Community Research and Development Information Service - CORDIS

ERC

ERIKLINDAHLERC2007 Result In Brief

Project reference: 209825
Funded under: FP7-IDEAS-ERC
Country: Sweden

New state-of-the-art techniques for molecular simulation

Biomolecular simulations are now used ubiquitously in general life sciences research. An incredible virtual microscope, we are still however a long way from using this tool to comprehend disease at the molecular level for the design of effective drugs.
New state-of-the-art techniques for molecular simulation
The EU-funded ERIKLINDAHLERC2007 (Multiscale and distributed computing algorithms for biomolecular simulation and efficient free energy calculations) project has advanced the state-of-the-art in biomolecular simulations. They successfully increased the performance of such tools by several orders of magnitude and enhanced prediction accuracy.

Researchers made remarkable progress starting with the development of GROMACS version 4.6. This software can handle different kinds of biomolecules (e.g. proteins, lipids and nucleic acids) along with their corresponding force fields via CHARMM27-specific features. They demonstrated about 40X higher performance and a tenfold improvement in parallelisation for simulations. Moreover, this freely available simulation code was faster by a factor of five than the current state-of-the-art.

Implementation of free energy calculations based on the "Bennett Acceptance Ratio" allowed on-the-fly analysis with improved accuracy even in distributed and cloud computing environments. Furthermore, project members successfully automated free energy input data creation while statistical analyses provided standard error estimates for all free energy calculations.

The icing on the cake, however, was the development of a novel framework enabling "parallel adaptive molecular dynamics" that combines distributed computing with molecular dynamics simulations. Available on the Copernicus website, this open-source tool automatically schedules thousands of tightly coupled simulations while distributed computing enhances sampling efficiency by several orders-of-magnitude.

Adaptive simulations for free energy calculations and automated generation of simulation parameters for small molecules have helped reduce the number of simulations required. As a result, Copernicus can determine solvation and binding free energies of a large series of compounds within a few hours. Moreover, this costs less than a euro per compound.

These new tools helped researchers conduct simulations of important biological systems such as gated ion channels and membrane proteins. Key discoveries were made leading to several publications in journals with significant implications for drug design. For instance, the study of ligand-gated ion channels provided novel insight into the dual-site allosteric modulation mechanism. This could prove useful in designing novel pairs of on/off anaesthetic drugs.

In all, the tools and findings from the ERIKLINDAHLERC2007 study could improve our knowledge about degenerative diseases like Creutzfeldt-Jakob and aid in the rational design of drugs. The application of these methodologies could also be extended beyond biomedicine to polymer science and nanotechnology.

Related information

Keywords

Biomolecular simulation, drug, GROMACS, free energy calculation, distributed computing, Copernicus
Record Number: 188500 / Last updated on: 2016-09-09
Domain: Biology, Medicine