[phenixbb] Strong anomalous signal but AutoSol fails
mohamed noor
mohamed.noor34 at gmail.com
Tue Aug 11 10:42:33 PDT 2015
Hi Diana, Tom
I am able to obtain an MR solution with a different crystal form in P41212.
Autobuild and refinement to 3.8 A gives me an Rfree of about 30 %. When I
go back to my higher resolution data in the point group P321, Phaser runs
with the search model that I used for the P41212 data failed. However, I
can get a solution (TFZ=14, LLG=400) with Phaser.MRage using the lower
resolution (P41212) model.
Autobuilding-refining this solution gives an Rfree of 48 %. So I think
something is wrong somewhere but I can't figure out where. XDS selected
P321 and this was more or less confirmed by Pointless and Xtriage. When I
checked FRAME.cbf, all spots were predicted and the number of spots
rejected per frame were quite low (5-10 rejected out of 200 or so strong
reflections).
I remember having a pseudosymmetry case in a different protein recently
whereby I had good anomalous signal in P321 (yes, same thing again!!) with
the Rfree being stuck in 35 % (1.2 A dataset). But I reprocessed it to P3
and refinement and map was much better (Rfree around 22 %) although the
anomalous signal was not there.
In short, I am currently running a few more Phaser runs in P3, P6 and P622
point groups and checking for all the possible space groups.
Thanks again.
On Mon, Aug 10, 2015 at 5:09 PM, Diana Tomchick <
Diana.Tomchick at utsouthwestern.edu> wrote:
> I second the recommendation on calculating a rotation function, though you
> may need some assistance with interpreting the results of the calculation.
>
> Running any phasing program (including AutoSol) at lower resolution will
> give you higher absolute values for the FOM but that doesn’t translate to a
> real solution.
>
> At this point in the game, I would start to question whether the Laue
> symmetry that you chose for your data processing is correct—it is possible
> that you should have a higher symmetry, or you have pseudosymmetry. Either
> possibility could really complicate the location of MR models and heavy
> atom sites.
>
> If you can use SHELX to find heavy atom sites, then it’s been my
> experience that 80% of the time you can put your data into PHENIX and
> AutoSol will give you a solution. The other 20% of the time requires more
> manual effort and fiddling.
>
> Diana
>
> **************************************************
> Diana R. Tomchick
> Professor
> Departments of Biophysics and Biochemistry
> University of Texas Southwestern Medical Center
> 5323 Harry Hines Blvd.
> Rm. ND10.214A
> Dallas, TX 75390-8816
> Diana.Tomchick at UTSouthwestern.edu
> (214) 645-6383 (phone)
> (214) 645-6353 (fax)
>
> > On Aug 9, 2015, at 10:29 AM, Terwilliger, Thomas Charles <
> terwilliger at lanl.gov> wrote:
> >
> > Hi Mohamed,
> >
> > I don't know of any way to know the number of copies in the ASU exactly,
> no.
> >
> > It does seem that the solutions you have found so far are not likely to
> be correct, as you suggest. Here is one more general approach to try: as
> you have a high degree of NCS, try to use it to evaluate your solutions. I
> assume there is 1 Fe per molecule. With high NCS in the crystal, it is
> likely (but not for sure) that the molecules in the ASU have some kind of
> symmetric arrangement. So you could look at any potential substructure
> solutions you find and see if there is any symmetry in the sites. You can
> just look at them in Coot with symmetry turned on or you can use
> phenix.find_ncs to look for NCS in the sites.
> >
> > Additionally you could calculate a rotation function from your native
> data and from your anomalous differences and see if there are NCS rotation
> axes that are present (for example a 13-fold rotation axis or something
> like that).
> >
> > All the best,
> > Tom T
> >
> > From: mohamed noor [mohamed.noor34 at gmail.com]
> > Sent: Sunday, August 09, 2015 4:39 AM
> > To: Terwilliger, Thomas Charles
> > Cc: PHENIX user mailing list
> > Subject: Re: [phenixbb] Strong anomalous signal but AutoSol fails
> >
> > One more thing, is there any way of knowing how many copies there should
> be in the ASU? Right now, I am doing it blindly as there is a huge range
> based on Matthew's coefficient calculation.
> >
> > On Sun, Aug 9, 2015 at 11:37 AM, mohamed noor <mohamed.noor34 at gmail.com>
> wrote:
> > Hi Tom
> >
> > When I used data in P 3(2)21, Hyss returned a CC of 0.2 for 10 sites and
> in P 3, I obtained a CC of 0.32 for 17 sites (out of 20 requested). From
> the documentation, CC 0.2 is considered possible and 0.3 is good. However,
> this does not translate into good models built by AutoBuild with the R
> factor being in the 45-50 % and low model CC (0.25). The FOM in AutoSol
> also seems to not reflect what it should be even if it is better than 0.4.
> Interestingly, when I input a dataset with low anomalous signal to 5 A (out
> of 3.5 A), I got an FOM of > 0.4 and Bayes-CC of 45 - but bad model.
> >
> > I also tried to run Phaser/MRage in all possible SG within both point
> groups without any luck.
> >
> > Mohamed
> >
> > On Fri, Aug 7, 2015 at 10:26 PM, Terwilliger, Thomas Charles <
> terwilliger at lanl.gov> wrote:
> > Hi Mohamed,
> >
> > This looks generally fine to me. One thing that stands out though is
> the line:
> >
> > 4.0- 3.5 0.79 3.23 0.24 4001 2577
> >
> > where the half-dataset CC is 0.24, much greater than for lower
> resolutions. This suggests that something is systematically wrong in this
> resolution shell (or elsewhere). It may be useful to cut the data at 4 A
> just for this reason. As Diana points out it may be useful for other
> reasons as well to cut the resolution at 4 A for finding the sites.
> >
> > I would suggest taking all of the datasets you have that are more or
> less isomorphous and putting them in a directory and then using
> scale_and_merge to put them all together (this is what it is designed
> for). It should be able to down-weight the ones that are most different
> from the average.
> >
> > You could also try the brute-force option in phenix.hyss to try very
> hard to find the sites.
> >
> > All the best,
> > Tom T
> >
> > From: mohamed noor [mohamed.noor34 at gmail.com]
> > Sent: Friday, August 07, 2015 10:55 AM
> > To: Terwilliger, Thomas Charles
> > Cc: PHENIX user mailing list
> > Subject: Re: [phenixbb] Strong anomalous signal but AutoSol fails
> >
> > Hi Tom
> >
> > Just to check that I am on the right track, I have attached below the
> output from phenix.anomalous_signal using a single XDS_ASCII.HKL file from
> a single crystal. It is one of a few datasets, but based on Xtriage seems
> to be the best one.
> >
> > The space group is P 3(1/2) 2 1. All datasets were collected at the peak
> wavelength (fluorescence scan). My initial feeling that there is a strong
> signal comes from CORRECT.LP. Aimless and Xtriage. IIRC, the resolution
> difference between the optimistic and pessimistic measurability is about
> 0.3-0.4 A.
> >
> >
> > Estimation of anomalous signal in a dataset
> >
> > Estimating B-value for anomalous substructure as 91.2 based on
> > overall B-value of 75.4 (Note: you can set this with
> b_value_anomalous=xx)
> >
> > Getting scaled data and half-datasets with scale_and_merge
> > Log file will be: scale.log
> >
> > Files for half_dataset CC:
> ['TEMP4/XDS_ASCII_SG150_1800frames_reprocFriedel_0_1.sca']
> > Files for half_dataset CC:
> ['TEMP4/XDS_ASCII_SG150_1800frames_reprocFriedel_0_1.sca']
> > Files for half_dataset CC:
> ['TEMP4/XDS_ASCII_SG150_1800frames_reprocFriedel_0_1.sca']
> > Scaled data are in: scaled_data.mtz
> > Half-dataset A is in: half_dataset_a.mtz
> > Half-dataset B is in: half_dataset_b.mtz
> > Using scaled data in analysis
> >
> >
> > Setting up estimator for CC*
> >
> > -------------------Summary of signal in this dataset
> ------------------------
> >
> > Shell
> > CCano Nrefl Nrefl
> > Resolution Esqr I/sigI half anom half
> > 47.8- 7.0 0.50 19.55 0.32 2389 2360
> > 7.0- 6.5 0.69 10.63 0.11 621 608
> > 6.5- 6.0 0.87 8.47 0.08 850 818
> > 6.0- 5.5 0.88 7.52 0.08 1189 1141
> > 5.5- 5.0 0.72 5.88 0.09 1709 1588
> > 5.0- 4.5 0.76 4.59 0.07 2521 2247
> > 4.5- 4.0 0.98 3.25 0.05 3881 3268
> > 4.0- 3.5 0.79 3.23 0.24 4001 2577
> > 3.5- 3.3 1.72 1.99 -0.02 3420 2118
> >
> > Cumulative
> >
> > ----------------------Data quality----------------- Best guess of
> expected
> > results of finding
> sites
> > ------ and
> phasing--------
> >
> > CCano Nrefl P(Substr)
> > Resolution Skew Esqr half anom CC* Signal +/- (%) FOM*
> +/-
> > 47.8- 7.0 0.02 0.47 0.32 2389 0.51 10.1 1.3 68 0.2
> 0.0
> > 47.8- 6.5 0.02 0.50 0.28 3010 0.48 10.7 1.7 72 0.2
> 0.1
> > 47.8- 6.0 0.02 0.57 0.24 3860 0.43 10.8 2.8 72 0.2
> 0.1
> > 47.8- 5.5 0.05 0.64 0.20 5049 0.44 12.6 2.4 78 0.2
> 0.1
> > 47.8- 5.0 0.01 0.66 0.17 6758 0.39 12.8 3.3 78 0.2
> 0.1
> > 47.8- 4.5 0.01 0.69 0.14 9279 0.36 13.4 3.4 82 0.2
> 0.1
> > 47.8- 4.0 0.00 0.77 0.12 13160 0.33 14.3 3.6 88 0.2
> 0.1
> > 47.8- 3.5 0.00 0.77 0.15 17161 0.39 18.4 3.8 98 0.3
> 0.1
> > 47.8- 3.3 0.00 0.88 0.14 20581 0.31 15.7 2.1 95 0.2
> 0.0
> >
> >
> > On Fri, Aug 7, 2015 at 2:30 PM, Terwilliger, Thomas Charles <
> terwilliger at lanl.gov> wrote:
> > Hi Mohamed,
> >
> > You might try running phenix.anomalous_signal on your data (may require
> finding your unmerged data and running phenix.scale_and_merge first). This
> will give you an idea if you should be able to solve your SAD dataset.
> >
> > See:
> http://www.phenix-online.org/version_docs/1.10pre-2124/reference/anomalous_signal.html
> >
> > All the best,
> > Tom T
> >
> >
> >
> >
> >
> > From: phenixbb-bounces at phenix-online.org [
> phenixbb-bounces at phenix-online.org] on behalf of mohamed noor [
> mohamed.noor34 at gmail.com]
> >
> > Sent: Thursday, August 06, 2015 3:16 PM
> >
> > To: PHENIX user mailing list
> >
> > Subject: [phenixbb] Strong anomalous signal but AutoSol fails
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > Dear developers
> >
> >
> >
> >
> > I have a low resolution anomalous dataset which Aimless suggests has an
> effective resolution to 3.3 A and anomalous signal to 3.5 A. However, SAD
> phasing with AutoSol is not successful with the final R factor around 50 %.
> >
> >
> >
> >
> >
> > I also have another dataset collected at a remote wavelength without
> anomalous signal to 3 A but they are not isomorphous (> 2 A difference in c
> axis).
> >
> >
> >
> >
> >
> >
> > The anomalous signal comes from the ligand heme c, which is bound
> covalently to the protein, so its occupancy should be 1. The protein is
> quite small with about 120 residues. Xtriage suggests an NCS of 6 to 20
> with most likely number to be 13.
> >
> >
> >
> >
> > Is there any reason why a reasonable solution cannot be found? There is
> no twinning.
> >
> >
> >
> >
> > I am using the latest nightly 1.10 pre2124.
> >
> >
> >
> >
> > Thanks.
> >
> >
> >
> >
> >
> >
> >
> >
> > _______________________________________________
> > phenixbb mailing list
> > phenixbb at phenix-online.org
> > http://phenix-online.org/mailman/listinfo/phenixbb
> > Unsubscribe: phenixbb-leave at phenix-online.org
>
>
> ________________________________
>
> UT Southwestern
>
>
> Medical Center
>
>
>
> The future of medicine, today.
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20150811/cb19f643/attachment-0001.htm>
More information about the phenixbb
mailing list