Fri, 08/03/2018 - 14:09
On recent Design Puzzles, I have been making barrels from loops and parallel sheets. I have tried barrels with triangular cross-sections and square cross-sections. Below is about Puzzle 1546 where I tried square barrels. Sequence and Secondary Structure: In Puzzle 1546, I used 2 different starting sequences & secondary structure settings, each with 107 residues: 000000000111111111122222222223333333333444444444455555555556 123456789012345678901234567890123456789012345678901234567890 eeHeeHeeHeeHeeHeeHeeHeeHeeeeHeeeeHeeeeHeeeeHeeeeHeeeeHeeeeHe Je1a=vtLvtLvtLvtLvtLvtLvtLvtLtvtvLtvtvLtvtvLtvtvLtvtvLtvtvLtvtvLt Je1b=tvLtvLtvLtvLtvLtvLtvLtvLvtvtLvtvtLvtvtLvtvtLvtvtLvtvtLvtvtLv 00000000000000000000000000000000000000011111111 66666666677777777778888888888999999999900000000 12345678901234567890123456789012345678901234567 eeeHeeeeHeeeeHeeeeHeeeeHeeHeeHeeHeeHeeHeeHeeHee Je1a=vtvLtvtvLtvtvLtvtvLtvtvLvtLvtLvtLvtLvtLvtLvtLvt Je1b=tvtLvtvtLvtvtLvtvtLvtvtLtvLtvLtvLtvLtvLtvLtvLtv Each contains many sheets (e) of length 2 or 4 made of valine (v) and threonine (t) residues separated by helices (H) of length 1 made of leucine (L) residues. I chose these residues because I had read that the best sheet residues are tyrosine (y), phenylalanine (f), tryptophan (w), threonine (t), valine (v), or isoleucine (i) while the best helix residues are methionine (m), alanine (a), leucine (l), glutamate (e), or lysine (k). My goal was to have the hydrophilic threonines on the outside of the barrel and the hydrophobic valines on the inside of the barrel. I also wanted the leucines to end on the outside of the barrel, despite being hydrophobic. Surprisingly, in previous puzzles, leucines seemed to work better at these positions than other (generally hydrophilic) residues that I tried. I chose single-residue helices instead of multiple-residue loops between the sheets because alpha-helices turn about 100 degrees per residue. For triangle-shaped barrels, I wanted 120 degree turns between sheets. For square-shaped barrels, I wanted 90 degree turns between sheets. Having 100 degree turns seemed like a good compromise. Using less residues might also help the barrels fold properly in the lab, letting residues meant to hydrogen-bond with each other be closer in the sequence to each other. Once I set the desired sequence and secondary structure, I went into the selection interface, selected all 107 residues, and then did 5 2 5 2 to idealize the structure. This made the sheets flat and put bends at the helices. Next I unfroze everything. Then I used the recipe FreezeSelectSome1 (https://fold.it/portal/recipe/102617) to freeze only the backbones for the sheet residues. This left the helices free to move and left all sidechains free to move. Then I saved the structures as Je1a or Je1b.