Foraging kinetics of C. elegans

(a) Average experimental population reorientation rate (black line) in a rolling 2-minute window. Blue bins represent probability of observed reorientation rate. (b) Abrupt transitions were identified by performing two linear regressions on observed reorientation curves. Transition times were defined by the intersection of the regressions. (c) An example of an experimental reorientation curve with an abrupt reorientation transition. (d) An example of an experimental reorientation curve that lacked an abrupt reorientation transition. (e) Distribution of slope differences and transition times from regressions fit to the experimental data. Insets are individual examples of experimental cumulative reorientation curves. Number of worms (N) = 1631. Dashed line represents the median slope difference. All data curated from López-Cruz et al5.

Stochastic modeling of foraging kinetics of C. elegans

(a) Outline of the Gillespie algorithm. Step 1: two random numbers (r1 and r2) are drawn from a uniform distribution [0,1]. Step 2: The time of an event is randomly assigned based on the total rates (a0) and r1. Step 3: The event at t + τ is determined by the probability of the event occurring. (b) (Top) The parameters α and γ are assigned based on fitting a decay curve (red) to the observed average reorientation rate (blue). (Bottom) Average model population reorientation rate (black line) in a rolling 2-minute window. Red bins represent probability of observed reorientation rate. N = 1631, α = 1.54 min-1, γ = 0.07 min-1. (c) (Top) An example of a modeled abrupt reorientation transition. (Bottom) An example of a modeled reorientation curve that lacked an abrupt reorientation transition. (d) Distribution of slope differences and transition times from regressions fit to the experimental (blue) and modeled (red) data. Insets are individual examples of experimental and modeled cumulative reorientation curves. Dashed line represents the median experimental slope difference. (e) Examples of experimental (blue) and modeled (red) cumulative reorientation curves, with similar stochastic dynamics. Sudden changes in rate are indicated with arrows. The M-decay for each model is shown in orange. (f) Examples of modeled data when the reorientation rate is constant. Sudden changes in rate are indicated with arrows. α = 1.5.

Scaling of foraging kinetics of C. elegans

(a) The cumulative distributions of run lengths for different foraging dynamics. Single rate data were generated using a single reorientation rate (α), as in Figure 2f. The Lévy walk data were generated using a bounded power law distribution (Equation 8, μ=2), with the same boundaries as the experimental data.