * * * * * * * * * * * * * * Diana R. Tomchick Professor University of Texas Southwestern Medical Center Department of Biophysics 5323 Harry Hines Blvd. Rm. ND10.214A Dallas, TX 75390-8816, U.S.A. Email: [email protected] 214-645-6383 (phone) 214-645-6353 (fax)

On Nov 15, 2013, at 11:37 PM, Edna Patricia Matta, Ms wrote:

...

Hello,

Recently I did submit a structure to the PDB but they found that some atoms had zero value for the B-factors. I am new in this filed and I do not know what to do. There have been reports of problems with some versions of phenix about this. I used Phenix 1.8.

I repeted the refinement doing restrain NCS-related B-factors but I do not find where to check for the the B-factor values in the the log file.

Somebody can help me? Thanks a lot, Edna

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

________________________________

UT Southwestern Medical Center The future of medicine, today.

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

If you see this problem using the latest Phenix version then please send me files (off-list) necessary to reproduce it and I will investigate. If interested what was the "problem" causing zero B-factors please see below. Thanks, Pavel In a nutshell the problem is this. Assuming total model structure factor is (for simplicity): F ~ exp(-Boverall_scale * s**2) * exp(-Batoms * s**2) it's clear that any combination of Boverall_scale and Batoms that maintains Boverall_scale+Batoms=const will not change total model structure factors (and therefore R-factors, maps, etc). The new bulk-solvent and overall scaling algorithm that we implemented sometime in April 2012 and that went into 1.8_1069 originally did not care about returning isotropic component of Boverall_scale back to atoms (Batoms): http://journals.iucr.org/d/issues/2013/04/00/dz5273/dz5273.pdf In fact, it assumed that overall contribution (overall B that is equal for all atoms) should stay in Boverall_scale (indeed, why keep common B in all individual atomic B-factors!?), and the rest goes into Batoms. This behavior typically generates user concerns about "too small" B-factors. An alternative agreement is to postulate that the matrix Boverall_scale is traceless meaning that overall B-factor goes into individual atomic B-factors. This behavior occasionally generates user concerns about "too large" B-factors. A few months later (sometime in end summer/fall 2012) we changed scaling behavior such that overall B-factor stays in individual atomic B-factors. All in all, either way is correct and a matter of agreement and preferences, but definitely not a bug.

Hi Morten, POLYGON is the tool for this. Definition "too low/high" is relative and depends on resolution. Pavel On 11/25/13 4:29 AM, Morten Grøftehauge wrote:

It would be nice if the validation procedures looked for B-factor outliers that were too low as well as too high. A very low B-factor can be indicative of the wrong atom, e.g. a water molecule that is actually an ion.

Morten

On 18 November 2013 21:38, Pavel Afonine mailto:[email protected]> wrote:

If you see this problem using the latest Phenix version then please send me files (off-list) necessary to reproduce it and I will investigate.

If interested what was the "problem" causing zero B-factors please see below.

Thanks, Pavel

In a nutshell the problem is this. Assuming total model structure factor is (for simplicity):

F ~ exp(-Boverall_scale * s**2) * exp(-Batoms * s**2)

it's clear that any combination of Boverall_scale and Batoms that maintains Boverall_scale+Batoms=const will not change total model structure factors (and therefore R-factors, maps, etc).

The new bulk-solvent and overall scaling algorithm that we implemented sometime in April 2012 and that went into 1.8_1069 originally did not care about returning isotropic component of Boverall_scale back to atoms (Batoms):

http://journals.iucr.org/d/__issues/2013/04/00/dz5273/__dz5273.pdf http://journals.iucr.org/d/issues/2013/04/00/dz5273/dz5273.pdf

In fact, it assumed that overall contribution (overall B that is equal for all atoms) should stay in Boverall_scale (indeed, why keep common B in all individual atomic B-factors!?), and the rest goes into Batoms. This behavior typically generates user concerns about "too small" B-factors.

An alternative agreement is to postulate that the matrix Boverall_scale is traceless meaning that overall B-factor goes into individual atomic B-factors. This behavior occasionally generates user concerns about "too large" B-factors.

A few months later (sometime in end summer/fall 2012) we changed scaling behavior such that overall B-factor stays in individual atomic B-factors.

All in all, either way is correct and a matter of agreement and preferences, but definitely not a bug.

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

-- Morten K Grøftehauge, PhD Pohl Group Durham University

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

Engin, you can see it as a histogram. Also phenix.refine log shows histograms of B-factors that will capture this. Most of the time the problem is not about not having the information but not knowing what to look for or not willing to look. Then eventually overlooked problems get discovered later down the road by someone else. Pavel On 11/25/13 7:59 AM, Engin Özkan wrote:

I think what was meant is Phenix report not only *residues* with "suspiciously high B-factors" as it does in the GUI (thanks!), but also those with suspiciously low. As said, one could then detect that one water that was actually a big ion. The overall comparisons reported by POLYGON are useful for determining refinement parameterization and such, but not details like this.

Engin

On 11/25/13, 9:40 AM, Pavel Afonine wrote:

...

Hi Morten,

POLYGON is the tool for this. Definition "too low/high" is relative and depends on resolution.

Pavel

On 11/25/13 4:29 AM, Morten Grøftehauge wrote:

...

It would be nice if the validation procedures looked for B-factor outliers that were too low as well as too high. A very low B-factor can be indicative of the wrong atom, e.g. a water molecule that is actually an ion.

Morten

On 18 November 2013 21:38, Pavel Afonine mailto:[email protected]> wrote:

If you see this problem using the latest Phenix version then please send me files (off-list) necessary to reproduce it and I will investigate.

If interested what was the "problem" causing zero B-factors please see below.

Thanks, Pavel

In a nutshell the problem is this. Assuming total model structure factor is (for simplicity):

F ~ exp(-Boverall_scale * s**2) * exp(-Batoms * s**2)

it's clear that any combination of Boverall_scale and Batoms that maintains Boverall_scale+Batoms=const will not change total model structure factors (and therefore R-factors, maps, etc).

The new bulk-solvent and overall scaling algorithm that we implemented sometime in April 2012 and that went into 1.8_1069 originally did not care about returning isotropic component of Boverall_scale back to atoms (Batoms):

http://journals.iucr.org/d/__issues/2013/04/00/dz5273/__dz5273.pdf http://journals.iucr.org/d/issues/2013/04/00/dz5273/dz5273.pdf

In fact, it assumed that overall contribution (overall B that is equal for all atoms) should stay in Boverall_scale (indeed, why keep common B in all individual atomic B-factors!?), and the rest goes into Batoms. This behavior typically generates user concerns about "too small" B-factors.

An alternative agreement is to postulate that the matrix Boverall_scale is traceless meaning that overall B-factor goes into individual atomic B-factors. This behavior occasionally generates user concerns about "too large" B-factors.

A few months later (sometime in end summer/fall 2012) we changed scaling behavior such that overall B-factor stays in individual atomic B-factors.

All in all, either way is correct and a matter of agreement and preferences, but definitely not a bug.

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

-- Morten K Grøftehauge, PhD Pohl Group Durham University

_______________________________________________ phenixbb mailing list [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

On Mon, Nov 25, 2013 at 8:21 AM, Engin Özkan wrote:

I was looking at this histogram when I wrote that email :) but that histogram is not clickable and does not report residue names (unlike the list just below it). I always look at the histogram (and its usually Poisson-like shape is intriguing). It was just a suggestion to add the other end of the distribution to the list of outlier residues reported by Phenix.

Good point - I will add this. The next version of the validation in Phenix (phenix.molprobity, not yet available in the GUI) will already flag waters with suspiciously low B-factors, although the threshold for suspicion is arbitrarily defined right now. -Nat

Yes, your are right, it would be good to have it.. Pavel On 11/25/13 8:21 AM, Engin Özkan wrote:

I was looking at this histogram when I wrote that email :) but that histogram is not clickable and does not report residue names (unlike the list just below it). I always look at the histogram (and its usually Poisson-like shape is intriguing). It was just a suggestion to add the other end of the distribution to the list of outlier residues reported by Phenix.

Engin

On 11/25/13, 10:12 AM, Pavel Afonine wrote:

...

Engin,

you can see it as a histogram. Also phenix.refine log shows histograms of B-factors that will capture this.

Most of the time the problem is not about not having the information but not knowing what to look for or not willing to look. Then eventually overlooked problems get discovered later down the road by someone else.

Pavel

On 11/25/13 7:59 AM, Engin Özkan wrote:

...

I think what was meant is Phenix report not only *residues* with "suspiciously high B-factors" as it does in the GUI (thanks!), but also those with suspiciously low. As said, one could then detect that one water that was actually a big ion. The overall comparisons reported by POLYGON are useful for determining refinement parameterization and such, but not details like this.

Engin

On 11/25/13, 9:40 AM, Pavel Afonine wrote:

...

Hi Morten,

POLYGON is the tool for this. Definition "too low/high" is relative and depends on resolution.

Pavel

On 11/25/13 4:29 AM, Morten Grøftehauge wrote:

...

It would be nice if the validation procedures looked for B-factor outliers that were too low as well as too high. A very low B-factor can be indicative of the wrong atom, e.g. a water molecule that is actually an ion.

Morten

On 18 November 2013 21:38, Pavel Afonine mailto:[email protected]> wrote:

If you see this problem using the latest Phenix version then please send me files (off-list) necessary to reproduce it and I will investigate.

If interested what was the "problem" causing zero B-factors please see below.

Thanks, Pavel

In a nutshell the problem is this. Assuming total model structure factor is (for simplicity):

F ~ exp(-Boverall_scale * s**2) * exp(-Batoms * s**2)

it's clear that any combination of Boverall_scale and Batoms that maintains Boverall_scale+Batoms=const will not change total model structure factors (and therefore R-factors, maps, etc).

The new bulk-solvent and overall scaling algorithm that we implemented sometime in April 2012 and that went into 1.8_1069 originally did not care about returning isotropic component of Boverall_scale back to atoms (Batoms):

http://journals.iucr.org/d/__issues/2013/04/00/dz5273/__dz5273.pdf http://journals.iucr.org/d/issues/2013/04/00/dz5273/dz5273.pdf

In fact, it assumed that overall contribution (overall B that is equal for all atoms) should stay in Boverall_scale (indeed, why keep common B in all individual atomic B-factors!?), and the rest goes into Batoms. This behavior typically generates user concerns about "too small" B-factors.

An alternative agreement is to postulate that the matrix Boverall_scale is traceless meaning that overall B-factor goes into individual atomic B-factors. This behavior occasionally generates user concerns about "too large" B-factors.

A few months later (sometime in end summer/fall 2012) we changed scaling behavior such that overall B-factor stays in individual atomic B-factors.

All in all, either way is correct and a matter of agreement and preferences, but definitely not a bug.

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

-- Morten K Grøftehauge, PhD Pohl Group Durham University

_______________________________________________ phenixbb mailing list [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

---

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

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

Thanks a lot for all the answers :) Pavel was very kind to re-run my data with the latest version of the software and the problem was solved!! A great week to all, Edna Edna Patricia Matta, MSc. PhD Postdoctoral Fellow McGill University Goodman Cancer Centre 1160 Pine Avenue West, Room 614 Montreal, Quebec, CANADA, H3A 1A3 [http://www.med.mcgill.ca/nahum/images/tel.png] TEL: 1-514-398-7275 [http://www.med.mcgill.ca/nahum/images/fax.png] FAX: 1-514-398-1287 ________________________________ From: [email protected] [[email protected]] on behalf of Anshul Bhardwaj [[email protected]] Sent: Monday, November 18, 2013 3:02 PM To: PHENIX user mailing list Subject: Re: [phenixbb] zero B-factors Hello Edna, I am in the same boat, I had similar B-factor=0 problem with phenix.refine ver. 1.8_1069. Apparently, it was a bug in that specific version of PHENIX, which was fixed subsequently in later versions. Almost more than a year after PDB deposition to RCSB database, RCSB annotators found the issue and requested to send revised PDB with properly refined B-factors. The simplest solution is to refine the structure with latest official stable PHENIX version 1.8.4-1496, that takes care for the problem. Hope this helps! Best, Anshul Anshul Bhardwaj, Ph.D. Faculty, Department of Biochemistry and Molecular Biology Manager, Kimmel Cancer Center X-ray Crystallography and Molecular Interactions Thomas Jefferson University 233 South 10th Street, BLSB suite 822 Philadelphia, PA 19107 Tel: (215) 503-4587 [email protected] http://www.jefferson.edu/jmc/departments/biochemistry/faculty-staff/faculty/... http://www.kimmelcancercenter.org/kcc/kccnew/research/resources/xray/index.h... ________________________________ From: [email protected] on behalf of Edna Patricia Matta, Ms Sent: Saturday, November 16, 2013 12:37 AM To: [email protected] Subject: [phenixbb] zero B-factors Hello, Recently I did submit a structure to the PDB but they found that some atoms had zero value for the B-factors. I am new in this filed and I do not know what to do. There have been reports of problems with some versions of phenix about this. I used Phenix 1.8. I repeted the refinement doing restrain NCS-related B-factors but I do not find where to check for the the B-factor values in the the log file. Somebody can help me? Thanks a lot, Edna The information contained in this transmission contains privileged and confidential information. It is intended only for the use of the person named above. If you are not the intended recipient, you are hereby notified that any review, dissemination, distribution or duplication of this communication is strictly prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message. CAUTION: Intended recipients should NOT use email communication for emergent or urgent health care matters.