Skip to content

Processing TPT data (5Y78 & 5Y79)

Jump to bottom
keitaroyam edited this page Jan 26, 2018 · 7 revisions

The following describes how triose-phosphate/phosphate translocator (TPT) datasets can be processed using KAMO (documentation in Japanese / English).

References

  • Original paper
    • Lee et al. (2017) "Structure of the triose-phosphate/phosphate translocator reveals the basis of substrate specificity." Nature Plants doi: 10.1038/s41477-017-0022-8
  • Structures described here

Raw data

  • Available in SBGrid Data Bank under CC0
  • Collected on BL32XU, SPring-8 using 1 Å wavelength and 10-20×2-10 μm2 beam
  • P21212, a~107, b~165, c~41 Å
  • Pi-bound (5Y78)
    • 98 small-wedge (10°/dataset) datasets with MX225-HS CCD detector
    • One 30° dataset with MX225-HS CCD detector
    • 624 small-wedge (3-10°/dataset) datasets with EIGER X 9M detector
  • 3-PGA-bound (5Y79)
    • 332 small-wedge (3-10°/dataset) datasets with EIGER X 9M detector
  • 1°/frame (CCD) or 0.1°/frame (EIGER), shutterless

How data were processed in the original paper

Processing of individual datasets

GUI command 'kamo' was used by following parameters: unit_cell="41 106 165 90 90 90" space_group="p222" auto_frame_exclude_spot_based=true for 5Y78 and unit_cell="106 165 41 90 90 90" space_group="P21212" for 5Y79. XDS versions used were May 1, 2016 BUILT=20160617 (5Y78) and Oct 15, 2015 (5Y79). 598 (5Y78) and 250 (5Y79) datasets were processed and prepared for merging.

Merging

Hierarchical clustering was performed using CC at 2.5 Å (5Y78) or BLEND (5Y79) and second-largest clusters were used. Outlier rejection was performed based on frame-CC, and em.b+bfactor (5Y78) or em.b (5Y79). For 5Y78, kamo.multi_merge was executed at 1.9 Å resolution and the selected result was re-scaled at 2.1 Å manually.

Following is the the tables in XSCALE.LP that was used for structure analyses.

5Y78 (319 datasets merged):

 SUBSET OF INTENSITY DATA WITH SIGNAL/NOISE >= -3.0 AS FUNCTION OF RESOLUTION
 RESOLUTION     NUMBER OF REFLECTIONS    COMPLETENESS R-FACTOR  R-FACTOR COMPARED I/SIGMA   R-meas  CC(1/2)  Anomal  SigAno   Nano
   LIMIT     OBSERVED  UNIQUE  POSSIBLE     OF DATA   observed  expected                                      Corr

     6.26      167898    1858      1868       99.5%      10.4%     11.1%   167898   58.55     10.5%    99.9*     9    0.953    1285
     4.44      278098    3031      3079       98.4%      19.8%     16.4%   278098   42.92     19.9%    99.9*    11*   1.063    2478
     3.63      326933    3872      3903       99.2%      30.1%     26.7%   326933   29.80     30.3%    99.7*     5    1.011    3307
     3.15      421298    4543      4543      100.0%      62.7%     65.9%   421298   15.35     63.0%    99.5*     5    0.846    3976
     2.82      488967    5162      5162      100.0%     110.8%    122.3%   488967    9.44    111.4%    98.9*     0    0.760    4596
     2.57      537092    5606      5606      100.0%     187.0%    211.0%   537091    6.13    187.9%    97.5*     0    0.734    5035
     2.38      568717    6128      6128      100.0%     358.9%    416.0%   568717    3.51    360.8%    93.2*     2    0.710    5555
     2.23      553581    6570      6570      100.0%     774.0%    886.1%   553581    2.03    778.6%    82.7*     1    0.662    6003
     2.10      599576    6933      6933      100.0%     -99.9%    -99.9%   599576  -99.00    -99.9%    68.6*     0    0.602    6363
    total     3942160   43703     43792       99.8%     130.3%    148.1%  3942159   12.59    131.1%    99.5*     2    0.764   38598

Note that R=-99.9% and I/SIGMA=-99.00 seem to happen in XDS when R value became negative whatever I/SIGMA value actually is. The actual stats can be calculated for example phenix.merging_statistics (cannot use the same binning though).

5Y79 (199 datasets merged):

 SUBSET OF INTENSITY DATA WITH SIGNAL/NOISE >= -3.0 AS FUNCTION OF RESOLUTION
 RESOLUTION     NUMBER OF REFLECTIONS    COMPLETENESS R-FACTOR  R-FACTOR COMPARED I/SIGMA   R-meas  CC(1/2)  Anomal  SigAno   Nano
   LIMIT     OBSERVED  UNIQUE  POSSIBLE     OF DATA   observed  expected                                      Corr

     6.59       63653    1599      1607       99.5%       7.3%      7.5%    63653   55.77      7.4%   100.0*     7    0.938    1082
     4.66      105801    2688      2688      100.0%      10.7%     10.5%   105801   41.13     10.8%    99.9*    11*   0.947    2173
     3.81      145537    3375      3389       99.6%      14.6%     13.1%   145537   38.25     14.8%    97.6*     9    1.005    2861
     3.30      166496    3994      3994      100.0%      29.4%     29.3%   166496   20.78     29.8%    99.8*     3    0.865    3469
     2.95      183575    4484      4484      100.0%      56.9%     60.8%   183575   11.71     57.6%    99.3*     3    0.787    3967
     2.69      215992    5032      5032      100.0%      92.1%    100.8%   215992    7.95     93.2%    98.7*     3    0.756    4492
     2.49      232070    5366      5367      100.0%     160.9%    180.1%   232070    4.86    162.8%    96.7*     5    0.732    4849
     2.33      211534    5697      5698      100.0%     273.0%    313.9%   211534    2.71    276.8%    90.1*     0    0.686    5179
     2.20      169164    5947      5948      100.0%     462.9%    530.8%   169164    1.57    471.4%    74.9*     1    0.643    5434
    total     1493822   38182     38207       99.9%      52.8%     57.5%  1493822   14.54     53.5%    99.4*     4    0.778   33506
Clone this wiki locally