Molecular Spiders with Memory

Synthetic bio-molecular spiders with “legs” made of single-stranded segments of DNA can move on a surface which is also covered by single-stranded segments of DNA complementary to the leg DNA. In experimental realizations, when a leg detaches from a segment of the surface for the first time it alters that segment, and legs subsequently bound to these altered segments more weakly. Inspired by these experiments we investigate spiders moving along a one-dimensional substrate, whose legs leave newly visited sites at a slower rate than revisited sites. For a random walk (one-leg spider) the slowdown does not effect the long time behavior. For a bipedal spider, however, the slowdown generates an effective bias towards unvisited sites, and the spider behaves similarly to the excited walk. Surprisingly, the slowing down of the spider at new sites increases the diffusion coefficient and accelerates the growth of the number of visited sites.

💡 Research Summary

The paper “Molecular Spiders with Memory” investigates how synthetic DNA‑based molecular spiders move on a substrate that is itself coated with complementary single‑stranded DNA. In experimental realizations, when a spider’s leg binds to a fresh site for the first time, the substrate strand is chemically altered (e.g., by cleavage or conformational change). This alteration reduces the binding affinity of any subsequent leg that attempts to re‑bind to that same site, effectively creating a “memory” of visitation. The authors translate this experimentally observed memory into a one‑dimensional stochastic model and study two archetypal walkers: a single‑leg walker (ordinary random walk) and a bipedal walker (two‑leg spider).

For the single‑leg case, the slowdown at newly visited sites does not affect the asymptotic scaling. The transition matrix remains symmetric, the walk stays recurrent, and the mean‑square displacement grows as ⟨x²(t)⟩∼Dt with D identical to that of a standard random walk. Consequently, the number of distinct sites visited, V(t), follows the classic t½ law.

The situation changes dramatically for the bipedal spider. Because the two legs must remain within a fixed maximal separation, a leg that lands on a fresh site experiences a reduced detachment rate, while the other leg, typically on a previously visited (and thus weakened) site, detaches more quickly. This asymmetry generates an effective bias toward unvisited sites. Mathematically, the process maps onto an “excited walk” where the transition probabilities depend on the visitation history: p_new < p_old for detachment, yet the overall motion is biased forward because the slower detachment on new sites keeps one leg anchored while the other leg hops forward.

The authors combine exact calculations for small systems, mean‑field approximations, and extensive Monte‑Carlo simulations to quantify the consequences of this bias. They find that the diffusion coefficient D of the bipedal spider is a monotonic increasing function of the memory strength (the ratio of detachment rates between new and revisited sites). Counter‑intuitively, the slowdown at fresh sites does not hinder diffusion; instead, it forces the spider to continually explore new territory, thereby enlarging the effective step length and raising D.

Moreover, the growth law for the number of distinct sites visited accelerates. While a simple random walk yields V(t)∼t½, the memory‑induced bias gives V(t)∼t^α with α≈0.55–0.65 depending on the degree of slowdown. This faster coverage is directly linked to the effective forward drift: the spider spends more time on the frontier of its explored region, reducing the probability of returning to already visited sites.

The paper also discusses the broader implications for nanorobotics. By deliberately engineering a memory mechanism—e.g., designing substrate strands that become less sticky after the first binding—one can program a molecular walker to preferentially move into unexplored regions, achieving faster surface scanning or targeted delivery without external control fields. The authors outline extensions to higher dimensions, to spiders with more than two legs, and to more realistic kinetic schemes that include stochastic binding energies and cooperative effects.

In summary, the study reveals that a simple biochemical memory, which one might expect to impede motion, can in fact enhance both the diffusion coefficient and the rate of space exploration for multi‑leg molecular spiders. This insight opens a new design paradigm for autonomous DNA nanomachines, where controlled “forgetting” of visited sites becomes a powerful tool for steering stochastic motion at the nanoscale.