How big do computer simulations of membrane phase separation need to be?

What is it about?

We are interested in a phenomenon known as membrane phase separation, the tendency of some lipid membranes (including some found in our cells) to separate into regions rich in ordered lipids and cholesterol and areas with more fluid lipids; this process is analogous to the separation of oil and water. We wanted to understand what it is about certain lipids that make them separate, and how favorable that separation is. We use computer simulations to so, because the comparable experiments are often difficult or even impossible. However, there is a challenge: small simulations will produce different answers from big ones. To figure out how big the simulations need to be, we ran otherwise identical calculations with systems ranging from a few hundred lipids to more than 10,000, and concluded that 8-10 thousand lipids are generally needed to get "bulk-like" behavior.

Featured Image

Why is it important?

We are the first group to work out a way to figure out not just if phase separation is favorable but by how much; this is something that's not readily available from simulations or experiments, but it's valuable if we want to work out *why* things are happening.