"A tangled problem," by William R. Taylor and Kuang Lin. Nature, 2 January 2003.
The authors, both mathematical biologists, suggest that as proteins lack the two qualities associated with tying shoelaces---dexterity and intent---perhaps there are better models than shoelace-tying for analyzing the properties of proteins. The authors suggest we experiment with a meter length of beads, covering the end beads with an adhesive putty, "pouring" the beads from hand to hand, and checking for knots after those two end beads stick together. After the admittedly unscientifically small sample of 20 tries, about one quarter of the "protein strands" are knotted. Given these small results and what is known about the structures of about 2,000 proteins, the authors pose, "why are there so few knots?" The key seems to lie in their "stickiness" (amino-acid residues). The authors discuss topological and covalent knots, loops and "pseudo-knots" and suggest that with the vast amount of forthcoming structural-genomics programs "we may soon have a large enough collection of folds to get a better idea of how frequent and important all forms of knots are to protein structure and function."
--- Annette Emerson