NMRlipids databank: First draft of the manuscript

The first draft of the manuscript describing the NMRlipids databank is now available in this repository (main text and SI). It contains descriptions of the databank structure (as discussed in the Current status and structure post) and quality evaluation (as discussed in the Form factor quality evaluation update post). In addition, applications of the NMRlipids databank are demonstrated by analysing lipid flip-flops and water anisotropic diffusion from all the 714 simulations that are currently in the databank.

I have an ambitious plan to submit the manuscript by the end of this year due to the upcoming end of my funding and grant application deadlines. Therefore, I would appreciate receiving comments about the manuscript relatively soon. The results about lipid flip-flops and anisotropic water diffusion that are currently in the manuscript are summarized below.

The project is organized in three GitHub repositories. The Databank repository contains the core databank, the DataBankManuscript repository contains the analyses and results for the manuscript, and the DataBankManuscriptText repository contains the manuscript text. You can add your comments to issues of any of these repositories or to this blog post.

Detection rare phenomena using NMRlipids databank: Cholesterol flip-flops Flip-flops between bilayer leaflets were observed for cholesterol, DCHOL (18,19-di-nor-cholesterol), DOG (1,2-dioleoyl-sn-glycerol), and SDG (1-stearoyl-2-docosahexaenoyl-sn-glycerol), but not for other lipids in the NMRlipids databank. For cholesterol we observed 635 flip-flop events in 78 different simulations. Therefore, we analyzed the how cholesterol flip-flop rate depends on membrane properties. Cholesterol flip-flop rates and their averages over fixed ranges of x-axis values are plotted as a function of membrane thickness, lateral density and order in Fig. 1. The results reveal a non-linear correlation between decreasing cholesterol flip-flop rate and membrane packing.

Fig 1: A Illustration of cholesterol flip-flop. B-D Cholesterol flip-flops analyzed from the databank as a function of membrane thickness, area per lipid, and acyl chain order. Values from  simulations with non-zero flip-flop rates are shown with blue dots. Averages over fixed range of x-axis values are shown with black dots.

Extending the scope of MD simulations to new fields using the NMRlipids databank: Water diffusion anisotropy in membrane systems We first calculated the water permeability through membranes from all simulations in the NMRlipids databank. Observed permeabilities and their averages over fixed ranges of values at x-axis are shown in Figs. 2 B-E as a function of temperature, membrane thickness, area per lipid, and acyl chain order. the permeability of water decreases on average when membranes become more packed, i.e., with decreasing area per lipid and increasing thickness and acyl chain order (Figs. 2 C-E). To analyze how water diffusion anisotropy depends on membrane properties in a multi-lamellar lipid bilayer system, we calculated also the water diffusion parallel to the membrane surface from all simulations in the NMRlipids databank. The parallel diffusion coefficient of water decreases with reduced hydration and increases with the temperature, but dependencies on membrane area per lipid, thickness, or fraction of charged lipids were not observed in Figs. 2 and S5. Significant
increase in the diffusion anisotropy with membrane packing was observed.

Fig 2: A Water diffusion and permeability through membranes, and lateral diffusion along the membrane illustrated in a multilamellar stack of lipid bilayers. B-E Water permeation through membranes analyzed from the databank as a function of temperature, thickness, area per lipid, and acyl chain order. Values from simulations with non-zero permeation values are shown with blue dots. Averages over fixed range of x-axis values are shown with black dots. Insert in B) shows the Arrhenius plot of permeation (ln (P) vs. 1/T) that gives 17 ± 4 kT for the average
activation energy for water permeation through lipid bilayer. Inserts in C) and D) show the region where the dependence could be considered approximately linear. F Lateral diffusion of water as a function of hydration level. Experimental points for DMPC bilayers at 313 K at different hydration levels are shown. G-H Diffusion anistoropy of water as a function of thickness and area per lipid.