Use of Segments in PDB Files

Question

[khouli]

The PDB file format specifies that columns 73-76 of the ATOM and HETATM records are reserved for something called a "Segment identifier". PyMOL also makes references to "segments". In both cases, whatever a segment is it seems to go unused and is undocumented.

What was the purpose of segments? Is it currently used? Was it ever used? Did it go obsolete?

Answer 1

[GouthamAtla]

Some description is given here about their use.

Segment IDs are used like the more common chain IDs to distinguish different molecular segments that are not covalently bound to each other.

[João Rodrigues]

Adding to what Geek_y mentioned, segment IDs allow for more than 1 character, so they can be seen as an extension to the chain identifier (1-char only) for very large structures. We use it in our software quite often exactly because of this reason.

[khouli]

This doesn't seem like the original intent of the field. The PDB file format predates widespread usage of molecular dynamics and even today that's not the intended usage. On the other hand, maybe the original intent was simply as a generic "notepad".

[João Rodrigues]

You're probably right with the last sentence. However, although the field and the file-format wasn't probably meant for MD, CHARMM did form the basis of some molecular modelling programs - XPLOR-NIH, CNS, etc. - that are producing structures deposited as PDB files in the RCSB PDB. So the usage written in the link is still probably widespread and likely "correct". Also, as I mentioned above, as structures got bigger, it started being used for additional chain naming.

Regardless, the PDB format is being phased out in favour of mmCIF, which has none of these problems.