Cartoon for C-alpha only structures

[Tom Goddard]

Show ribbon for C-alpha only structures, for example, 4bml.

[Tom Goddard]

Another example is 1grl where the tails of the protein chain are C-alpha only but most of the residues have side chains.

[Conrad Huang]

Structure.polymers is returning an empty vector<Chain::Residues> instance, so the ribbon code never sees anything it needs to render.

[Eric Pettersen]

polymers() on these structures have included the CA-only parts for some time now, and yet still no ribbon depiction -- reassigning to Conrad.

[Conrad Huang]

Changed ribbon code to use default rules for determining polymer connectedness (consider missing structure, consider chain id changes). This makes 4bml and 1grl look better because the CA traces become part of the ribbon (because we consider the missing structure bonds) but looks a little odd with "real" missing structures like 1www. Is this acceptable? Alternative is to add more flags to polymer() to be even smarter about chain breaks.

[goddard@…]

I don’t think ribbons should smoothly join chain breaks, like in 1www where residues 61-66 are missing, the ribbon should be broken their instead of smoothly going from residue 60 to 67.  This seems like a separate issue than ribbons for C-alpha structures.  I think missing residues should cause a ribbon break in both C-alpha and all atom structures.

[Conrad Huang]

Eric will add another argument to polymer() so that chain traces get connected as polymers but truly missing residues cause a break.

[Eric Pettersen]

Okay, the ribbon code now uses the new options to polymers() to include chain traces but break at other missing structure.

[Conrad Huang]

Eric's fix makes 2bml, 1grl and 1www all look correct.