1. Answers to Questions

  • Question 1: How to identify templates of the target sequence
    Method: X_Ray diffraction.
    Resolution: 2.5 Å
    Chains: 2; A (\alpha chain) and B (\beta chain)
  • Question 2: PHENIX real space refine
    NOTE: Actual values depend on previous fitting and they may differ from the ones shown in this appendix.
    CC(mask) value: 0.778
    Residue with lower correlation value: 142 ARG (Misfit at the end of the chain)
    Correlation value: 0.186172531458
    Second residue with lower correlation value: 1 MET (Post-translation processing)
    Correlation value: 0.348504275208
    Correlation value of HEME group: 0.81328813 (To get this value, Select Residue Type (Other) and Show CC below (0.9 or 1.0)).
  • Question 3: PHENIX real space refine
    NOTE: Actual values depend on previous fitting and they may differ from the ones shown in this appendix.
    CC(mask) value has improved to 0.787.
    A 142 ARG correlation has improved to 0.4282267700789.
    HEME group correlation has not improved (0.81007005253).
  • Question 4: PHENIX real space refine
    NOTE: Actual values depend on previous fitting and they may differ from the ones shown in this appendix.
    CC(mask) value has improved to 0.805.
    A 142 ARG correlation has improved to 0.474205806292.
    HEME group correlation has also improved to 0.821341112742.
  • Question 5: CCP4 Refmac
    NOTE: Actual values depend on previous fitting and they may differ from the ones shown in this appendix.
    Tab. 1.2 Refmac results:

    Statistic

    Initial

    Final

    R factor

    0.3865

    0.3441

    Rms BondLength

    0.0142

    0.0165

    Rms BondAngle

    2.0081

    1.9696

    Rms ChirVolume

    0.1401

    0.0844

    The improvement is quite remarkable.

  • Question 6: CCP4 Refmac
    Tab. 1.3 Refmac results:

    Statistic

    Initial

    Final

    R factor

    0.3506

    0.3488

    Rms BondLength

    0.0137

    0.0150

    Rms BondAngle

    1.6843

    1.8655

    Rms ChirVolume

    0.0783

    0.0783

    Why: Because the starting values were already very good.

  • Question 7: CCP4 Refmac
    NOTE: Actual values depend on previous fitting and they may differ from the ones shown in this appendix.
    Tab. 1.4 :Refmac results. RSRAC stands for Real Space Refine after Coot.

    Refmac

    RSRAC

    Coot

    Statistic

    Initial

    Final

    Initial

    Final

    R factor

    0.4869

    0.4855

    0.4971

    0.4825

    Rms BondLength

    0.0176

    0.0212

    0.0136

    0.0193

    Rms BondAngle

    1.9186

    2.3549

    1.8053

    0.2382

    Rms ChirVolume

    0.1112

    0.1055

    0.1470

    0.1043
    Starting and final Rfactor values seem to be worse when we do not generate a mask volume around the atomic structure when Refmac runs both after Coot + Real Space Refine (compare with (Table 4) and after Coot (compare with Table 3). Whitout using a delimiting mask, the whole volume is considered, even if the structure fits to a small part of the volume. The use of mask is thus especially indicated when map and model show different sizes. However, no differences are detected when the volume generated by the extract unit cell protocol or normalized volume generated by Coot are used (data not shown).
  • Question 8: EMRinger
    NOTE: Actual values depend on previous fitting and they may differ from the ones shown in this appendix.
    EMRinger score after ChimeraX rigid fit: 3.86.
    EMRinger score after Coot: 2.37; Manual refinement depends on each user and in this case, for instance, we did not pay attention to rotamers.
    EMRinger score after PHENIX real space refine after Coot: 5.38
    EMRinger score after Refmac after Coot: 2.87; With Refmac parameters used, the improvement got with PHENIX real space refine after Coot is clearly higher than the improvement got with Refmac after Coot.
    EMRinger score after Refmac after PHENIX real space refine after Coot: 5.34; Refmac does not improve very much the result (because it was already good). With the EMRinger statistic, we can say that the modeling workflow is helpful to get a quite good model.
  • Question 9: Validation CryoEM
    NOTE: Actual values depend on previous fitting and they may differ from the ones shown in this appendix.
    Tab. 1.5 Validation statistics of human metHgb \alpha subunit model. RSRAC stands for Real Space Refine after Coot. Rama stands for Ramachandran.

    Statistic

    ChimeraX

    Coot

    PHENIX RSRAC

    Refmac after Coot

    Refmac after RSRAC

    5NI1

    CC(mask)

    0.569

    0.725

    0.787

    0.803

    0.801

    0.843

    EMRinger score

    3.86

    2.37

    5.38

    2.87

    5.34

    3.98

    RMS (Bonds)

    0.0188

    0.0183

    0.0090

    0.020

    0.0191

    0.0126

    RMS (Angles)

    2.41

    2.02

    1.30

    1.94

    1.84

    1.43

    Rama favored (%)

    97.14

    97.12

    95.68

    97.12

    95.68

    94.24

    Rama allowed (%)

    2.15

    2.88

    4.32

    2.88

    4.32

    5.76

    Rama outliers (%)

    0.71

    0.00

    0.00

    0.00

    0.00

    0.00

    Rotamer outliers (%)

    1.75

    24.78

    0.00

    23.01

    1.77

    0.88

    Clash score

    70.34

    26.24

    1.81

    21.73

    1.36

    2.26

    Overall score

    2.66

    3.12

    1.24

    3.02

    1.35

    1.39

    C beta deviations

    1

    7

    0

    2

    0

    0

    RMSD

    0.841

    0.447

    0.456

    0.414

    0.384

    0.0
  • Question 10: Validation CryoEM
    NOTE: Actual values depend on previous fitting and they may differ from the ones shown in this appendix.
    Tab. 1.6 Validation statistics of human metHgb :math:`beta`subunit model. RSRAC stands for Real Space Refine after Coot. Rama stands for Ramachandran.

    Statistic

    ChimeraX

    Coot

    PHENIX RSRAC

    Refmac after Coot

    Refmac after RSRAC

    5NI1

    CC(mask)

    0.524

    0.690

    0.776

    0.765

    0.767

    0.830

    E MRinger score

    1.13

    3.93

    4.76

    3.70

    5.32

    4.87

    RMS (Bonds)

    0.0313

    0.0169

    0.0078

    0.0191

    0.0183

    0.0117

    RMS (Angles)

    2.17

    1.97

    1.33

    1.95

    1.87

    1.40

    Rama favored (%)

    96.55

    97.92

    96.53

    97.22

    95.14

    95.83

    Rama allowed (%)

    3.45

    2.08

    3.47

    2.78

    4.86

    4.17

    Rama outliers (%)

    0.0

    0.00

    0.00

    0.00

    0.00

    0.00

    Rotamer outliers (%)

    1.68

    29.66

    0.85

    27.97

    5.93

    0.00

    Clash score

    75.93

    34.24

    3.89

    25.57

    2.16

    4.32

    Overall score

    2.75

    3.16

    1.40

    3.14

    1.92

    1.50

    C beta deviations

    0

    8

    0

    1

    0

    0

    RMSD

    0.935

    0.495

    0.470

    0.441

    0.494

    0.0
  • Question 11: Building the asymmetric unit
    Tab. 1.7 Validation statistics of human metHgb unit cell model. RSR stands for Real Space Refine. Rama stands for Ramachandran.

    Statistic

    ChimeraX rigid fit

    PHENIX RSRAC

    Refmac after RSRAC

    5NI1

    CC(mask)

    0.787

    0.808

    0.789

    0.840

    E MRinger score

    4.64

    4.58

    4.35

    4.11

    RMS (Bonds)

    0.0187

    0.0093

    0.0182

    0.0122

    RMS (Angles)

    1.860

    1.380

    1.840

    1.410

    Rama favored (%)

    95.41

    95.41

    94.70

    95.05

    Rama allowed (%)

    4.59

    4.59

    5.30

    4.95

    Rama outliers (%)

    0.00

    0.00

    0.00

    0.00

    Rotamer outliers (%)

    3.90

    0.00

    3.90

    0.43

    Clash score

    5.31

    3.54

    3.32

    3.53

    Overall score

    2.05

    1.46

    1.94

    1.49

    C beta deviations

    0

    0

    0

    0

    RMSD

    0.494

    0.509

    0.537

    0.00
  • Question 12: The whole macromolecule
    Tab. 1.8 Validation statistics of whole human metHgb model. RSR stands for Real Space Refine. Rama stands for Ramachandran.

    Statistic

    ChimeraX operate

    PHENIX RSR

    Refmac after RSR

    5NI1

    CC(mask)

    0.810

    0.803

    0.792

    0.842

    E MRinger score

    4.95

    4.70

    4.05

    4.18

    RMS (Bonds)

    0.0093

    0.0076

    0.0181

    0.0122

    RMS (Angles)

    1.390

    1.350

    1.860

    1.410

    Rama favored (%)

    95.41

    95.41

    95.41

    95.23

    Rama allowed (%)

    4.59

    4.59

    4.59

    4.77

    Rama outliers (%)

    0.00

    0.00

    0.00

    0.00

    Rotamer outliers (%)

    0.00

    0.00

    5.41

    0.43

    Clash score

    4.97

    3.21

    2.54

    3.53

    Overall score

    1.58

    1.43

    1.94

    1.48

    C beta deviations

    0

    0

    0

    0

    RMSD

    0.579

    0.642

    0.454

    0.00