I attached the summary letter. Thanks for your time. Regards... Raja ----- Original Message ----- From: "Ralf W. Grosse-Kunstleve" Date: Tuesday, November 17, 2009 11:48 am Subject: Re: [phenixbb] fixing geometry for deposition To: [email protected]

...

pdb validation summary letter gives a list of covalent bond angles greater than 6 times standard deviation for dna part of my molecule. I am wondering if someone can tell how to fix that. I used PHENIX to refine my molecule. I did not use any restraints for the dna part.

Do you mean hydrogen-bond restraints? -- At 2.9A the structure would probably be completely distorted if you didn't have any restraints. The 6 times standard deviation should not happen if we use the same restraints as the pdb is expecting. If you send me (off-list) the inputs and the pdb summary letter we'll take a closer look.

...

Also a few amino acids are listed for which phy and psy fall outside Ramachandran plot. I like to know what is the best way to fix these phy, psy angles.

A few outliers are expected. Did you look at them in coot?

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

Hi Pavel, I used the following commands at the last stage of refinement: phenix.refine 1205A-p21212.mtz nov138_001.pdb simulated_annealing=false ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None main.bulk_solvent_and_scale=false strategy=rigid_body+individual_sites xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=1 output.write_maps=true output.prefix=nov139 --overwrite Regarding amino acids phy, psy..... I tried real space refinement/regularization in coot, but getting the same list of amino acids again and again in the pdb validation summary letter. I also tried mutation on its own to get the ideal geometry, still got the same result. Did I do anything wrong? Raja ----- Original Message ----- From: Pavel Afonine Date: Tuesday, November 17, 2009 11:50 am Subject: Re: [phenixbb] fixing geometry for deposition To: PHENIX user mailing list

Hi Raja,

...

pdb validation summary letter gives a list

of covalent bond angles greater than 6 times standard deviation for dna part of my molecule. I am wondering if someone can tell how to fix that. I used PHENIX to refine my molecule. I did not use any restraints for the dna part.

...

did you completely turn off all the geometry restraints for that selected part of your structure? If so, then it is not surprising that it got distorted during refinement at 2.9A resolution, and therefore I would suggest re-running phenix.refine using all the restraints that phenix.refine normally uses by default.

...

Also a few amino acids are listed for which

phy and psy fall outside Ramachandran plot. I like to know what is the best way to fix these phy, psy angles.

...

1) Can't you do it in Coot? (especially when there are only a few such outliers, as you say).

2) Try adding riding H atoms and re-run phenix.refine (with and w/o weights optimization to see which option gives better results).

3) Try a quick geometry regularization going into trying "2)". Repeat the same with and w/o H.

This is a pretty frequent question, so I hope someone someone can tell a success story about it...

...

Final refined parameters are

REMARK Start: r_work = 0.3576 r_free = 0.3345 bonds = 0.009 angles = 1.344 REMARK Final: r_work = 0.2517 r_free = 0.2845 bonds = 0.008 angles = 1.209

at 2.9A resolution

Looks reasonable given the resolution...

Pavel.

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

Hi Pavel, I finished 4 runs as you suggested. See below: run1. phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 output.write_maps=true output.prefix=nov17 REMARK Start: r_work = 0.2803 r_free = 0.3036 bonds = 0.012 angles = 1.488 REMARK Final: r_work = 0.2438 r_free = 0.2994 bonds = 0.009 angles = 1.309 b_ave=70.09 run2. with wxc and wxu optimization phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 optimize_wxc=true optimize_wxu=true output.write_maps=true output.prefix=nov17-opt REMARK Start: r_work = 0.2803 r_free = 0.3036 bonds = 0.012 angles = 1.488 REMARK Final: r_work = 0.2143 r_free = 0.2961 bonds = 0.032 angles = 3.281 b_ave=64.71 phenix.reduce may21_pdbset.pdb > may21_pdbset_h_added.pdb run3. with hydrogen phenix.refine 1205A-p21212.mtz nov138_001_h_added.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 output.write_maps=true output.prefix=nov17-wh REMARK Start: r_work = 0.2860 r_free = 0.3123 bonds = 0.085 angles = 1.672 REMARK Final: r_work = 0.2381 r_free = 0.3029 bonds = 0.017 angles = 1.685 b_ave=77.7 run4. with both hydrogen and optimization phenix.refine 1205A-p21212.mtz nov138_001_h_added.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 optimize_wxc=true optimize_wxu=true output.write_maps=true output.prefix=nov17-wh-opt REMARK Start: r_work = 0.2860 r_free = 0.3123 bonds = 0.085 angles = 1.672 REMARK Final: r_work = 0.2447 r_free = 0.3032 bonds = 0.020 angles = 1.922 b_ave=80.75 It looks output from run1 is the best, adding hydrogen and optimizing wxc and wxu did not help much. I have had little better Rfree and rmsd at the end of my last run. You are right. the commands I used at the last stage looks weird. Before that run I have used "strategy=rigid_body+individual_sites+group_adp+tls" and also included bulk_solvent and scaling. The map is pretty good. But, the number of reflections used in the refinement was more than that in mtz file and that was the warning in the pdb validation summary letter. I think that's because default extension of high resolution range in PHENIX. So, I needed to run 1 macro cycle with a little lower high resolution 2.9A with the last refined pdb. I found those are the combination of commands gives the best R, Rfree and rmsd for bonds and angles. As soon as I was including bulk solven and scaling correction and/or b_group the values were increasing. Although I don't have clear explanation, but that is what I found. Any way these are very little differences and might not be so important. I fixed the problem in dna. The problem was related with the occupancy of the 2 alternate conformations of dna. But, I still have 8 amino acids' (out of 400) phy, psy out side Ramachandran plot. Probably this does not mater much if I I deposite this co-ordinate in PDB. Thanks for all your suggestions though. Regards... Raja ----- Original Message ----- From: Pavel Afonine Date: Tuesday, November 17, 2009 12:20 pm Subject: Re: [phenixbb] fixing geometry for deposition To: PHENIX user mailing list

Hi Raja,

...

I used the following commands at the last stage of

refinement:>

...

phenix.refine 1205A-p21212.mtz nov138_001.pdb simulated_annealing=false ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None main.bulk_solvent_and_scale=false strategy=rigid_body+individual_sites xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=1 output.write_maps=true output.prefix=nov139 --overwrite

The above command seems extremely weird to me for the following reasons: 1) Bulk-solvent correction and anisotropic scaling always has to be done (unless you are experimenting with fake data or these corrections have been already applied to Fobs, which is bad ideas anyway). By using "main.bulk_solvent_and_scale=false" you turn bulk-solvent correction and anisotropic scaling off.

2) By default, phenix.refine does 3 refinement macro-cycles, but often (depending how far you are from the final model) this is not enough for refinement to converge, so increasing it to 5 or so is a good idea. Doing just one macro-cycle (refinement.main.number_of_macro_cycles=1) does not make much sense to me. By the way, you can use shortcuts like "main.number_of_m=5".

3) You can drop this off the list since it is the default setting anyway: "simulated_annealing=false".

4) The refinement strategy: "strategy=rigid_body+individual_sites". Normally, you do the rigid body refinement at initial stages of refinement when your model is poor, and not at the final run. This is because the rigid-body refinement can be very rude on your model: bonds can be broken between rigid groups since no restraints is used, for example. Also, it's strange that the ADP refinement is turned off - it's always good to do, and at 2.8A you can still refined individual ADPs. Well, I'm not mention using TLS...

So, I would modify the above command as following:

phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 output.write_maps=true output.prefix=test --overwrite

and try running it as is, and with adding keywords "optimize_wxc=true optimize_wxu=true", using a model with or w/o H riding H atoms, etc... what I wrote in my previous email.

...

Regarding amino acids phy, psy..... I tried real space refinement/regularization in coot,

This is not the same as what I suggested to try in my previous email. Anyway, I would first try the modified command above and only then try other suggested things.

Good luck, Pavel.

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

Hi Pavel, Whenever I run ncs using the default parameters I get the following: The current limit is: refinement.ncs.excessive_distance_limit=1.5 The number of distances exceeding this limit is: 364 Please correct your model or redefine the limit, e.g. with: refinement.ncs.excessive_distance_limit=3 To disable this message completely define: refinement.ncs.excessive_distance_limit=None It runs fine when you put refinement.ncs.excessive_distance_limit=None in the command line. My question is what would be the best distance for the program to run without this message. thanks, Shya

Hi Raja,

here are a few comments:

- if you used TLS previously and then run a refinement without TLS (that is "strategy=..." does not include "tls"), the previous TLS refinement gets invalidated, because phenix.refine will automatically convert ANISOU into isotropic equivalent. It says it somewhere in .log file.

- the "run 1" looks better than the others indeed (judging by the R-factors and bond/angles RMSDs at this resolution).

- your original concern was the geometry problems, so using H in refinement, as well as optimizing weights, may help fixing them. So I would check the geometry too for all the runs below, and not only R-factors.

- yesterday Kendall suggested a set of good ideas to try out regarding NCS.

- I did not understand what you mean by "default extension of high resolution range in PHENIX". phenix.refine uses all reflections that it finds in your input data files, unless specified otherwise using high/low resolution cutoffs and sigma cutoffs. phenix.refine does not collect any additional data for you -:)

- at 2.9A it is very likely that you still ok to refine individual ADPs and switching to group ADP can increase R-factors. So no surprises here. I doubt that using bulk-solvent correction and anisotropic scaling increases the R-factors; probably you tried it in combination with something else, and that "something else" was the cause of increased R-values.

- are you using the latest version of PHENIX ?

Pavel.

On 11/18/09 8:47 AM, Raja Dey wrote:

...

Hi Pavel, I finished 4 runs as you suggested. See below: run1. phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 output.write_maps=true output.prefix=nov17 REMARK Start: r_work = 0.2803 r_free = 0.3036 bonds = 0.012 angles = 1.488 REMARK Final: r_work = 0.2438 r_free = 0.2994 bonds = 0.009 angles = 1.309 b_ave=70.09

run2. with wxc and wxu optimization phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 optimize_wxc=true optimize_wxu=true output.write_maps=true output.prefix=nov17-opt REMARK Start: r_work = 0.2803 r_free = 0.3036 bonds = 0.012 angles = 1.488 REMARK Final: r_work = 0.2143 r_free = 0.2961 bonds = 0.032 angles = 3.281 b_ave=64.71

phenix.reduce may21_pdbset.pdb > may21_pdbset_h_added.pdb

run3. with hydrogen phenix.refine 1205A-p21212.mtz nov138_001_h_added.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 output.write_maps=true output.prefix=nov17-wh REMARK Start: r_work = 0.2860 r_free = 0.3123 bonds = 0.085 angles = 1.672 REMARK Final: r_work = 0.2381 r_free = 0.3029 bonds = 0.017 angles = 1.685 b_ave=77.7

run4. with both hydrogen and optimization phenix.refine 1205A-p21212.mtz nov138_001_h_added.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 optimize_wxc=true optimize_wxu=true output.write_maps=true output.prefix=nov17-wh-opt REMARK Start: r_work = 0.2860 r_free = 0.3123 bonds = 0.085 angles = 1.672 REMARK Final: r_work = 0.2447 r_free = 0.3032 bonds = 0.020 angles = 1.922 b_ave=80.75

It looks output from run1 is the best, adding hydrogen and optimizing wxc and wxu did not help much. I have had little better Rfree and rmsd at the end of my last run.

You are right. the commands I used at the last stage looks weird. Before that run I have used "strategy=rigid_body+individual_sites+group_adp+tls" and also included bulk_solvent and scaling. The map is pretty good. But, the number of reflections used in the refinement was more than that in mtz file and that was the warning in the pdb validation summary letter. I think that's because default extension of high resolution range in PHENIX. So, I needed to run 1 macro cycle with a little lower high resolution 2.9A with the last refined pdb. I found those are the combination of commands gives the best R, Rfree and rmsd for bonds and angles. As soon as I was including bulk solven and scaling correction and/or b_group the values were increasing. Although I don't have clear explanation, but that is what I found. Any way these are very little differences and might not be so important.

I fixed the problem in dna. The problem was related with the occupancy of the 2 alternate conformations of dna. But, I still have 8 amino acids' (out of 400) phy, psy out side Ramachandran plot. Probably this does not mater much if I I deposite this co-ordinate in PDB. Thanks for all your suggestions though.

Regards... Raja

----- Original Message ----- From: Pavel Afonine Date: Tuesday, November 17, 2009 12:20 pm Subject: Re: [phenixbb] fixing geometry for deposition To: PHENIX user mailing list

...

Hi Raja,

...

I used the following commands at the last stage of

refinement:>

...

phenix.refine 1205A-p21212.mtz nov138_001.pdb

simulated_annealing=false ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None main.bulk_solvent_and_scale=false strategy=rigid_body+individual_sites xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=1 output.write_maps=true output.prefix=nov139 --overwrite

...

The above command seems extremely weird to me for the following reasons: 1) Bulk-solvent correction and anisotropic scaling always has to be done (unless you are experimenting with fake data or these corrections have been already applied to Fobs, which is bad ideas anyway). By using "main.bulk_solvent_and_scale=false" you turn bulk-solvent correction and anisotropic scaling off.

2) By default, phenix.refine does 3 refinement macro-cycles, but often (depending how far you are from the final model) this is not enough for refinement to converge, so increasing it to 5 or so is a good idea. Doing just one macro-cycle (refinement.main.number_of_macro_cycles=1) does not make much sense to me. By the way, you can use shortcuts like "main.number_of_m=5".

3) You can drop this off the list since it is the default setting anyway: "simulated_annealing=false".

4) The refinement strategy: "strategy=rigid_body+individual_sites". Normally, you do the rigid body refinement at initial stages of refinement when your model is poor, and not at the final run. This is because the rigid-body refinement can be very rude on your model: bonds can be broken between rigid groups since no restraints is used, for example. Also, it's strange that the ADP refinement is turned off - it's always good to do, and at 2.8A you can still refined individual ADPs. Well, I'm not mention using TLS...

So, I would modify the above command as following:

phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 output.write_maps=true output.prefix=test --overwrite

and try running it as is, and with adding keywords "optimize_wxc=true optimize_wxu=true", using a model with or w/o H riding H atoms, etc... what I wrote in my previous email.

...

Regarding amino acids phy, psy..... I tried real space

refinement/regularization in coot,

This is not the same as what I suggested to try in my previous email. Anyway, I would first try the modified command above and only then try other suggested things.

Good luck, Pavel.

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

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

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

By the way, it may be worth of repeating that the .geo file contains the list of all the geometry restraints used in phenix.refine refinement (bonds, angles, planarities, chiralities, dihedrals, non-bonded, NCS, custom defined restraints and edits), for each and every atom, making it a complete foot-print of geometry restraints used to obtain the final refined structure. Pavel. On 11/18/09 11:12 AM, Pavel Afonine wrote:

Hi Shya,

inspect the .geo file for something like "EXCESSIVE DISTANCE" (or something similar, I don't remember exactly), and that will give you an idea by how much you exceed the default 1.5 value.

It might be worth of looking at those atoms in NCS copies that have too large deviations between NCS copies, and if you find that it is for the reason, then you may want to exclude them from NCS.

Good luck! Pavel.

On 11/18/09 11:07 AM, [email protected] wrote:

...

Hi Pavel,

Whenever I run ncs using the default parameters I get the following: The current limit is: refinement.ncs.excessive_distance_limit=1.5 The number of distances exceeding this limit is: 364 Please correct your model or redefine the limit, e.g. with: refinement.ncs.excessive_distance_limit=3 To disable this message completely define: refinement.ncs.excessive_distance_limit=None

It runs fine when you put refinement.ncs.excessive_distance_limit=None in the command line. My question is what would be the best distance for the program to run without this message. thanks, Shya

...

Hi Raja,

here are a few comments:

- if you used TLS previously and then run a refinement without TLS (that is "strategy=..." does not include "tls"), the previous TLS refinement gets invalidated, because phenix.refine will automatically convert ANISOU into isotropic equivalent. It says it somewhere in .log file.

- the "run 1" looks better than the others indeed (judging by the R-factors and bond/angles RMSDs at this resolution).

- your original concern was the geometry problems, so using H in refinement, as well as optimizing weights, may help fixing them. So I would check the geometry too for all the runs below, and not only R-factors.

- yesterday Kendall suggested a set of good ideas to try out regarding NCS.

- I did not understand what you mean by "default extension of high resolution range in PHENIX". phenix.refine uses all reflections that it finds in your input data files, unless specified otherwise using high/low resolution cutoffs and sigma cutoffs. phenix.refine does not collect any additional data for you -:)

- at 2.9A it is very likely that you still ok to refine individual ADPs and switching to group ADP can increase R-factors. So no surprises here. I doubt that using bulk-solvent correction and anisotropic scaling increases the R-factors; probably you tried it in combination with something else, and that "something else" was the cause of increased R-values.

- are you using the latest version of PHENIX ?

Pavel.

On 11/18/09 8:47 AM, Raja Dey wrote:

...

Hi Pavel, I finished 4 runs as you suggested. See below: run1. phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 output.write_maps=true output.prefix=nov17 REMARK Start: r_work = 0.2803 r_free = 0.3036 bonds = 0.012 angles = 1.488 REMARK Final: r_work = 0.2438 r_free = 0.2994 bonds = 0.009 angles = 1.309 b_ave=70.09

run2. with wxc and wxu optimization phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 optimize_wxc=true optimize_wxu=true output.write_maps=true output.prefix=nov17-opt REMARK Start: r_work = 0.2803 r_free = 0.3036 bonds = 0.012 angles = 1.488 REMARK Final: r_work = 0.2143 r_free = 0.2961 bonds = 0.032 angles = 3.281 b_ave=64.71

phenix.reduce may21_pdbset.pdb > may21_pdbset_h_added.pdb

run3. with hydrogen phenix.refine 1205A-p21212.mtz nov138_001_h_added.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 output.write_maps=true output.prefix=nov17-wh REMARK Start: r_work = 0.2860 r_free = 0.3123 bonds = 0.085 angles = 1.672 REMARK Final: r_work = 0.2381 r_free = 0.3029 bonds = 0.017 angles = 1.685 b_ave=77.7

run4. with both hydrogen and optimization phenix.refine 1205A-p21212.mtz nov138_001_h_added.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 optimize_wxc=true optimize_wxu=true output.write_maps=true output.prefix=nov17-wh-opt REMARK Start: r_work = 0.2860 r_free = 0.3123 bonds = 0.085 angles = 1.672 REMARK Final: r_work = 0.2447 r_free = 0.3032 bonds = 0.020 angles = 1.922 b_ave=80.75

It looks output from run1 is the best, adding hydrogen and optimizing wxc and wxu did not help much. I have had little better Rfree and rmsd at the end of my last run.

You are right. the commands I used at the last stage looks weird. Before that run I have used "strategy=rigid_body+individual_sites+group_adp+tls" and also included bulk_solvent and scaling. The map is pretty good. But, the number of reflections used in the refinement was more than that in mtz file and that was the warning in the pdb validation summary letter. I think that's because default extension of high resolution range in PHENIX. So, I needed to run 1 macro cycle with a little lower high resolution 2.9A with the last refined pdb. I found those are the combination of commands gives the best R, Rfree and rmsd for bonds and angles. As soon as I was including bulk solven and scaling correction and/or b_group the values were increasing. Although I don't have clear explanation, but that is what I found. Any way these are very little differences and might not be so important.

I fixed the problem in dna. The problem was related with the occupancy of the 2 alternate conformations of dna. But, I still have 8 amino acids' (out of 400) phy, psy out side Ramachandran plot. Probably this does not mater much if I I deposite this co-ordinate in PDB. Thanks for all your suggestions though.

Regards... Raja

----- Original Message ----- From: Pavel Afonine Date: Tuesday, November 17, 2009 12:20 pm Subject: Re: [phenixbb] fixing geometry for deposition To: PHENIX user mailing list

...

Hi Raja,

...

I used the following commands at the last stage of

refinement:>

...

phenix.refine 1205A-p21212.mtz nov138_001.pdb

simulated_annealing=false ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None main.bulk_solvent_and_scale=false strategy=rigid_body+individual_sites xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=1 output.write_maps=true output.prefix=nov139 --overwrite

...

The above command seems extremely weird to me for the following reasons: 1) Bulk-solvent correction and anisotropic scaling always has to be done (unless you are experimenting with fake data or these corrections have been already applied to Fobs, which is bad ideas anyway). By using "main.bulk_solvent_and_scale=false" you turn bulk-solvent correction and anisotropic scaling off.

2) By default, phenix.refine does 3 refinement macro-cycles, but often (depending how far you are from the final model) this is not enough for refinement to converge, so increasing it to 5 or so is a good idea. Doing just one macro-cycle (refinement.main.number_of_macro_cycles=1) does not make much sense to me. By the way, you can use shortcuts like "main.number_of_m=5".

3) You can drop this off the list since it is the default setting anyway: "simulated_annealing=false".

4) The refinement strategy: "strategy=rigid_body+individual_sites". Normally, you do the rigid body refinement at initial stages of refinement when your model is poor, and not at the final run. This is because the rigid-body refinement can be very rude on your model: bonds can be broken between rigid groups since no restraints is used, for example. Also, it's strange that the ADP refinement is turned off - it's always good to do, and at 2.8A you can still refined individual ADPs. Well, I'm not mention using TLS...

So, I would modify the above command as following:

phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 output.write_maps=true output.prefix=test --overwrite

and try running it as is, and with adding keywords "optimize_wxc=true optimize_wxu=true", using a model with or w/o H riding H atoms, etc... what I wrote in my previous email.

...

Regarding amino acids phy, psy..... I tried real space

refinement/regularization in coot,

This is not the same as what I suggested to try in my previous email. Anyway, I would first try the modified command above and only then try other suggested things.

Good luck, Pavel.

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

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

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

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

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

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

Hi Maia, just looking at distances is not enough: in addition, one needs to look at how those residues appear in maps. It's a part of planned improvements of NCS refinement. However, for the moment phenix.refine stops prompting the user to check the NCS groups and problem residues. I saw a number of cases where this helped to identify wrong NCS selections due to atom selection syntax errors. Pavel. On 11/21/09 8:56 AM, Maia Cherney wrote:

Hi Pavel, I would like to make a suggestion about the use of excessive_distance_limit. Instead of shutting down the refinement due to excessive_distance over the limit why not use it for excluding only those residues from NCS refinement and do the ncs refinement for the rest of the residues.

Maia

Pavel Afonine wrote:

...

By the way, it may be worth of repeating that the .geo file contains the list of all the geometry restraints used in phenix.refine refinement (bonds, angles, planarities, chiralities, dihedrals, non-bonded, NCS, custom defined restraints and edits), for each and every atom, making it a complete foot-print of geometry restraints used to obtain the final refined structure.

Pavel.

On 11/18/09 11:12 AM, Pavel Afonine wrote:

...

Hi Shya,

inspect the .geo file for something like "EXCESSIVE DISTANCE" (or something similar, I don't remember exactly), and that will give you an idea by how much you exceed the default 1.5 value.

It might be worth of looking at those atoms in NCS copies that have too large deviations between NCS copies, and if you find that it is for the reason, then you may want to exclude them from NCS.

Good luck! Pavel.

On 11/18/09 11:07 AM, [email protected] wrote:

...

Hi Pavel,

Whenever I run ncs using the default parameters I get the following: The current limit is: refinement.ncs.excessive_distance_limit=1.5 The number of distances exceeding this limit is: 364 Please correct your model or redefine the limit, e.g. with: refinement.ncs.excessive_distance_limit=3 To disable this message completely define: refinement.ncs.excessive_distance_limit=None

It runs fine when you put refinement.ncs.excessive_distance_limit=None in the command line. My question is what would be the best distance for the program to run without this message. thanks, Shya

...

Hi Raja,

here are a few comments:

- if you used TLS previously and then run a refinement without TLS (that is "strategy=..." does not include "tls"), the previous TLS refinement gets invalidated, because phenix.refine will automatically convert ANISOU into isotropic equivalent. It says it somewhere in .log file.

- the "run 1" looks better than the others indeed (judging by the R-factors and bond/angles RMSDs at this resolution).

- your original concern was the geometry problems, so using H in refinement, as well as optimizing weights, may help fixing them. So I would check the geometry too for all the runs below, and not only R-factors.

- yesterday Kendall suggested a set of good ideas to try out regarding NCS.

- I did not understand what you mean by "default extension of high resolution range in PHENIX". phenix.refine uses all reflections that it finds in your input data files, unless specified otherwise using high/low resolution cutoffs and sigma cutoffs. phenix.refine does not collect any additional data for you -:)

- at 2.9A it is very likely that you still ok to refine individual ADPs and switching to group ADP can increase R-factors. So no surprises here. I doubt that using bulk-solvent correction and anisotropic scaling increases the R-factors; probably you tried it in combination with something else, and that "something else" was the cause of increased R-values.

- are you using the latest version of PHENIX ?

Pavel.

On 11/18/09 8:47 AM, Raja Dey wrote:

...

Hi Pavel, I finished 4 runs as you suggested. See below: run1. phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 output.write_maps=true output.prefix=nov17 REMARK Start: r_work = 0.2803 r_free = 0.3036 bonds = 0.012 angles = 1.488 REMARK Final: r_work = 0.2438 r_free = 0.2994 bonds = 0.009 angles = 1.309 b_ave=70.09

run2. with wxc and wxu optimization phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 optimize_wxc=true optimize_wxu=true output.write_maps=true output.prefix=nov17-opt REMARK Start: r_work = 0.2803 r_free = 0.3036 bonds = 0.012 angles = 1.488 REMARK Final: r_work = 0.2143 r_free = 0.2961 bonds = 0.032 angles = 3.281 b_ave=64.71

phenix.reduce may21_pdbset.pdb > may21_pdbset_h_added.pdb

run3. with hydrogen phenix.refine 1205A-p21212.mtz nov138_001_h_added.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 output.write_maps=true output.prefix=nov17-wh REMARK Start: r_work = 0.2860 r_free = 0.3123 bonds = 0.085 angles = 1.672 REMARK Final: r_work = 0.2381 r_free = 0.3029 bonds = 0.017 angles = 1.685 b_ave=77.7

run4. with both hydrogen and optimization phenix.refine 1205A-p21212.mtz nov138_001_h_added.pdb ncs_groups.params main.ncs=true ncs.find_automatically=false refinement.ncs.excessive_distance_limit=None xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=5 optimize_wxc=true optimize_wxu=true output.write_maps=true output.prefix=nov17-wh-opt REMARK Start: r_work = 0.2860 r_free = 0.3123 bonds = 0.085 angles = 1.672 REMARK Final: r_work = 0.2447 r_free = 0.3032 bonds = 0.020 angles = 1.922 b_ave=80.75

It looks output from run1 is the best, adding hydrogen and optimizing wxc and wxu did not help much. I have had little better Rfree and rmsd at the end of my last run.

You are right. the commands I used at the last stage looks weird. Before that run I have used "strategy=rigid_body+individual_sites+group_adp+tls" and also included bulk_solvent and scaling. The map is pretty good. But, the number of reflections used in the refinement was more than that in mtz file and that was the warning in the pdb validation summary letter. I think that's because default extension of high resolution range in PHENIX. So, I needed to run 1 macro cycle with a little lower high resolution 2.9A with the last refined pdb. I found those are the combination of commands gives the best R, Rfree and rmsd for bonds and angles. As soon as I was including bulk solven and scaling correction and/or b_group the values were increasing. Although I don't have clear explanation, but that is what I found. Any way these are very little differences and might not be so important.

I fixed the problem in dna. The problem was related with the occupancy of the 2 alternate conformations of dna. But, I still have 8 amino acids' (out of 400) phy, psy out side Ramachandran plot. Probably this does not mater much if I I deposite this co-ordinate in PDB. Thanks for all your suggestions though.

Regards... Raja

----- Original Message ----- From: Pavel Afonine Date: Tuesday, November 17, 2009 12:20 pm Subject: Re: [phenixbb] fixing geometry for deposition To: PHENIX user mailing list

> Hi Raja, > > > > >> I used the following commands at the last stage of >> >> >> > refinement:> > > > >> phenix.refine 1205A-p21212.mtz nov138_001.pdb >> >> >> > simulated_annealing=false ncs_groups.params main.ncs=true > ncs.find_automatically=false > refinement.ncs.excessive_distance_limit=None > main.bulk_solvent_and_scale=false strategy=rigid_body+individual_sites > xray_data.high_resolution=2.9 refinement.main.number_of_macro_cycles=1 > output.write_maps=true output.prefix=nov139 --overwrite > > > >> >> > The above command seems extremely weird to me for the following > reasons: > 1) Bulk-solvent correction and anisotropic scaling always has to be > done > (unless you are experimenting with fake data or these corrections > have > been already applied to Fobs, which is bad ideas anyway). By using > "main.bulk_solvent_and_scale=false" you turn bulk-solvent > correction and > anisotropic scaling off. > > 2) By default, phenix.refine does 3 refinement macro-cycles, but > often > (depending how far you are from the final model) this is not enough > for > refinement to converge, so increasing it to 5 or so is a good idea. > Doing just one macro-cycle > (refinement.main.number_of_macro_cycles=1) > does not make much sense to me. By the way, you can use shortcuts > like > "main.number_of_m=5". > > 3) You can drop this off the list since it is the default setting > anyway: "simulated_annealing=false". > > 4) The refinement strategy: "strategy=rigid_body+individual_sites". > Normally, you do the rigid body refinement at initial stages of > refinement when your model is poor, and not at the final run. This > is > because the rigid-body refinement can be very rude on your model: > bonds > can be broken between rigid groups since no restraints is used, for > example. Also, it's strange that the ADP refinement is turned off - > it's > always good to do, and at 2.8A you can still refined individual > ADPs. > Well, I'm not mention using TLS... > > So, I would modify the above command as following: > > phenix.refine 1205A-p21212.mtz nov138_001.pdb ncs_groups.params > main.ncs=true ncs.find_automatically=false > refinement.ncs.excessive_distance_limit=None > xray_data.high_resolution=2.9 output.write_maps=true > output.prefix=test > --overwrite > > and try running it as is, and with adding keywords > "optimize_wxc=true > optimize_wxu=true", using a model with or w/o H riding H atoms, > etc... > what I wrote in my previous email. > > > > >> Regarding amino acids phy, psy..... I tried real space >> >> >> > refinement/regularization in coot, > > This is not the same as what I suggested to try in my previous > email. > Anyway, I would first try the modified command above and only then > try > other suggested things. > > Good luck, > Pavel. > > _______________________________________________ > phenixbb mailing list > [email protected] > http://www.phenix-online.org/mailman/listinfo/phenixbb > > > > _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

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

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

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

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

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

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

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

On Nov 22, 2009, at 11:50 AM, Maia Cherney wrote:

Hi Pavel, I just installed the phenix-1.5-2 version and got some problems using my previous .def file (from 1.4-4 version).

The result of the refinement was a new pdb file, but no mtz file with map coefficients. The error message:

Sorry: Duplicate mtz_label_amplitudes:FOFCWT

I guess, I need to remove some lines from my old .def in a new refinement run, but can I get the mtz without rerunning the refinement (it takes a very long time to rerun it with tls etc.)

Two methods: 1) Run phenix.create_maps (or click "Create Maps" in the GUI). Note that by default, this program does not automatically fill missing F(obs) with F(calc) in the standard weighted difference maps, but phenix.refine does. You'll probably want to turn this on. 2) phenix.model_vs_data coords.pdb data.mtz --map=2mFo-DFc_filled phenix.model_vs_data coords.pdb data.mtz --map=mFo-DFc_filled ("_filled" is optional) -Nat -------------------- Nathaniel Echols Lawrence Berkeley Lab 510-486-5136 [email protected]

With F+ and F- you can generate Dano maps in CCP4 using phases from your current model. ________________________________________ From: Maia Cherney [[email protected]] Sent: Sunday, November 22, 2009 7:00 PM To: PHENIX user mailing list Subject: Re: [phenixbb] regarding ncs distance - .geo file Hi Randy, Pavel I want to verify some sulfur atoms in my ligands. I would like to use sulfur anomalous dispersion. I know the protein structure. What is the best way to generate an anomalous difference map? Is there a "phenix.find_anomalous_scatterers_from_model" or similar command in phenix? Maia _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

Apologies Pavel, Its CCP4ese for anomalous difference map

Hi X-gurus,

just to remove some of my ignorance,... and sorry in advance for a stupid question: What is "(...) Dano maps in CCP4 using phases from your current model (...)". If you tell me exactly what it is I might be able to add this to PHENIX family of tools (or at least put it to to-do list)...

All the best! Pavel.

On 11/22/09 5:36 PM, Mayer, Mark (NIH/NICHD) [E] wrote:

...

With F+ and F- you can generate Dano maps in CCP4 using phases from your current model. ________________________________________ From: Maia Cherney [[email protected]] Sent: Sunday, November 22, 2009 7:00 PM To: PHENIX user mailing list Subject: Re: [phenixbb] regarding ncs distance - .geo file

Hi Randy, Pavel

I want to verify some sulfur atoms in my ligands. I would like to use sulfur anomalous dispersion. I know the protein structure. What is the best way to generate an anomalous difference map? Is there a "phenix.find_anomalous_scatterers_from_model" or similar command in phenix?

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

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

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

-- +++++++++++++++++++++++++++++++++++++++++++++++++++++++ Mark Mayer Ph.D. LCMN NICHD NIH DHHS Bldg 35, Room 3B 1002 MSC 3712 35 Lincoln Drive Bethesda MD 20892 3712 Phone: 301-496-9346 (office); 301-496-9347 (lab); FAX 301-496-2396 Lab web site: http://snb.nichd.nih.gov/index.htm Send packages, Fedex and anything requiring a signature to: Bldg 35, Room 3B 1004 35 Lincoln Drive Bethesda MD 20892

On Nov 22, 2009, at 4:00 PM, Maia Cherney wrote:

I want to verify some sulfur atoms in my ligands. I would like to use sulfur anomalous dispersion. I know the protein structure. What is the best way to generate an anomalous difference map? Is there a "phenix.find_anomalous_scatterers_from_model" or similar command in phenix?

You can generate the anomalous difference map using either of the same commands in my last message - the only difference is that the map type is simply "anomalous". -------------------- Nathaniel Echols Lawrence Berkeley Lab 510-486-5136 [email protected]

Hi, I am wondering what is the criteria for occupancy refinement of ligands. I noticed that R factors change very little, but the ligand B-factors change significantly . On the other hand, the occupancy is refined to the second digit after the decimal point. How can I find out the error for the refined occupancy of ligands? Maia

Hi, I am wondering what is the criteria for occupancy refinement of ligands. I noticed that R factors change very little, but the ligand B-factors change significantly . On the other hand, the occupancy is refined to the second digit after the decimal point. How can I find out the error for the refined occupancy of ligands? Maia

Hi Maia, I think the criteria for occupancy refinement of ligands is similar to a decision to add an alt conformation for an amino acid. I don't refine occupancy of a ligand unless the difference map indicates that we have to. Sometimes part of the igand may be conformationally mobile and show poor density, but I personally don't think this justifies occupancy refinement without evidence from the difference map. I agree with Pavel that you shouldn't expect much change in overall statistics, unless the ligand has very low occupancy., or you have a very small protein. We typically see 0.5-1% difference in R factors from refining with ligand versus without for nuclear receptor igand binding domains of about 250 amino acids, and we see very small differences from occupancy refinement of the ligands. Regarding the error, I have noticed differences of 10% percent occupancy depending on what you set the starting occupancy before refinement. That is, if the starting occupancy starts at 1, you might end up with 50%, but if you start it at 0.01, you might get 40%. I don't have the expertise to explain why this is, but I also don't think it is necessarily important. I think it is more important to convince yourself that the ligand binds how you think it does. With steroid receptors, the ligand is usually planer, and tethered by hydrogen bonds on two ends. That leaves us with with four possible poses, so if in doubt, we will dock in the ligand in all of the four orientations and refine. So far, we have had only one of several dozen structures where the ligand orientation was not obvious after this procedure. I worry about a letter to the editor suggesting that the electron density for the ligand doesn't support the conclusions of the paper, not whether the occupancy is 40% versus 50%. You might also want to consider looking at several maps, such as the simple or simulated annealing composite omit maps. These can be noisy, so also try the kicked maps ( http://www.phenix-online.org/pipermail/phenixbb/2009-September/002573.html), which I have become a big fan of. Regards, Kendall Nettles On 11/24/09 3:07 PM, "[email protected]" wrote: Hi, I am wondering what is the criteria for occupancy refinement of ligands. I noticed that R factors change very little, but the ligand B-factors change significantly . On the other hand, the occupancy is refined to the second digit after the decimal point. How can I find out the error for the refined occupancy of ligands? Maia _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

Re: [phenixbb] occupancy refinementHi, I just remembered that the 10% difference in occupancy was most likely the result of not reaching the convergence. If you continue refinement, then the occupancy converges. Pavel, my question is this. If occupancy depends on B-factor, how come the occupancy refinement gives you such an accurate number for an occupancy (let's say 0.44). Maia ----- Original Message ----- From: chern To: PHENIX user mailing list Sent: Tuesday, November 24, 2009 9:08 PM Subject: Re: [phenixbb] occupancy refinement Thank you Kendall and Pavel for your responces. I really want to determine the occupancy of my ligand. I saw one suggestion to try different refinements with different occupancies and compare the B-factors. The occupancy with a B-factor that is at the level with the average protein B-factors, is a "true" occupancy. I also noticed the dependence of the ligand occupancy on the initial occupancy. I saw the difference of 10 to 15%, that is why I am wondering if the second digit after the decimal point makes any sence. Maia ----- Original Message ----- From: Kendall Nettles To: PHENIX user mailing list Sent: Tuesday, November 24, 2009 8:22 PM Subject: Re: [phenixbb] occupancy refinement Hi Maia, I think the criteria for occupancy refinement of ligands is similar to a decision to add an alt conformation for an amino acid. I don't refine occupancy of a ligand unless the difference map indicates that we have to. Sometimes part of the igand may be conformationally mobile and show poor density, but I personally don't think this justifies occupancy refinement without evidence from the difference map. I agree with Pavel that you shouldn't expect much change in overall statistics, unless the ligand has very low occupancy., or you have a very small protein. We typically see 0.5-1% difference in R factors from refining with ligand versus without for nuclear receptor igand binding domains of about 250 amino acids, and we see very small differences from occupancy refinement of the ligands. Regarding the error, I have noticed differences of 10% percent occupancy depending on what you set the starting occupancy before refinement. That is, if the starting occupancy starts at 1, you might end up with 50%, but if you start it at 0.01, you might get 40%. I don't have the expertise to explain why this is, but I also don't think it is necessarily important. I think it is more important to convince yourself that the ligand binds how you think it does. With steroid receptors, the ligand is usually planer, and tethered by hydrogen bonds on two ends. That leaves us with with four possible poses, so if in doubt, we will dock in the ligand in all of the four orientations and refine. So far, we have had only one of several dozen structures where the ligand orientation was not obvious after this procedure. I worry about a letter to the editor suggesting that the electron density for the ligand doesn't support the conclusions of the paper, not whether the occupancy is 40% versus 50%. You might also want to consider looking at several maps, such as the simple or simulated annealing composite omit maps. These can be noisy, so also try the kicked maps ( http://www.phenix-online.org/pipermail/phenixbb/2009-September/002573.html), which I have become a big fan of. Regards, Kendall Nettles On 11/24/09 3:07 PM, "[email protected]" wrote: Hi, I am wondering what is the criteria for occupancy refinement of ligands. I noticed that R factors change very little, but the ligand B-factors change significantly . On the other hand, the occupancy is refined to the second digit after the decimal point. How can I find out the error for the refined occupancy of ligands? Maia _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb ---------------------------------------------------------------------------- _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb ------------------------------------------------------------------------------ _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

Hi Pavel, It looks like all different refined occupancies starting from different initial occupancies converged to the same number upon going through very many cycles of refinement. Maia Pavel Afonine wrote:

Hi Maia,

the atom parameters, such as occupancy, B-factor and even position are interdependent in some sense. That is, if you have somewhat incorrect occupancy, that B-factor refinement may compensate for it; if you misplaced an atom the refinement of its occupancy or/and B-factor will compensate for this. Note in all the above cases the 2mFo-DFc and mFo-DFc maps will appear almost identical, as well as R-factors.

So, I think your goal of finding a "true" occupancy is hardly achievable.

Although, I think you can approach it by doing very many refinements (say, several hundreds) (where you refine occupancies, B-factors and coordinates) each refinement starting with different occupancy and B-factor values, and make sure that each refinement converges. Then select a subset of refined structures with similar and low R-factors (discard those cases where refinement got stuck for whatever reason and R-factors are higher) (and probably similar looking 2mFo-DFc and mFo-DFc maps in the region of interest). Then see where the refined occupancies and B-factors are clustering, and the averaged values will probably give you an approximate values for occupancy and B. I did not try this myself but always wanted to.

If you have a structure consisting of 9 carbons and one gold atom, then I would expect that the "second digit" in gold's occupancy would matter. However, if we speak about dozen of ligand atoms (which are probably a combination of C,N,O) out of a few thousands of atoms of the whole structure, then I would not expect the "second digit" to be visibly important.

Pavel.

On 11/24/09 8:08 PM, chern wrote:

...

Thank you Kendall and Pavel for your responces. I really want to determine the occupancy of my ligand. I saw one suggestion to try different refinements with different occupancies and compare the B-factors. The occupancy with a B-factor that is at the level with the average protein B-factors, is a "true" occupancy. I also noticed the dependence of the ligand occupancy on the initial occupancy. I saw the difference of 10 to 15%, that is why I am wondering if the second digit after the decimal point makes any sence. Maia

----- Original Message ----- *From:* Kendall Nettles mailto:[email protected] *To:* PHENIX user mailing list mailto:[email protected] *Sent:* Tuesday, November 24, 2009 8:22 PM *Subject:* Re: [phenixbb] occupancy refinement

Hi Maia, I think the criteria for occupancy refinement of ligands is similar to a decision to add an alt conformation for an amino acid. I don’t refine occupancy of a ligand unless the difference map indicates that we have to. Sometimes part of the igand may be conformationally mobile and show poor density, but I personally don’t think this justifies occupancy refinement without evidence from the difference map. I agree with Pavel that you shouldn’t expect much change in overall statistics, unless the ligand has very low occupancy., or you have a very small protein. We typically see 0.5-1% difference in R factors from refining with ligand versus without for nuclear receptor igand binding domains of about 250 amino acids, and we see very small differences from occupancy refinement of the ligands.

Regarding the error, I have noticed differences of 10% percent occupancy depending on what you set the starting occupancy before refinement. That is, if the starting occupancy starts at 1, you might end up with 50%, but if you start it at 0.01, you might get 40%. I don’t have the expertise to explain why this is, but I also don’t think it is necessarily important. I think it is more important to convince yourself that the ligand binds how you think it does. With steroid receptors, the ligand is usually planer, and tethered by hydrogen bonds on two ends. That leaves us with with four possible poses, so if in doubt, we will dock in the ligand in all of the four orientations and refine. So far, we have had only one of several dozen structures where the ligand orientation was not obvious after this procedure. I worry about a letter to the editor suggesting that the electron density for the ligand doesn’t support the conclusions of the paper, not whether the occupancy is 40% versus 50%.

You might also want to consider looking at several maps, such as the simple or simulated annealing composite omit maps. These can be noisy, so also try the kicked maps ( http://www.phenix-online.org/pipermail/phenixbb/2009-September/002573.html), http://www.phenix-online.org/pipermail/phenixbb/2009-September/002573.html%2... which I have become a big fan of.

Regards, Kendall Nettles

On 11/24/09 3:07 PM, "[email protected]" wrote:

Hi, I am wondering what is the criteria for occupancy refinement of ligands. I noticed that R factors change very little, but the ligand B-factors change significantly . On the other hand, the occupancy is refined to the second digit after the decimal point. How can I find out the error for the refined occupancy of ligands?

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

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

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

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

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

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

You might also consider excluding the regions with poor geometry from the NCS groups, to see if they are in fact different. You are also forcing NCS regardless of the RMS distance. I would refine a few rounds without NCS and with the default refinement parameters, and then then turn the NCS back on. Look in the .geo file for excessive distances, and examine these regions in Coot using the NCS ghosts and maps to decide if they should be excluded from the NCS groups.

Thanks Kendall, it is very good point. We should probably do something like this to automatically adjust NCS restraints during refinement. Pavel.