Massive MuMMI Martinis

A huge consortium effort from the Livermore Lab on RAS dynamics, featuring a staggering 120,000 independent Martini simulations, has now been published: Ingolfsson et al., PNAS, 2022.

Impressive work based on the MuMMI (Multiscale Machine-learned Modeling Infrastructure) workflow.

MuMMI.jpg

Small molecules

adts202100391-gra-0001-m.jpgA good start of the new year: Our paper on how to parameterise small molecules with Martini 3 is now published !

You can read about best practices on chosing bead types, handling constraints, and selecting virtual sites for planar compounds, as well as find a large table with best bead types for a wide selection of chemical fragments and numerous validated topologies for small compounds.

Alessandri et al., Adv. Theory & Simul. 2100391. https://doi.org/10.1002/adts.202100391

GPCR-arrestin complex formation

toc2.pngMartini-based simulations capture the spontaneous binding of beta-arrestin2 to beta2-adrenergic receptor in a similar pose as the crystallographically resolved structure of rhodopsin/arresin-1. Cool !

See the paper from Pluhackova et al.: https://doi.org/10.3389/fcell.2021.807913

The complex formation is found to be dependent on specific lipids, as well as on phosphorylation state. Parameters for phosphorylated serine and threonine in the framework of Martini 2/2P are provided in the SI.

Cell-scale membrane envelope

ci1c01050_0012.gif

 

From the lab of Tajkhorshid: a protocol for setting up simulations of cell-scale membrane envelopes, using Martini.

Paving the way for whole cell simulations !

See Vermaas et al., JCIM, online, for details.

Deep eutectic solvents

sc1c06521_0010.gifDeep eutectic solvents (DESs) are a more environmentally friendly, cost-effective, and recyclable alternative for ionic liquids. Since the number of possible deep eutectic solvents is very large, there are needs for effective methods to predict the physicochemical nature of possible new deep eutectic solvents that are not met by the currently available models.

To meet this challenge, we parameterized and validated a first set of DESs compatible with Martini 3, and showed its application in simulating liquid-liquid extraction processes.

For details, check Vainikka et al., ACS Sust. Chem. Engin., online.

Martini 3 versus force matching

An interesting study from the Wilson lab shows how Martini 3 can reproduce complex chromonic self-assembly, whereas specifically parameterized CG models obtained with force matching (FM) cannot. To see the details and the underlying reasons for the failure of FM, see: Yu & Wilson, J. Mol. Liq., online.