TITO is an AI model for molecular dynamics that "skips frames" in "molecular movies".
Some context: Molecular dynamics simulates how atoms move. It has important applications in many areas, including drug discovery.
Standard simulations move in tiny steps, around a femtosecond, because fast bond vibrations have to be resolved for the calculation to stay stable. But useful molecular events unfold over nanoseconds, microseconds or longer.
That leaves researchers doing billions of steps. It gets kinda computationally burdensome.
TITO, from researchers at Chalmers and the University of Gothenburg, learns from existing simulation data how atomic configurations change over longer time gaps, then samples those jumps directly.
The team says it is more than 10,000 times faster than conventional simulations.
In Science Advances, they showed it across on 12,530 small organic molecules and more than 1,000 short peptides.
“As far as we know, this is the first time this has been done in a way that works for many different molecules,” says Simon Olsson, research leader and associate professor in computer science and engineering at Chalmers and the University of Gothenburg
This is still far from simulating a real drug binding to a protein in a cell-like environment. But faster molecular simulations could help researchers test more candidate molecules computationally before deciding what to make or study in the lab. (6/7)