AIMS
Artificial Intelligence for Molecular Sciences — a web platform that puts interactive molecular dynamics, AI-assisted design, and data-science analytics in the hands of anyone curious enough to open a browser.
Connecting bits and atoms
The material world is made of atoms and molecules. In recent decades, data and computation — the in silico engine — have transformed molecular science. AIMS sits at the strategic intersection of artificial intelligence and molecular sciences, making that intersection accessible in the browser.
AIMS currently supports two types of projects. In both, AI can generate novel molecules and materials that can then be investigated with molecular modeling and simulation.
Molecular modeling
Interactive molecular dynamics lets users design and perform computational experiments to test their own hypotheses. Play with the simulation below to experience full-speed computation right in the browser.
With built-in visual analytics — including brushing and linking — users can explore quantitative structure-activity/property relationships (QSAR/QSPR) of molecule collections. A classic QSPR is the relationship between the boiling points and the carbon atom counts of alkanes — see this case study built in AIMS.
Drug discovery
Small molecules account for 90% of all drugs sold. The example below shows the process of screening an inhibitor for an HIV protease from a collection of small molecules using AIMS. Interactive molecular docking is under development. Filters implementing Lipinski's Rule of Five operate in the hyperspace of molecular properties, as shown below.