cartoon hides too many atoms of nucleic acids

[Elaine Meng]

Maybe the default behavior of cartoon should be to not hide any atoms of nucleic acids. The current behavior is to hide all but the bases, whicha are quite far from the cartoon.

[Conrad Huang]

Cartoon hides backbone atoms, so I assume that the hidden atoms are all defined as "backbone". Do I maintain a separate set of "atoms to hide" or do we change the definition of "backbone"?

[Elaine Meng]

Not sure if the question was directed to me.  I don’t have an opinion on how it’s implemented, I just think cartoon should not hide so many atoms.  Possibly all atoms should still be shown, but I’m not militant about that.
Elaine

[pett]

There are two kinds of "backbone" defined in the C++ layer, a "minimal" backbone and a "maximal" backbone. The minimal backbone is the smallest set of atoms needed to connect one residue to the next. For amino acids that would be C, CA, and N, and for nucleic acids O3', C3', C4', C5', O5', and P. The maximal set is all atoms that residues of that type always have in common, so for nucleic acids that includes the sugar. Neither set is really correct for ribbons. For amino acids you want to hide the maximal set (i.e. the O too). For nucleic acids hiding the maximal is bad because it hides the sugars, but hiding the minimal set will show OP1 and OP2, which is no good either.

Currently the Atom "is_backbone" functions works off the maximal backbone set (i.e. atoms not in the side chain).

[pett]

I'm thinking of smartening up the C++ layer to add a third set of atoms, the "ribbonable" set. Then Atom's "is_backbone" would take an argument: Residue.BB_MIN, BB_MAX, or BB_RIBBON and return the appropriate value. If that sounds like a good enough idea, let me know and I'll proceed.

[Conrad Huang]

Works for me.

On 11/4/2015 11:49 AM, Chimera2 wrote:

[pett]

Okay, with the revamped is_backbone functionality, Chimera 2 hides the appropriate nucleic acid backbone atoms. Now, the resulting ribbon is _way_ too narrow, but that's another issue!