A crystallographer's guide to interpreting WHATIF output

A crystallographer's guide to interpreting WHATIF output

© 1997-1998 - Gerard J Kleywegt

With the new release of the PDB's Web-based data-deposition procedure AutoDep, macromolecular crystallographers receive, and have to reply to, the output of the validation tests of the program WhatIf of Vriend and Hooft (EMBL Heidelberg). The authors have written an explanation of the output.

This document aims to help practical crystallographers decide what to do if certain "errors" or "warnings" are issued:

Some can be ignored
Others are the responsibility of the PDB
Yet others can easily be fixed
Sometimes a detailed analysis of the local structure and (most importantly) the electron density will be required
In rare circumstances, the correctness of the entire model may be in question

So here goes for the list of possible error and warning messages as it exists today (4 July, 1997). Note that these are my personal opinions (they are all mine and, no, you can't have them, and you don't have to agree with them, so there !). Comments, suggestions, fan-mail and hate-mail can be E-mailed. Useful comments may be added to this page (let me know in case you specifically do not want this to happen with your comments when you mail them).

The format of this page is simple - each entry consists of:

one or more related error or warning messages as issued by WHATIF
my personal comment (in red)
any additional comments received from other people (in blue)
an example of WHATIF output for this type of error/warning

Click on an error or warning message or note in the list below to go to the comment and example:

Error: Missing unit cell information
Error: Non-triclinic spacegroup-symbol is missing
Error: C-unique setting for "P 21"
Warning: C-unique setting for "B 2"
Warning: Cell is non-standard
Warning: Triclinic cell with mixed acute and obtuse angles
Warning: Non-triclinic cell with acute angle
Warning: Unconventional orthorhombic cell
Warning: Class of conventional cell differs from CRYST1 cell
Warning: Unconventional cell on CRYST1
Error: Matthews Coefficient (Vm) very high
Warning: Matthews Coefficient (Vm) high
Error: Matthews Coefficient (Vm) too low
Error: Duplicate specification of SCALE cards
Error: Invalid SCALE matrix
Error: The SCALE matrix gives a left handed axis system
Error: Scale matrix represents wrong crystal class
Error: Negated value in scale matrix
Error: The SCALE matrix is nonstandard
Error: Value in first row of scale matrix mistyped
Error: Value in second row of scale matrix mistyped
Error: Inconsistent SCALE and CRYST1 cards
Note: Symmetry information inconsistent
Warning: New symmetry found
Warning: Conventional cell is pseudo-cell
Warning: Determinant of MTRIX is not exactly 1
Warning: MTRIX is not a pure rotation matrix
Warning: Missing atoms
Warning: C-terminal oxygen atoms missing
Error: C-terminal groups where they should not be
Warning: Duplicate atoms encountered
Warning: Unexpected atoms encountered
Warning: Rounded coordinates detected
Warning: Aspartic acid convention problem
Warning: Glutamic acid convention problem
Warning: Phenylalanine convention problem
Warning: Arginine nomenclature problem
Warning: Tyrosine convention problem
Warning: Leucine nomenclature problem
Warning: Valine nomenclature problem
Error: Isoleucine nomenclature problem
Error: Threonine nomenclature problem
Warning: Chirality deviations detected
Warning: High improper dihedral angle deviations
Error: Weights outside the 0.0 -- 1.0 range
Error: Chain names not unique
Error: Water clusters without contacts with non-water atoms
Warning: Water molecules need moving
Warning: Temperature factors given as "U", not as "B"
Warning: Average B-factor problem
Warning: Average B-factor out of normal range
Warning: More than 5\% of buried atoms has low B-factor
Warning: More than 2\% of buried atoms has low B-factor
Error: The B-factors of bonded atoms show signs of over-refinement
Error: Atoms too close to symmetry axes
Error: Abnormally short interatomic distances
Warning: Unusual bond lengths
Warning: Unusual bond angles
Warning: High bond length deviations
Warning: High bond angle deviations
Warning: Low bond length variability
Warning: Low bond angle variability
Warning: Possible cell scaling problem
Warning: Directionality in bond lengths
Warning: Unusual PRO puckering amplitudes
Warning: Unusual PRO puckering phases
Warning: Torsion angle evaluation shows unusual residues
Error: Ramachandran Z-score very low
Warning: Ramachandran Z-score low
Warning: Omega angles too tightly restrained
Warning: Omega angle restraints not strong enough
Error: chi-1/chi-2 angle correlation Z-score very low
Warning: chi-1/chi-2 angle correlation Z-score low
Warning: Backbone torsion angle evaluation shows unusual conformations
Error: Side chain planarity problems
Error: Connections to aromatic rings out of plane
Warning: Inside/Outside residue distribution unusual
Warning: Abnormal packing environment for some residues
Warning: Abnormal packing environment for sequential residues
Error: Abnormal average packing environment
Warning: Structural average packing environment a bit worrysome
Warning: Low packing Z-score for some residues
Warning: Abnormal packing Z-score for sequential residues
Error: Abnormal structural average packing Z-score
Warning: Structural average packing Z-score a bit worrysome
Warning: Backbone oxygen evaluation
Warning: Unusual rotamers
Warning: Unusual backbone conformations
Error: Backbone conformation Z-score very low
Warning: Backbone conformation Z-score low
Error: HIS, ASN, GLN side chain flips
Warning: Buried unsatisfied hydrogen bond donors
Warning: Buried unsatisfied hydrogen bond acceptors

Error: Missing unit cell information
Comment: in existing PDB entries, this is an omission of the PDB. This "error" is unlikely to occur in the context of AutoDep.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Symmetry related problems

Error: Missing unit cell information

No SCALE matrix is given in the PDB file. 
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Non-triclinic spacegroup-symbol is missing
Comment: in existing PDB entries, this is an omission of the PDB. This "error" is unlikely to occur in the context of AutoDep.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Non-triclinic spacegroup-symbol is missing

The CRYST1 card present in the PDB file gives a valid non-triclinic cell,
but the space group symbol is not given. 

The CRYST1 cell dimensions

    A    =  73.380  B   =  80.259  C    =  95.121
    Alpha=  90.000  Beta=  90.000  Gamma=  90.000
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: C-unique setting for "P 21"
Comment: if you have a good reason for using this, I wouldn't worry about this "error" !

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: C-unique setting for "P 21"

The space group symbol on the CRYST1 card of the PDB file is ``P 21''
(monoclinic), but the cell setting is C-unique. 

The CRYST1 cell dimensions

    A    =  52.230  B   =  70.290  C    =  95.470
    Alpha=  90.000  Beta=  90.000  Gamma=  99.590

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: C-unique setting for "B 2"
Comment: if you have a good reason for using this, I wouldn't worry about this "warning" !

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: C-unique setting for "B 2"

The space group symbol on the CRYST1 card of the PDB file is ``B 2''
(monoclinic), which is a non-standard c-unique setting for ``C 2''. 

The CRYST1 cell dimensions

    A    = 185.500  B   =  72.000  C    =  73.000
    Alpha=  90.000  Beta=  90.000  Gamma=  77.700

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Cell is non-standard
Comment: unlikely to occur with modern indexing software. Nothing to lose any sleep over, though.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Cell is non-standard

The unit cell in the CRYST1 card of the PDB file contains angles outside
the range 60-120 degrees. A conventional cell has angles within these limits. 

The CRYST1 cell dimensions

    A    =  41.690  B   =  34.620  C    =  32.130
    Alpha=  90.000  Beta= 127.600  Gamma=  90.000

Space group name: P 21 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Triclinic cell with mixed acute and obtuse angles
Comment: unlikely to occur with modern indexing software. Nothing to lose any sleep over, though.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Triclinic cell with mixed acute and obtuse angles

The crystallographic unit cell does not conform to the convention that a
triclinic cell should be specified as having either
three obtuse (type II) or three acute angles (type I). 

The CRYST1 cell dimensions

    A    =  32.100  B   =  46.200  C    =  25.400

    Alpha= 105.900  Beta= 113.300  Gamma=  69.700

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Non-triclinic cell with acute angle
Comment: unlikely to occur with modern indexing software. Nothing to lose any sleep over, though.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Non-triclinic cell with acute angle

The crystallographic unit cell does not conform to the convention that
all non-orthogonal angles in a non-triclinic cell should be obtuse. 

The CRYST1 cell dimensions

    A    = 185.500  B   =  72.000  C    =  73.000
    Alpha=  90.000  Beta=  90.000  Gamma=  77.700

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Unconventional orthorhombic cell
Comment: unlikely to occur with modern indexing software. Nothing to lose any sleep over, though.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Unconventional orthorhombic cell

The primitive P 2 2 2 or P 21 21 21 cell specified does not conform to the
convention that the axes should be given in order of increasing length. 

The CRYST1 cell dimensions

    A    = 116.700  B   = 103.900  C    =  48.340
    Alpha=  90.000  Beta=  90.000  Gamma=  90.000

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Class of conventional cell differs from CRYST1 cell
Comment: this is potentially a serious problem, since it could indicate a spacegroup error. In the worst case, go back to the original images and re-index, making sure you use the highest possible symmetry.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Class of conventional cell differs from CRYST1 cell

The crystal class of the conventional cell is different from the crystal
class of the cell given on the CRYST1 card. If the new class is supported
by the coordinates this is an indication of a wrong space group assignment. 

The CRYST1 cell dimensions

    A    =  85.070  B   =  77.980  C    =  88.860

    Alpha=  90.000  Beta= 118.630  Gamma=  90.000

Dimensions of a reduced cell

    A    =  77.980  B   =  85.070  C    =  88.819

    Alpha= 118.582  Beta=  90.000  Gamma=  90.000

Dimensions of the conventional cell

    A    =  85.070  B   = 155.991  C    =  77.980

    Alpha=  90.000  Beta=  90.000  Gamma=  89.969

Transformation to conventional cell

 |  1.000000  0.000000  0.000000|
 | -1.000000  0.000000 -2.000000|
 |  0.000000  1.000000  0.000000|

Crystal class of the cell: MONOCLINIC 

Crystal class of the conventional CELL: ORTHORHOMBIC 

Space group name: P 21 

Bravais type of conventional cell is: C 
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Unconventional cell on CRYST1
Comment: this usually involves swapping two axes so that the shortest one comes first. Unlikely to occur with modern indexing software. Nothing to lose any sleep over.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Unconventional cell on CRYST1

The derived ``conventional cell'' is different from the cell given on the
CRYST1 card. 

The CRYST1 cell dimensions

    A    =  55.300  B   =  59.400  C    =  42.500
    Alpha=  90.000  Beta=  99.100  Gamma=  90.000

Dimensions of a reduced cell

    A    =  42.500  B   =  55.300  C    =  59.400
    Alpha=  90.000  Beta=  90.000  Gamma=  99.100

Dimensions of the conventional cell

    A    =  42.500  B   =  59.400  C    =  55.300
    Alpha=  90.000  Beta=  99.100  Gamma=  90.000

Transformation to conventional cell

 |  0.000000  0.000000  1.000000|
 |  0.000000 -1.000000  0.000000|
 |  1.000000  0.000000  0.000000|

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Matthews Coefficient (Vm) very high
Warning: Matthews Coefficient (Vm) high

Error: Matthews Coefficient (Vm) too low
Comment: usually this has a trivial reason in existing PDB files. If you get this message for a new model, it may be due to an incorrect value of "Z" as indicated below. Fix on your deposition form.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Matthews Coefficient (Vm) very high

The Matthews coefficient [REF] is defined as the density of the protein
structure in cubic Angstroms per Dalton. Normal values are between 1.5
(tightly packed, little room for solvent) and 4.0 (loosely packed, much
space for solvent). Some very loosely packed structures can get values
a bit higher than that. 

Numbers this high are almost always caused by giving the wrong value for
Z on the CRYST1 card. 

Molecular weight of all polymer chains: 65310.648 
Volume of the Unit Cell V= 31006604.0 
Cell multiplicity: 2 
Matthews coefficient for observed atoms Vm= 237.378 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Duplicate specification of SCALE cards
Comment: unlikely to happen with AutoDep

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Duplicate specification of SCALE cards

There is more than one "SCALE" matrix present in the PDB file. 
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Invalid SCALE matrix
Comment: unlikely to happen with AutoDep

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Invalid SCALE matrix

The SCALE matrix contains elements larger than 0.5. This matrix represents
a very small cell. 

Possible cause: Probably one or more of the values is mistyped, or the SCALE
cards do not conform to the FORMAT given in the PDB specification. 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: The SCALE matrix gives a left handed axis system
Comment: if you manage to do this and still refine your structure you deserve an award !

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: The SCALE matrix gives a left handed axis system

The SCALE matrix given in the PDB file represents a left handed axis system
for the cell. This can cause weird problems
(such as negative cell volumes). 

Scale matrix as derived from SCALE

 |  0.007435  0.010515  0.010515|
 |  0.007435  0.000000  0.010515|
 | -0.011641  0.000000  0.000000|

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Scale matrix represents wrong crystal class
Comment: unlikely to happen with AutoDep
Comment from Gert Vriend (Gert.Vriend@EMBL-Heidelberg.de), EMBL, at 19980719:
Dear Gerard,
I was looking at your 'judgement' of the WHAT_CHECK checks. It must have been a lot of work puting that whole file on the web... However, there are some problems with the remarks you make. For example, you write about most CRYST/SCALE problems that they are unlikely to happen now that we have AUTODEP. However, see below Enrique's quote:
"Dear Gert,
Your test for SCALE and CRYST is superior to what we do. All that we do is to calculate a matrix based on the CRYST info and to compare this with the SCALE matrix. Differences greater than some criteria are returned to the depositor.
enrique"
That means that AUTODEP will not be able to deal with alternative settings etc., that can be 'encoded' in the SCALE matrix. Further, our checks do not only detect problems, but in roughly 50% of all cases give suggestions for improvements. And there are more things to say...
I think that, although it would cost you some time, you should be a bit more careful in the 'judgement' of WHAT_CHECK checks. I would say, try to be just as careful as we are in detecting the 'errors'.
Greetings
Gert Vriend
```
      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Scale matrix represents wrong crystal class

The crystal class for the unit cell on the CRYST1 card is different from the
crystal class derived from the SCALE matrix.
The cell on the CRYST1 card is compatible with the SPACE group. 

The CRYST1 cell dimensions

    A    =  58.850  B   =  58.850  C    = 259.220

    Alpha=  90.000  Beta=  90.000  Gamma=  90.000

Cell as derived from the SCALE matrix

    A    =  58.851  B   =  58.851  C    = 260.793

    Alpha=  90.000  Beta=  83.666  Gamma=  90.000

Space group name: P 43 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
   
```

Error: Negated value in scale matrix
Comment: unlikely to happen with AutoDep

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Negated value in scale matrix

One or more of the values of the scale matrix are wrong. 

Possible cause: Comparison with the matrix derived from the CRYST1 card
reveals that values have been inverted in sign. 

SCALE matrices (as given and from CRYST1) 

  0.019685  0.000000 -0.002977     0.019685  0.000000  0.002977
  0.000000  0.009355  0.000000     0.000000  0.009355  0.000000
  0.000000  0.000000  0.027861     0.000000  0.000000  0.027861

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: The SCALE matrix is nonstandard
Comment: unlikely to happen with AutoDep

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: The SCALE matrix is nonstandard

The SCALE matrix given in the PDB file represents a crystallographic unit
cell in which one or more angles are outside the range 60-120 degrees. The
CRYST1 card does not contain such values. 

Possible cause: Values in the SCALE cards are wrong, or the PDB file format
is not strictly adhered to. 

SCALE matrices (as given and from CRYST1) 

  0.007435  0.010515  0.010515     0.007435  0.000000  0.000000
  0.007435  0.000000  0.010515     0.000000  0.007435  0.000000
 -0.011641  0.000000  0.000000     0.000000  0.000000  0.011641
 

The CRYST1 cell dimensions

    A    = 134.500  B   = 134.500  C    =  85.900
    Alpha=  90.000  Beta=  90.000  Gamma=  90.000

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Value in first row of scale matrix mistyped

Error: Value in second row of scale matrix mistyped
Comment: unlikely to happen with AutoDep

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Value in first row of scale matrix mistyped

The SCALE matrix is incompatible with the CRYST1 cell. 

Possible cause: one or more of the values in the first row of the SCALE
matrix are wrong. 

Scale matrix as derived from SCALE

 |  0.023558  0.000000  0.002855|
 |  0.000000  0.013193  0.000000|
 |  0.000000  0.000000  0.023554|

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Inconsistent SCALE and CRYST1 cards
Comment: unlikely to happen with AutoDep

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Inconsistent SCALE and CRYST1 cards

The SCALE matrix and CRYST1 cards from the PDB file are inconsistent. 

Possible cause: A simple element by element comparison of the matrices can
not locate an obvious cause. 

The CRYST1 cell dimensions

    A    = 439.800  B   = 426.900  C    = 421.200
    Alpha=  90.000  Beta=  90.000  Gamma=  90.000

Cell as derived from the SCALE matrix

    A    = 441.301  B   = 427.306  C    = 421.932
    Alpha=  89.999  Beta=  89.952  Gamma=  90.001

SCALE matrices (as given and from CRYST1) 

  0.001454  0.000011  0.001738     0.002274  0.000000  0.000000
  0.000017  0.002340 -0.000029     0.000000  0.002342  0.000000
 -0.001819  0.000032  0.001519     0.000000  0.000000  0.002374
 
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Note: Symmetry information inconsistent
Comment: unlikely to happen with AutoDep

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Note: Symmetry information inconsistent

The SCALE and CRYST1 information disagree. 

Possible cause: The CRYST1 dimensions were rounded. 

The CRYST1 cell dimensions

    A    = 380.100  B   = 379.300  C    = 350.900
    Alpha=  90.000  Beta=  90.000  Gamma=  90.000

Cell as derived from the SCALE matrix

    A    = 380.041  B   = 379.290  C    = 350.904
    Alpha=  89.982  Beta=  90.013  Gamma=  89.997

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: New symmetry found
Comment: this is a potentially serious message, since it may point to a spacegroup error. Unfortunately, the WhatIf output has been reduced; it used to provide a more detailed analysis and useful information for investigating the problem.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: New symmetry found

In the conventional cell, independent molecules in the asymmetric unit
seemingly become symmetry relatives. This fact needs manual checking. 
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Conventional cell is pseudo-cell
Comment: good to know, but I'm baffled as to why this is issued as a "warning" instead of a "note" !

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Conventional cell is pseudo-cell

The extra symmetry that would be implied by the transition to the previously
mentioned conventional cell has not been observed. It must be concluded that
the crystal lattice has pseudo-symmetry. 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Determinant of MTRIX is not exactly 1
Comment: this means that the operator you used to generate molecule "B" from molecule "A", for instance, makes the B molecule larger or smaller than the A molecule ! Get a new operator with a determinant of 1.000000, for instance using X-PLOR, O or LSQMAN ! You probably want to do a further cycle of (rigid-body) refinement and/or averaging with this new operator.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Determinant of MTRIX is not exactly 1

The Determinant of a matrix given on a set of MTRIX cards differs from 1
by more than 0.002 

Matrix

 | -0.498700  0.866700  0.012480|
 | -0.865900 -0.498800  0.037820|
 |  0.039000  0.008050  0.992100|

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: MTRIX is not a pure rotation matrix
Comment: Calculate a new NCS operator, for instance using X-PLOR, O or LSQMAN ! You probably want to do a further cycle of (rigid-body) refinement and/or averaging with this new operator.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: MTRIX is not a pure rotation matrix

The matrix given on a set of MTRIX cards is not a pure rotation matrix. 

Matrix

 |  0.373740 -0.056207  0.940855|
 | -0.073400 -0.996800 -0.033512|
 |  0.909789 -0.054363 -0.376740|

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Missing atoms
Comment: you probably had a very good reason not to model these atoms; explain in a REMARK record.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Missing atoms

The atoms listed in the table below are missing from the entry. If many
atoms are missing, the other checks can become less sensitive. Some checks
(notably packing quality) will give worse results if the atoms are missing
than when they would have been modelled first. 

   1 ASN  (   2 )      CB
   1 ASN  (   2 )      CG
   1 ASN  (   2 )      OD1
   1 ASN  (   2 )      ND2
 305 LYS  ( 306 )      CB
 305 LYS  ( 306 )      CG
 305 LYS  ( 306 )      CD
 305 LYS  ( 306 )      CE
 305 LYS  ( 306 )      NZ

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: C-terminal oxygen atoms missing
Comment: they are probably called "OT1" and "OT2" or something. AutoDep ought to be able to handle this (IMHO).

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: C-terminal oxygen atoms missing

The C-atoms listed in the table below belong to a C-terminal residue in a
protein chain, but the C-terminal oxygen ("O2" or "OXT") that it should be
bound to was not found. 

 162 LYS  ( 162 )      C

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: C-terminal groups where they should not be
Comment: probably arises from chain breaks where density for a part of the structure is not visible, and the refinement program requires a "pseudo carboxy terminus". Remove the offending atom from your deposited PDB file.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: C-terminal groups where they should not be

The C-atoms listed in the table below belong to a non-C-terminal residue
in a protein chain, but nevertheless a terminating group was found for them. 

 121 PRO  ( 129 )      C

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Duplicate atoms encountered
Comment: check to find the cause of this and fix it

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Duplicate atoms encountered.

While reading the PDB file, at least one atom was encountered twice. This
might be caused by missing alternate-conformation flags, or sometimes by
naming the second C-terminal oxygen O instead of OXT or O2. The duplicates
have been discarded. 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Unexpected atoms encountered
Comment: check to find the cause of this and fix it

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Unexpected atoms encountered

While reading the PDB file, at least one atom was encountered that was not
expected in the residue. This might be caused by a naming convention problem.
The unexpected atoms have been discarded. 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Rounded coordinates detected
Comment: if your model is not a wire model, this is probably due to errors, e.g. peaks picked as water molecules but not refined positionally. Sort out and fix. (Note that, in contrast to the output below, decimal places suggest precision rather than accuracy.)

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Rounded coordinates detected

At least two atoms were detected with all three coordinates rounded to 1
decimal place. Since this is highly unlikely to occur accidentally, the
atoms listed in the table below were probably not refined. It could also
be that ALL atomic coordinates were rounded to 1 or 2 decimal places
(resulting in considerable loss of accuracy). 

 361 HOH  (HOH  )      597    28.000    28.000     0.000
 361 HOH  (HOH  )      598    26.000    26.000     0.000
 361 HOH  (HOH  )      612    26.800     0.000     0.000
 361 HOH  (HOH  )      614    15.500     0.000     0.000

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Aspartic acid convention problem
Warning: Glutamic acid convention problem
Warning: Phenylalanine convention problem
Warning: Arginine nomenclature problem
Warning: Tyrosine convention problem
Warning: Leucine nomenclature problem

Warning: Valine nomenclature problem
Comment: either a trivial problem, or done on purpose in the case of non-crystallographic symmetry to get similar torsion angles for corresponding residues in the various molecules. There should be a flag in AutoDep to fix (or not) these things automatically. If you want to fix them use ProCheck, MOLEMAN or MOLEMAN2.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Aspartic acid convention problem

The aspartic acid residues listed in the table below have their chi-2 not
between -90.0 and 90.0. 

  24 ASP  (  24 ) A
  46 ASP  (  46 ) A
  80 ASP  (  24 ) B
 102 ASP  (  46 ) B

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Isoleucine nomenclature problem

Error: Threonine nomenclature problem
Comment: unlikely to occur with modern refinement programs; if it does it is an error that should be fixed.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Isoleucine nomenclature problem

The isoleucine residues listed in the table below have their C-delta-1
attached to the wrong C-gamma. 

  99 ILE  (  99 ) A

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Chirality deviations detected
Comment: this means that tetrahedral carbons (not necessarily chiral, as suggested below !) have not been properly restrained in the refinement (or a bit of the structure has not been refined at all after some atoms were moved around during a rebuilding session). Fix your restraints, if necessary, and do an extra cycle of refinement.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Chirality deviations detected

The atoms listed in the table below have an improper dihedral value that is
deviating from expected values. 

Improper dihedrals are a measure of the chirality/planarity of the structure
at a specific atom. Values around -35 or +35 are expected for chiral atoms,
and values around 0 for planar atoms. Planar side chains are left out of the
calculations, these are better handled by the planarity checks. 

Three numbers are given for each atom in the table. The first is the Z-score
for the improper dihedral. The second number is the measured improper
dihedral. The third number is the expected value for this atom type. A final
column contains an extrawarning if the chirality for an atom is opposite to
the expected value. 

  40 ASN  (  40 )      CA     -4.6     22.6     33.8
  94 ALA  (  93 )      CA     -4.0     26.8     34.4
 107 MET  ( 106 )      CA     -4.6     23.8     34.1

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: High improper dihedral angle deviations
Comment: all or many of the impropers have been under-restrained. Adjust the restraints and do an extra cycle of refinement.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: High improper dihedral angle deviations

The RMS Z-score for the improper dihedrals in the structure is high. For well
refined structures this number is expected to be around 1.0. The fact that it
is higher than 1.5 in this structure could be an indication of overrefinement. 

 Improper dihedral Z-score : 1.775

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Weights outside the 0.0 -- 1.0 range
Comment: unlikely with modern refinement programs, but if it happens, fix the occupancies at more reasonable values, and do an extra cycle of refinement.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Weights outside the 0.0 -- 1.0 range

The atoms listed in the table below have their weight outside the 0.0--1.0
range. This problem is not hampering proper WHAT IF functioning, but it is
indicative of problems with the X-ray refinement. 

 108 ALA  ( 124 )      C   1.07
 108 ALA  ( 124 )      CB   1.07
 108 ALA  ( 124 )      O   1.07

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Chain names not unique
Comment: although there is only one example of this in the PDB (1TTT, and there it appears to be a shortcoming of the program), this could happen when you deposit. Fix.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Chain names not unique

The chain names listed below are given for more than one protein/DNA molecule in the structure. 

Chain identifier(s): D, E, F

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Water clusters without contacts with non-water atoms
Comment: check your maps (and B-factors) !

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Water clusters without contacts with non-water atoms

The water molecules listed in the table below are part of water molecule
clusters that do not make contacts with non-waters. These water molecules
are part of clusters that have a distance at least 1 Angstrom larger than
the sum of the Van der Waals radii to the nearest non-solvent atom. Because
these kinds of water clusters usually are not observed with X-ray diffraction
their presence could indicate a refinement artifact. The number in brackets
is the identifier of the water molecule in the input file. 

 804 HOH  (HOH  ) A    162
 805 HOH  (HOH  ) A    286

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Water molecules need moving
Comment: ignore if this was done for a good reason; otherwise, let the PDB fix it (there should be a flag in AutoDep to indicate this)

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Water molecules need moving

The water molecules listed in the table below were found to be significantly
closer to a symmetry related non-water molecule than to the ones given in the
coordinate file. For optimal viewing convenience revised coordinates for
these water molecules should be given. 

The number in brackets is the identifier of the water molecule in the input
file. Suggested coordinates are also given in the table. Please note that
alternative conformations for protein residues are not taken into account
for this calculation. 

  25 HOH  (HOH  )       45    2.846   24.342   15.050
  26 HOH  (HOH  )       56   11.843   13.097    5.059
  28 HOH  (HOH  )       51   19.230   32.397   17.970
  28 HOH  (HOH  )       54   19.389   29.689   16.110
  31 HOH  (HOH  )       25   14.785   30.653   19.485

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Temperature factors given as "U", not as "B"
Comment: either that, or occupancy and B-factor columns were exchanged, or you put something altogether different into the B-factor column. Fix or explain in a REMARK.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Temperature factors given as "U", not as "B"

The average temperature factor found is very low. Probably they are given
as "U" values, and not as "B" values. Values will be multiplied by
8-pi-squared for the analysis of B-factors. 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Average B-factor problem

Warning: Average B-factor out of normal range
Comment: there is no such thing as a normal average B-factor. There are several possible causes for high Bs. Check your Wilson plot to see if the average B makes sense (can only be done meaningfully if you have better than 3 A data, though). Can probably be ignored in most cases.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Average B-factor problem

The average B-factor for all buried protein atoms normally lies between
10--20. Values around 3--5 are expected for X-ray studies performed at
liquid nitrogen temperature. 

Because of the extreme value for the average B-factor, no further analysis
of the B-factors is performed. 

Average B-factor for buried atoms : 35.144 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: More than 5\% of buried atoms has low B-factor

Warning: More than 2\% of buried atoms has low B-factor
Comment: probably mostly a reflection of our inadequate methods for modelling B-factors at low resolution than anything else. Can probably be ignored most of the time.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: More than 5\% of buried atoms has low B-factor

For normal protein structures, no more than about 1 percent of the B factors
of buried atoms is below 5.0. The fact that this value is much higher in the
current structure could be a signal of overrefined B-factors, constaints to
too-low values, misuse of the B-factor field in the PDB file, or a scaling
problem. If the average B factor is low too, it is probably a low temperature
structure determination. 

Percentage of buried atoms with B less than 5 : 31.84 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: The B-factors of bonded atoms show signs of over-refinement
Comment: a lot of gobbledygook which is not to be taken seriously. At low resolution use grouped B-factors or tightly restrained bonded B-factors in order to reduce the number of degrees of freedom in your model. At high resolution, use Rfree to help you decide on a reasonable level for the bonded-B restraints.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: The B-factors of bonded atoms show signs of over-refinement

For each of the bond types in a protein a distribution was derived for the
difference between the square roots of the B-factors of the two atoms. All
bonds in the current protein were scored against these distributions. The
number given below is the RMS Z-score over the structure. For a structure
with completely restrained B-factors within residues, this value will be
around 0.35, for extremely high resolution structures refined with free
isotropic B-factors this number is expected to be near 1.0. Any value over
1.5 is sign of severe over-refinement of B-factors. 

RMS Z-score : 1.903 over 2783 bonds 
Average difference in B over a bond : 4.24 
RMS difference in B over a bond : 5.54 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Atoms too close to symmetry axes
Comment: check if atoms in special positions still are (i.e., they may have "drifted away") !

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Atoms too close to symmetry axes

The atoms listed in the table below are closer than 0.77 Angstrom to a
proper symmetry axis. This creates a bump between the atom and its symmetry
relative(s). It is likely that these represent refinement artefacts. 

2366 HOH  (HOH  )      169
2366 HOH  (HOH  )     1125
2366 HOH  (HOH  )     1166
2366 HOH  (HOH  )     1835
2368 HOH  (HOH  )       81
2368 HOH  (HOH  )       95
2368 HOH  (HOH  )      106
2368 HOH  (HOH  )     1175

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Abnormally short interatomic distances
Comment: check the density to make sure atoms are really there (don't take the "B1 + B2 > 80" too seriously, though). Also realise that according to this criterion there are 4,644 structures with errors in the PDB, and only 13 without ...

Comment from Kim Henrick (henrick@ebi.ac.uk), EBI, at 19971209: the program fails for PQR coordinates and for authors who submit coordinates in an unusual coordinate frame then the atoms first need to be re-orthogonalised, or WHATIF will be "unhappy" about the bump test.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Abnormally short interatomic distances

The pairs of atoms listed in the table below have an unusually short distance. 

The contact distances of all atom pairs have been checked. Two atoms are said
to `bump' if they are closer than the sum of their Van der Waals radii minus
0.40 Angstrom. For hydrogen bonded pairs a tolerance of 0.55 Angstrom is used.
The first number in the table tells you how much shorter that specific contact
is than the acceptable limit. The second distance is the distance between the
centers of the two atoms. 

The last text-item on each line represents the status of the atom pair. The
text `INTRA' means that the bump is between atoms that are explicitly listed
in the PDB file. `INTER' means it is an inter-symmetry bump. If the final
column contains the text 'HB', the bump criterium was relaxed because there
could be a hydrogen bond. Similarly relaxed criteria are used for 1--3 and
1--4 interactions (listed as 'B2' and 'B3', respectively). If the last column
is 'BF', the sum of the B-factors of the atoms is higher than 80, which makes
the appearance of the bump somewhat less severe because the atoms probably
aren't there anyway. 

Bumps between atoms for which the sum of their occupancies is lower than one
are not reported. In any case, each bump is listed in only one direction. 

  20 MET  (  20 )      SD   --  133 HOH  (HOH  )      175    0.315   2.535 INTRA HB
  45 ASN  (  45 )      ND2  --  110 GLY  ( 110 )      N      0.210   2.790 INTER
 122 VAL  ( 122 )      CB   --  134 HOH  (HOH  )      157    0.061   2.739 INTRA
  15 ASN  (  15 )      ND2  --   83 ILE  (  83 )      CD1    0.024   3.076 INTER

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Unusual bond lengths

Warning: Unusual bond angles
Comment: every structure has a few of these (remember that averages and standard deviations were calculated from small molecules, and it has yet to be shown that they are population statistics for biomacromolecules !). If there are many, you may have to increase the restraints and do an extra round of refinement.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Unusual bond lengths

The bond lengths listed in the table below were found to deviate more than
4 sigma from standard bond lengths (both standard values and sigma for amino
acid residues have been taken from Engh and Huber [REF], for DNA they were
taken from Parkinson et al [REF]). In the table below for each unusual bond
the bond length and the number of standard deviations it differs from the
normal value is given. 

Atom names starting with "<" belong to the previous residue in the chain.
If the second atom name is "--SS", the disulphide bridge has a deviating
length. 

  12   G  (  14 ) B  C8   N7    1.281  -4.0
  15   A  (  17 ) B  O4*  C1*   1.463   4.1
  17   A  (  21 ) B  O4*  C1*   1.463   4.1
  19   C  (  23 ) B  C1*  C2*   1.480  -4.8

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: High bond length deviations

Warning: High bond angle deviations
Comment: you may want to increase the restraints and do an extra round of refinement

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: High bond length deviations

Bond lengths were found to deviate more than normal from the mean standard
bond lengths (standard values for protein residues were taken from Engh and
Huber [REF], for DNA/RNA these values were taken from Parkinson et al [REF]).
The RMS Z-score given below is expected to be around 1.0 for a normally
restrained data set. The fact that it is higher than 1.5 in this structure
might indicate that the constraints used in the refinement were not strong
enough. This will also occur if a different bond length dictionary is used. 

Z-score for bond lengths: 1.764 
RMS-deviation in bond distances: 0.037 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Low bond length variability

Warning: Low bond angle variability
Comment: at low resolution high restraint weights for bond lengths and angles are necessary to reduce the number of degrees of freedom in a model. At high resolution, you could use Rfree to decide on a proper weight. Can safely be ignored in most cases. (Note: constraints can never be weak or strong, as suggested below; they are absolute and remove degrees of freedom from the model. Restraints, on the other hand, can be weak or strong.)

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Low bond length variability

Bond lengths were found to deviate less than normal from the mean Engh and
Huber [REF] and/or Parkinson et al [REF] standard bond lengths. The RMS
Z-score given below is expected to be around 1.0 for a normally restrained
data set. The fact that it is lower than 0.667 in this structure might
indicate that too-strong constraints have been used in the refinement.
This can only be a problem for high resolution X-ray structures. 

Z-score for bond lengths: 0.433 
RMS-deviation in bond distances: 0.010 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Possible cell scaling problem
Comment: this deserves investigation. With synchrotron data the phenomenon is not unusual since the wavelength may not have been known accurately. Also, some people use one set of cell constants (obtained from their very first crystal) for all subsequent complexes, mutants etc. This is poor practice ! Investigate !

Comment from Kim Henrick (henrick@ebi.ac.uk), EBI, at 19971209: I disagree with this. The problem doesn't come from an individaul wavelenght being unknown nor using the first or an average of the cells obtained in data reduction. This comes from wavelenght drift or the same beam having more than one wavelength perpendicular to each other so that data collected at one orientation is collected at a wavelength different to a following orientation. The effect is real only on sync data collection, however, no known program can or tries to correct for this observation. The WHATIF test is pointing out the fact one may have collected anisotropic wavelenght at different times in the same data collection, but the explanation is poor, and the correction is fairly meaningless as symmetry is broken to satisfy an isotropic distribution of bond lengths and angles. In some cases, the corrected cell dimensions are far off symmetry requirements and any attempt to use this cell to look at images and fit diffraction spots will be hopeless.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Possible cell scaling problem

Comparison of bond distances with Engh and Huber [REF] standard values for
protein residues and Parkinson et al [REF] values for DNA/RNA shows a
significant systematic deviation. It could be that the unit cell used in
refinement was not accurate enough. The deformation matrix given below gives
the deviations found: the three numbers on the diagonal represent the
relative corrections needed along the A, B and C cell axis. These values
are 1.000 in a normal case, but have significant deviations here (significant
at the 99.99\% confidence level) 

There are a number of different possible causes for the discrepancy. First
the cell used in refinement can be different from the best cell calculated.
Second, the value of lambda used for a synchrotron data set can be
miscalibrated. Finally, the discrepancy can be caused by a dataset that has
not been corrected for significant anisotropic thermal motion. 

Please note that the proposed scale matrix has NOT been constrained to obey
the space group symmetry. This is done on purpose. The distortions can give
you an indication of the accuracy of the determination. 

Unit Cell deformation matrix

 |  0.998473  0.000049  0.000017|
 |  0.000049  0.998579 -0.000039|
 |  0.000017 -0.000039  0.998875|

Proposed new scale matrix

 |  0.020398 -0.000001  0.004782|
 | -0.000001  0.013212  0.000001|
 |  0.000000  0.000000  0.011068|

With corresponding cell

    A    =  49.024  B   =  75.690  C    =  92.796
    Alpha=  90.006  Beta= 103.193  Gamma=  89.994

The CRYST1 cell dimensions

    A    =  49.100  B   =  75.800  C    =  92.900
    Alpha=  90.000  Beta= 103.200  Gamma=  90.000

Variance: 42.888 
(Under-)estimated Z-score: 4.827 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Directionality in bond lengths
Comment: see advice below

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Directionality in bond lengths

Comparison of bond distances with Engh and Huber [REF] standard values
for protein residues and Parkinson et al [REF] standard values for DNA/RNA
shows a significant systematic deviation. 

If this is not an XRAY structure this effect is hard to explain. Otherwise
you will have seen symmetry problems earlier. Please correct these and rerun
this check to see the possible implications on the cell axes. 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Unusual PRO puckering amplitudes
Comment: the example below is from one of my own low resolution structures. I stand duly corrected. Please forgive me for imparting such horror on this poor, defenseless proline residue !

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Unusual PRO puckering amplitudes

The proline residues listed in the table below have a puckering amplitude
that is outside of normal ranges. Puckering parameters were calculated by
the method of Cremer and Pople [REF]. Normal PRO rings have a puckering
amplitude Q between 0.20 and 0.45 Angstrom. If Q is lower than 0.20 Angstrom
for a PRO residue, this could indicate disorder between the two different
normal ring forms (with C-gamma below and above the ring, respectively).
If Q is higher than 0.45 Angstrom something could have gone wrong during
the refinement. 

   7 PRO  ( 244 ) A   0.46 HIGH

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Unusual PRO puckering phases
Comment: another example from the same structure. Let's do a nice "manual check of the data" ...

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Unusual PRO puckering phases

The proline residues listed in the table below have a puckering phase that
is not expected to occur in protein structures. Puckering parameters were
calculated by the method of Cremer and Pople [REF]. Normal PRO rings
approximately show a so-called envelope conformation with the C-gamma atom
above the plane of the ring (phi=+72 degrees), or a half-chair
conformation with C-gamma below and C-beta above the plane of the ring
(phi=-90 degrees). If phi deviates strongly from these values, this is
indicative of a very strange conformation for a PRO residue, and definitely
requires a manual check of the data. 

   1 PRO  ( 238 ) A  -58.1 half-chair C-beta/C-alpha (-54 degrees)
  10 PRO  ( 247 ) A  156.9 half-chair C-alpha/N (162 degrees)
  34 PRO  ( 271 ) A  -47.0 half-chair C-beta/C-alpha (-54 degrees)
  54 PRO  ( 291 ) A   25.6 half-chair N/C-delta (18 degrees)
 137 PRO  ( 374 ) A   51.0 half-chair C-delta/C-gamma (54 degrees)

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Torsion angle evaluation shows unusual residues
Comment: not very informative - are these Ramachandran outliers, or non-rotamers ? Check the density if you're worried.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Torsion angle evaluation shows unusual residues

The residues listed in the table below contain bad or abnormal torsion angles. 

These scores give an impression of how ``normal'' the torsion angles in
protein residues are. All torsion angles except omega are used for
calculating a `normality' score. Average values and standard deviations
were obtained from the residues in the WHAT IF database. These are used
to calculate Z-scores. A residue with a Z-score of below -2.0 is poor,
and a score of less than -3.0 is worrying. For such residues more than
one torsion angle is in a highly unlikely position. 

 229 TYR  (   9 ) B -3.0211
   8 TYR  (   9 ) A -3.0211
  81 TYR  (  82 ) A -2.6449
 302 TYR  (  82 ) B -2.6449
 426 PRO  ( 206 ) B -2.5862
 205 PRO  ( 206 ) A -2.5862
   9 PHE  (  10 ) A -2.0385
 230 PHE  (  10 ) B -2.0385

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Ramachandran Z-score very low

Warning: Ramachandran Z-score low
Comment: poor Ramachandran plots are always bad news ! If there are many outliers in a supposedly refined model, then seriously consider the possibility that parts of the structure are wrong !

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Ramachandran Z-score very low

The score expressing how well the backbone conformations of all residues
are corresponding to the known allowed areas in the Ramachandran plot is
very low. 

 Ramachandran Z-score : -7.127

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Omega angles too tightly restrained

Warning: Omega angle restraints not strong enough
Comment: the first warning can be safely ignored at low resolution; the second indicates that the restraint weights should be increased.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Omega angles too tightly restrained

The omega angles for trans-peptide bonds in a structure are expected to
give a gaussian distribution with the average around +178 degrees and a
standard deviation around 5.5 degrees. These expected values were obtained
from very accurately determined structures. Many protein structures are
too tightly constrained. This seems to be the case with the current
structure, as the observed standard deviation is below 4.0 degrees. 

Standard deviation of omega values : 1.426 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: chi-1/chi-2 angle correlation Z-score very low

Warning: chi-1/chi-2 angle correlation Z-score low
Comment: if the resolution is low, you are advised to go back to the graphics and use rotamers where they fit the density equally well as your own contraptions.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: chi-1/chi-2 angle correlation Z-score very low

The score expressing how well the chi-1/chi-2 angles of all residues are
corresponding to the populated areas in the database is very low. 

 chi-1/chi-2 correlation Z-score : -5.141

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Backbone torsion angle evaluation shows unusual conformations
Comment: check the density for Ramachandran outliers ! Also, some outliers may be caused by a peptide-flip error in the previous residue !

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Backbone torsion angle evaluation shows unusual conformations

The residues listed in the table below have abnormal backbone torsion angles. 

Residues with ``forbidden'' phi-psi combinations are listed, as well as
residues with unusual omega angles (deviating by more than 3 sigma from
the normal value). Please note that it is normal if about 5 percent of
the residues is listed here as having unusual phi-psi combinations. 

  25 SER  (  29 )   PRO omega poor
  26 PRO  (  30 )   Poor PRO-phi
  60 HIS  (  64 )   Poor phi/psi
  61 ALA  (  65 )   Poor phi/psi
  71 ASP  (  75 )   Poor phi/psi
 107 LYS  ( 111 )   Poor phi/psi
 124 GLY  ( 129 )   Poor phi/psi
 173 ASN  ( 178 )   Poor phi/psi
 190 PRO  ( 195 )   Poor PRO-phi
 196 PRO  ( 201 )   PRO omega poor
 197 PRO  ( 202 )   Poor phi/psi
 210 PRO  ( 215 )   Poor PRO-phi
 248 ASN  ( 253 )   Poor phi/psi

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Side chain planarity problems
Comment: you may want to increase the planarity restraint weights and do an extra round of refinement

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Side chain planarity problems

The side chains of the residues listed in the table below contain a planar
group that was found to deviate from planarity by more than 4.0 times the
expected value. For an amino acid residue that has a side chain with a
planar group, the RMS deviation of the atoms to a least squares plane was
determined. The number in the table is the number of standard deviations
this RMS value deviates from the expected value (0.0). 

  40 ASN  (  40 )     7.152
  72 ASP  (  72 )     6.482
  53 ASN  (  53 )     5.524
  89 ASP  (  89 )     5.386
 116 ASN  ( 116 )     4.235

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Connections to aromatic rings out of plane
Comment: somehow a restraint or two got lost (even though histidine is not an aromatic residue). Add appropriate restraints or increase their weights, and do an extra round of refinement.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Connections to aromatic rings out of plane

The atoms listed in the table below are connected to a planar aromatic
group in the sidechain of a protein residue but were found to deviate
from the least squares plane. 

For all atoms that are connected to an aromatic side chain in a protein
residue the distance of the atom to the least squares plane through the
aromatic system was determined. This value was divided by the standard
deviation from a distribution of similar values from a database of
small molecule structures. 

 300 HIS  ( 218 ) 2    CB   5.019
 146 HIS  ( 218 ) 1    CB   4.313

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Inside/Outside residue distribution unusual
Comment: hard to interpret in terms of individual residues. Could perhaps indicate possible register errors ???

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Inside/Outside residue distribution unusual

The distribution of residue types over the inside and the outside of the
protein is unusual. Normal values for the RMS Z-score below are between
0.84 and 1.16. The fact that it is higher in this structure could be caused
by transmembrane helices, by the fact that it is part of a multimeric
active unit, or by mistraced segments in the density. 

RMS inside/outside Z-score : 1.401 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Abnormal packing environment for some residues
Comment: difficult to interpret for individual residues, but if the overall score is very low this is almost invariably an indication of serious problems with the model. (If that is the case, one would expect several other global statistics also to be poor, e.g. the Ramachandran plot !)

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Abnormal packing environment for some residues

The residues listed in the table below have an unusual packing environment. 

The packing environment of the residues is compared with the average packing
environment for all residues of the same type in good PDB files. A low
packing score can indicate one of several things: Poor packing, misthreading
of the sequence through the density, crystal contacts, contacts with a
co-factor, or the residue is part of the active site. It is not uncommon
to see a few of these, but in any case this requires further inspection
of the residue. 

 103 GLU  ( 103 )    -6.18
 101 LYS  ( 101 )    -6.13
 100 LEU  ( 100 )    -5.59
 115 ASN  ( 115 )    -5.33

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Abnormal packing environment for sequential residues
Comment: merits scrutiny of the local density

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Abnormal packing environment for sequential residues

A stretch of at least three sequential residues with a questionable packing
environment was found. This could indicate that these residues are part of
a strange loop, but might also be an indication of misthreading. 

The table below lists the first and last residue in each stretch found, as
well as the average residue score of the series. 

 751 MET  ( 325 ) B   ---  754 GLY  ( 328 ) B    -5.57
 756 GLY  ( 332 ) B   ---  759 SER  ( 335 ) B    -4.68

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Abnormal average packing environment

Warning: Structural average packing environment a bit worrysome
Comment: worrisome ! This could indicate that part of the model is wrong.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Abnormal average packing environment

The average quality control value for the structure is very low. 

A molecule is certain to be incorrect if the average quality score is
below -3.0. Poorly refined molecules, very well energy minimized misthreaded
molecules and low homology models give values between -2.0 and -3.0. The
average quality of 200 highly refined Xray structures was -0.5+/-0.4 [REF]. 

Average for range 1 - 187 : -2.472 

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Low packing Z-score for some residues
Comment: many structures have a few of these. Check the density. If there are many outliers, worry !

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Low packing Z-score for some residues

The residues listed in the table below have an unusual packing environment
according to the 2nd generation quality check. The score listed in the table
is a packing normality Z-score: positive means better than average, negative
means worse than average. Only residues scoring less than -2.50 are listed
here. These are the "unusual" residues in the structure, so it will be
interesting to take a special look at them. 

 169 ALA  ( 181 )    -2.97
  14 LEU  (  26 )    -2.93
 396 LEU  ( 408 )    -2.62
  16 LEU  (  28 )    -2.59
 336 ARG  ( 348 )    -2.59
  28 LYS  (  40 )    -2.54

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Abnormal packing Z-score for sequential residues
Comment: check the density and tracing for such stretches

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Abnormal packing Z-score for sequential residues

A stretch of at least four sequential residues with a 2nd generation
packing Z-score below -1.75 was found. This could indicate that these
residues are part of a strange loop or that the residues in this range
are incomplete, but it might also be an indication of misthreading. 

The table below lists the first and last residue in each stretch found, as well as the average residue Z-score of the series. 

   9 LYS  (   9 ) O   ---   12 GLY  (  12 ) O    -2.40

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Abnormal structural average packing Z-score

Warning: Structural average packing Z-score a bit worrysome
Comment: worrisome ! This could indicate that part of the model is wrong.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Abnormal structural average packing Z-score

The quality control Z-score for the structure is very low. 

A molecule is certain to be incorrect if the Z-score is below -5.0. Poorly
refined molecules, very well energy minimized misthreaded molecules and low
homology models give values between -2.0 and -5.0. The average quality of
properly refined Xray structures is 0.0+/-1.0. 

 All   contacts    : Average = -1.409 Z-score =  -7.19

 BB-BB contacts    : Average = -0.762 Z-score =  -4.22

 BB-SC contacts    : Average = -1.344 Z-score =  -8.22

 SC-BB contacts    : Average = -0.546 Z-score =  -2.55

 SC-SC contacts    : Average = -1.005 Z-score =  -5.01

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Backbone oxygen evaluation
Comment: this is the pep-flip problem. Check the density.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Backbone oxygen evaluation

The residues listed in the table below have an unusual backbone oxygen
position. 

For each of the residues in the structure, a search was performed to find
5-residue stretches in the WHAT IF database with superposable C-alpha
coordinates, and some constraints on the neighboring backbone oxygens. 

In the following table the RMS distance between the backbone oxygen positions
of these matching structures in the database and the position of the backbone
oxygen atom in the current residue is given. If this number is larger than
1.5 a significant number of structures in the database show an alternative
position for the backbone oxygen. If the number is larger than 2.0 most
matching backbone fragments in the database have the peptide plane flipped.
A manual check needs to be performed to assess whether the experimental
data can support that alternative as well. The number in the last column
is the number of database hits (maximum 80) used in the calculation. It
is "normal" that some glycine residues show up in this list, but they
are still worth checking! 

 558 GLY  ( 558 )    2.88   14
 180 GLY  ( 180 )    1.92   10
 677 THR  ( 677 )    1.56   17

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Unusual rotamers
Comment: non-rotamers; check the density and see if a prefered rotamer perhaps fits the density equally well

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Unusual rotamers

The residues listed in the table below have a rotamer that is not seen very
often in the database of solved protein structures. This option determines
for every residue the position specific chi-1 rotamer distribution.
Thereafter it verified whether the actual residue in the molecule has the
most preferred rotamer or not. If the actual rotamer is the preferred one,
the score is 1.0. If the actual rotamer is unique, the score is 0.0. If
there are two preferred rotamers, with a population distribution of 3:2 and
your rotamer sits in the lesser populated rotamer, the score will be 0.66.
No value will be given if insufficient hits are found in the database. 

It is not necessarily an error if a few residues have rotamer values below
0.3, but careful inspection of all residues with these low values could be
worth it. 

 117 SER  ( 117 )     0.39
 108 GLU  ( 108 )     0.39

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Unusual backbone conformations
Comment: check density, Ramachandran, pep-flip.

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Unusual backbone conformations

For the residues listed in the table below, the backbone formed by itself
and two neighboring residues on either side is in a conformation that is
not seen very often in the database of solved protein structures. The number
given in the table is the number of similar backbone conformations in the
database with the same amino acid in the center. 

For this check, backbone conformations are compared with database structures
using C-alpha superpositions with some restraints on the backbone oxygen
positions. 

A residue mentioned in the table can be part of a strange loop, or there
might be something wrong with it or its directly surrounding residues. There
are a few of these in every protein, but in any case it is worth looking at! 

 101 LYS  ( 101 )    0
 126 ASP  ( 126 )    2

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: Backbone conformation Z-score very low

Warning: Backbone conformation Z-score low
Comment: could occur for small peptides (as below); if not, could be a reason to worry ?

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: Backbone conformation Z-score very low

A comparison of the backbone conformation with database proteins shows that
the backbone fold in this structure is very unusual. 

 Backbone conformation Z-score : -17.463

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Error: HIS, ASN, GLN side chain flips
Comment: inspect to see if you agree

      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Error: HIS, ASN, GLN side chain flips

Listed here are Histidine, Asparagine or Glutamine residues for which the
orientation determined from hydrogen bonding analysis are different from
the assignment given in the input. Either they could form energetically
more favorable hydrogen bonds if the terminal group was rotated by 180
degrees, or there is no assignment in the input file (atom type 'A') but
an assignment could be made. If a residue is marked ``flexible'' the flipped
conformation is only slightly better than the non-flipped conformation. 

 115 ASN  ( 115 )

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----

Warning: Buried unsatisfied hydrogen bond donors
Warning: Buried unsatisfied hydrogen bond acceptors
Comment: ignore for low-resolution models (no or very few water molecules)
Comment from Kim Henrick (henrick@ebi.ac.uk), EBI, at 19971209: If a residue is burried then, according to WHATIF, it must have all H-bonds satisfied - if it is exposed it doesn't - but waters are not used in both steps. In one case, a Tyr has all H-bonds satisfied, some to water, some not. It is exposed with no water included in the accesible surface calculation, but is buried if water is used. WHATIF says for this example that the side chain is buried and that un-satisfied H-bonds are present - now one can't use water to bury the side chain and then exclude the water in looking for H-bonds. Rob Hooft told me that it shouldn't have been considered buried as its relative ASA is 0.7% - therefore it is exposed. This is a very confusing test.
```
      
 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
Warning: Buried unsatisfied hydrogen bond donors

The buried hydrogen bond donors listed in the table below have a hydrogen
atom that is not involved in a hydrogen bond in the optimized hydrogen bond
network. 

Hydrogen bond donors that are buried inside the protein normally use all of
their hydrogens to form hydrogen bonds within the protein. If there are any
non hydrogen bonded buried hydrogen bond donors in the structure they will
be listed here. In very good structures the number of listed atoms will tend
to zero. 

   7 ASN  (  91 ) A    ND2
  17 ASN  ( 101 ) A    N
  54 THR  ( 138 ) A    N
  81 GLN  ( 165 ) A    NE2
  96 ALA  ( 180 ) A    N
[...]
 701 GLN  ( 422 ) B    N
 720 ASN  ( 441 ) B    ND2

 ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE ----- EXAMPLE -----
   
```

Latest update at 19 July, 1998.