nonstandard RNA residues cause cartoon (ribbon) problems

[Elaine Meng]

It looks like the nonstandard nucleic acid residues in a tRNA (for example, 1ehz) are not completely recognized. I say "not completely" because they are still treated as part of the polymer chain, but the cartoon (ribbon) is thin like peptide coil instead of having the nucleic-acid ribbon dimensions. The ribbon plane of the residue right after each thin segment is messed up, similar to the protein issue reported as #333. Looks like these nonstandard residues do have the standard ribose-phosphate backbone, with both the C5' path atom and C1' orientation atom, and are specified by command-line "nucleic"... they just need to be recognized correctly by the ribbon code, as far as I can tell. There are many modified bases in ribosomal and tRNA, so I don't expect this to be rare.

Reassign if I got the wrong person. :-)

[Conrad Huang]

"Fixed" in 8786937.

Added code to set non-amino-acid, non-nucleotide residue type based on the type of the previous residue in chain. So for drawing ribbons, unrecognized residues in proteins are treated as coils, while in nucleic acids they are treated as nucleotides. This simple heuristic will fail for chains with both amino acids and nucleotides under some circumstances but probability seems low.

[Elaine Meng]

I would argue that these ARE nucleotide residues, and rather common ones at that (and there are also some rather common modified amino acids that should be handled as amino acids).  Why not instead check for the presence of the guide and orientation atoms by name and simply use them, if present?  I think that would be much better.
Elaine

[Elaine Meng]

Oh sorry, you did say these nonstandard residues in RNA (and to a lesser extent DNA) would be treated as nucleotides. But I don't understand why you would treat nonstandard amino acids as coil if the SS info in the file says otherwise, unless they are lacking the needed backbone atoms. Most of them have the amino acid backbone.