Researchers use Nvidia GPUs to simulate a living cell

Image: Snapshot of the 20-minute 3D simulation, showing yellow and purple ribosomes, red and blue degradasomes, and smaller spheres representing DNA polymers and proteins.

Scientists at the University of Illinois at Urbana-Champaign have created a 3D simulation of a living minimal cell, using Nvidia GPUs to simulate 7,000 genetic information processes in 20 minutes. According to the scientists, this project is the longest and most complex cell simulation carried out to date.

“Even a minimal cell requires two billion atoms,” said Zaida Luthey-Schulten, professor of chemistry and co-director of the university’s Center for Living Cell Physics, in a statement. “You can’t make a 3D model like this at a realistic human timescale without GPUs. »

The simulation reproduces the physical and chemical characteristics of a minimal cell at the particle scale. A minimal cell has only the minimal set of genes necessary for its survival, functioning and replication. Nevertheless, by creating a dynamic model that mimics the behavior of a living cell, scientists were able to gain insight into the fundamental processes that occur in living cells.

“What we discovered is that fundamental behaviors emerge from the simulated cell – not because we programmed them, but because the kinetic parameters and lipid mechanisms were correct in our model,” says Zaida Luthey- Schulten.

One software and three GPUs

Zaida Luthey-Schulten co-developed GPU-accelerated software called Lattice Microbes, which is now available on the Nvidia NGC software hub to run the simulation. The researchers used Lattice Microbes with three Nvidia Volta architecture GPUs to simulate a minimal version of a parasitic bacteria called mycoplasma.

The simulation showed that the cell devoted most of its energy to transporting molecules across the cell membrane.

This research, published in the journal Cell, demonstrates the potential of whole-cell simulations, which might ultimately help scientists predict how changes to real-world cells would affect their function.

This research also shows how well Nvidia can use its GPUs to perform simulations. While the tech giant is working with scientists to create cellular-level simulations, it also plans to build a digital twin of Earth to simulate and predict climate change. He makes a big bet with its Omniverse platform, which is expected to simulate everything from digital twins of warehouses, factories, and factories to robots, self-driving cars, and humans.

Source : ZDNet.com