Hi Thomas, Chimera is not very memory efficient. I think it uses about 4 Kbytes per atom. So your all atom model, estimating 10 atoms per amino acid would be 50 million atoms and require 200 Gbytes of memory. Since you reduced to CA only that is about 5 million atoms which should only need about 20 Gbytes of memory. But that is just for representing the atoms, not the graphics, like ribbons for those. I think a ribbon is about 20x10 triangles per residue with about 30 bytes per triangle, so another 6 Kbytes for each of your CA, so 30 more Gbytes. Seems it might be possible with 64 Gbytes of memory, but not really surprising that it crashes. I wonder why you want to see ribbons for 5 million residues at once — I guess it might look cool, but that is more detail than most people will appreciate. PyMol or VMD are more memory efficient. Our next generation ChimeraX could do it pretty easily I think because it would only make one ribbon and draw it 8280 times — it is not released yet but I’ll send you a link if you want to try it. Tom
On Apr 15, 2016, at 3:21 PM, Klose, Thomas wrote:
Dear all,
I am trying to visualize a rather large structure (8280 chains with 620 amino acids each) as ribbons in Chimera, starting from a biological unit. I was able to reduce the required memory to something my Linux machine can handle (i7 with 64GB memory) by providing a file with just C-alpha atoms. Chimera successfully displays the copies as surfaces, and also starts to convert all the displayed chains into ribbons (chains as ribbon: finished), but then becomes unresponsive, runs with 100% CPU usage on one core for several hours and finally crashes. Is this a known problem/limitation and is there some workaround to display such large structures?
Thanks, Thomas _______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu <mailto:Chimera-users@cgl.ucsf.edu> Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users <http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users>