Movies of dynamically cross-linked actin fibers

Homogeneous meshworks

When the fibers are turned over faster than they can form bundles, the result is a homogeneous meshwork. For this we can run a small system (side view and top view). We verify that our results are independent of system size by also considering a larger system (side view and top view). In all of these movies, colored fibers are the ones in bundles (a bundle is a connected cluster of fibers, where each connection is two links spaced one quarter of the fiber length apart).

Bundle-embedded meshworks

When we increase the turnover time, there are more bundles. For this a small system (side view and top view) is too small and finite size effects will give a large error. We will consider a larger system (side view and top view) for this case.

Stress relaxtion test

We first perform a stress relaxation test to show how the network relaxes over a timescale of order seconds. The homogeneous meshwork and bundle-embedded meshwork relax on a similar timescale. These movies show 5 seconds of simulation time on a domain of size 2 microns, which is smaller than the domain we actually collect data on.

Small amplitude oscillatory shear rheology

The rheological studies in our paper are performed with SAOS rheology. Here are examples for a homogeneous meshwork and bundle-embedded meshwork. These movies show 5 seconds of simulation time on a domain of size 2 microns, which is smaller than the domain we actually collect data on. The maximum strain here is 0.2 (for visualization purposes), which is significantly outside the linear regime.

With Brownian fluctuations

We now add Brownian fluctuations (semiflexible bending fluctuations and translational and rotational diffusion) and look at the same network of filaments. We set the turnover time to 2.5 s (about 60% of the bundling time) and the persistence length to lp=17 μm (corresponding to actin filaments). Here is a movie comparing deterministic filaments (at left) with Brownian filaments (at right) with frequency ω=1 Hz and full hydrodynamic interactions. While the Brownian dynamics are faster, there isn't much difference in the morphologies. Decreasing to lp=1.7 μm, we see a substantial difference in the morphologies, as the Brownian bending fluctuations become visible to the naked eye, while the deterministic filaments remain pretty straight. We give a side-by-side movie movie comparing Brownian fibers with lp=1.7 μm, but with intra-fiber hydro on the left and full hydro on the right. There isn't much difference visually between the two, other than perhaps less bundling with hydrodynamics (something we also saw deterministically).