I am trying to parse the SCOP parseable files, specifically dir.des.scop.txt ver 1.75. But, I have been facing problems with the PDB residue identifiers in the file. This is a tab limited file and the PDB residue and chain identifier is the 6th column which. An example of this identifier is A:1-28,A:77-837 for domain d1kk8a2. Hence forward I'll represent that as domain_id (pdb_id chain&range). This particular example is straightforward, which means the domain d1kk8a2 consists of residues 1-28 and 77-837 of chain A of the corresponding PDB entry 1kk8. But some of them are unintuitive and like d3ckra1 (3ckr A:-2-385), what does the negative entry for the start mean? And this example d2p83b1 (2p83 B:61P-385), there is a P in the range (61P-385).
If you go to pdb website and search for the corresponding entries, then go to the sequence tab, you will see the alignment of the PDB chain and the corresponding SCOP domains. The ranges shown there correspond to the entries in SCOP, but it is difficult to make sense of it. Can someone please explain or provide some pointers. Thank you.