Random state transitions of knots: a first step towards modeling unknotting by type II topoisomerases (Q876535)

Type II topoisomerases are enzymes that change the topology of DNA by performing strand passage. In particular they unknot knotted DNA efficiently. In this article the authors investigate a model of strand passages that is based on knots on the cubic lattice and random strand passages in various projections of these lattice knots. This model might help to understand in what way the unknotting strand passages of Type II topoisomerases resemble or differ from random strand passages. In more detail: Lattice knots of all prime knots with less or equal to 8 crossings are generated. Using the BFACF-algorithm and Markov chains various lattice representatives (using different lengths) of these knots are generated. This process results in some small improvements of the minimal known lattice knot representations reported in [\textit{E. J. Janse van Rensburg} and \textit{S. D. Promislow}, J. Knot Theory Ramifications 4, No. 1, 115--130 (1995; Zbl 0843.57006)]. For a random sample of such lattice knots each knot was projected into the plane using 50-100 random directions. For each such projection the Dowker-Thistlethwaite-code (DT-code) was computed [\textit{C. H. Dowker} and \textit{M. B. Thistlethwaite}, Topology Appl. 16, 19--31 (1983; Zbl 0516.57002)]. Each such projection and the corresponding DT-code was simplified be performing Reidemeister I and II whenever possible with high probability. Given such a simplified DT-code, a sign change in a single randomly selected entry will now represent a random strand passage for the given knot type. For the different knot types between \(30,000- 60,000\) such strand passages were performed. The obtained knot spectrum was evaluated by computing of the HOMFLY polynomial. This resulted in a transition matrix of the various knot types. A matrix entry represents the probability that a random strand passage will change a given knot type to a second given knot type. Furthermore for a given knot type (the \(6_1\) twist knot) several consecutive strand passages were performed to study unknotting behavior that can be compared directly to experimental results in the laboratory. As the authors admit the model used here is too simplistic, it does not take into account the physical properties of DNA and can in its current form not be used to simulate knot probabilities that agree with those experimentally observed.