Why do chain identifiers in PDB have no standard starting chain id type?

Question

At first I assumed that PDB chain identifiers are just uppercase letters (A, B, C,..) but found out there are protein chain ids that are numbers (0-9) and lowercase letters as well.

For example,

• 1NAL only has chain ids (1 - 4). It starts with chain '1' (number).
• 103L has chain identfier 'A'. It starts with chain 'A' (uppercase letter).
• 5afi's chain id's range from [a-zA-Z0-6] since it has 52 unique protein chains. It starts with chain 'a' (lowercase letter)

Can someone please explain? Or are id's just arbitrarily assigned?

Answer 1

Chain IDs are assigned by authors who submit the structure to the wwPDB. According to the PDB spec:

Non-blank alphanumerical character is used for chain identifier.

Usually, the chains are assigned uppercase letters. But since digits and lowercase letters are also allowed, some people will use it. And a good reason to allow more characters is to accommodate for structures that have more polymer chains than there are letters in English alphabet. With lowercase letters and digits the limit is 62 chains. Which is still not enough for some structures. (Many programs can handle two-character chain IDs, but this common extension has not been embraced by the wwPDB.)