Some aspects of chirality have been mentioned already in the section with administrative problems. Here I will elaborate a bit more on this topic.

There are a few genuine d-amino acids in the PDB. For example gramicidin a. But I think that in most cases this is an error. I cannot give you much advice for what to do when you run into this problem. As a user of the structure, I guess you should read if the d-amino acids are real or not, and if they are erroneous, look for another structure. And if you are a crystallographer or an NMR spectroscopist, get help from a senior, start fresh using a homology model of your structure, or Email me for help.

A more hidden problem is the chirality on the Cβ of isoleucine. And I present one example here of a really bad case.


1ACP is a very old NMR structure. But as it was used to test the use of molecular dynamics in NMR structure refinement, I was a bit surprised to find many chirality problems. In the note below you find the abstract of the article describing 1ACP.

Supplemental material

In 1ACP I found these two isoleucines next to each other. They are residues 10 and 11 in the second MODEL.

Figure 35. Isoleucines 10 and 11 in MODEL 2 of 1ACP.

It was just funny to see that Isoleucine 10 is OK (in terms of chirality) while residue 11 has the wrong chirality on both the Cα and the Cβ.

WHAT_CHECK tells us about 1ACP:

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 extra warning if the chirality for an atom is opposite to the expected value.
   7 VAL   (   7-)  A  1   CA   -43.4   -30.19    33.23 Wrong hand
   8 LYS   (   8-)  A  1   C     -7.2   -10.86     0.10
  10 ILE   (  10-)  A  1   CA   -42.0   -31.81    33.24 Wrong hand
  11 ILE   (  11-)  A  1   CA   -42.8   -33.10    33.24 Wrong hand
  29 VAL   (  29-)  A  1   CA   -45.3   -32.85    33.23 Wrong hand
  46 LEU   (  46-)  A  1   CA   -45.4   -35.25    34.17 Wrong hand
  85 LYS   (   8-)  A  2   CA   -42.1   -34.73    33.92 Wrong hand
  88 ILE   (  11-)  A  2   CA   -43.0   -33.43    33.24 Wrong hand
  88 ILE   (  11-)  A  2   CB   -50.2   -34.93    32.30 Wrong hand
  96 GLN   (  19-)  A  2   CA   -35.8   -34.86    33.97 Wrong hand
 104 SER   (  27-)  A  2   CA   -37.8   -36.03    34.32 Wrong hand
 107 GLU   (  30-)  A  2   C    -11.1   -16.11    -0.05
 124 GLU   (  47-)  A  2   CA   -41.9   -35.08    33.98 Wrong hand
 149 ILE   (  72-)  A  2   CB   -49.5   -34.00    32.30 Wrong hand

The whole structure 1ACP is not very good. take a look at the Ramachandran plot of the second model (that is the one from which I took the two isoleucines):

Figure 36. Ramachandran plot for the second model of 1ACP.

Later, the same molecule was solved many times by X-ray. One example is 1L0H. Below, you can see how different a 12 year younger X-ray structure can be...

Figure 37. The very poor 1ACP structure.

Figure 38. The much better 1L0H structure.