Has an app that allows people to transform their hand-drawn structures into 3D augmented reality molecules caused a stir among chemists on Twitter.

Users say MolAR is not only educational, but also fun. “The app offers a quick visualization of the 3D structure of a molecule in augmented reality (AR), so that users can easily understand the molecular structures,” says team leader Todd Martinez from Stanford University in the USA. “This offers a far richer learning experience. Instead of static 2D structures on one side, the molecules appear in 3D on your desk. ‘

“My personal favorite is the ability to display molecular vibrations in 3D,” says the doctoral student and App user Dmytro Kravchuk from the University of Iowa in the USA. The tool also offers other highlights such as the “Draw” and “Recognize” functions, he adds. “Still at an early stage of development, these functions allow users to draw simple organic structures either on screen or on paper, scan the drawings and visualize the drawn molecules in 3D space.”

The app, programmed by Martinez’s colleague Sukolsak Sakshuwong, was first released for iOS in March and has now been updated to visualize hand-drawn structures and molecular vibrations. The idea originally came from Martinez’s colleague Umberto Raucci, after he had developed an Amazon Alexa app that enables voice control for quantum chemical calculations.

Martinez emphasizes that hand-drawn structure recognition is currently limited to hydrocarbons. “We are in the process of extending this to heteroatoms,” he says. The feature is based on a machine learning program called ChemPix, which was developed by Martinez’s colleague Hayley Weir. It converts the drawings into the computer-readable Smiles format (simplified molecular input-line input system). “The recognition software was trained on a large data set of structures with wavy lines, uneven bond angles, and a variety of background and paper types,” explains Martinez. However, since only simple hydrocarbon backbones were used for training, writes CH3 or Me for methyl groups can still confuse the neural network.

An image of a phone screen showing grapes lying on a market stall table.  There is a 3D chemical structure of resveratrol on the grapes

Kravchuk notes that while the software can easily recognize poorly drawn structures of non-cyclic hydrocarbons, recognizing cyclic structures is still a challenge if the handwriting is not perfect. “On the other hand, the detection of physical objects is well done and works by recognizing a water bottle as an H. out2O and coffee as a caffeine molecule. “

Jet-Sing Lee, postdoctoral fellow at Tokyo University in Japan, tried MolAr and says that he could perhaps continue to use it for teaching purposes – but also during coffee breaks with colleagues. “Chemistry-based apps that are well-designed and fun are rarely seen,” he comments. “It’s very cool that there is a pretty wide range of organic structures available, from simple molecules to larger carbon allotropes and biomolecules. An additional feature that should be included is the visualization of metal-containing structures, such as organometallic compounds. That would increase the appeal to a wider range of chemists. ‘

Martinez says new features are being developed that take into account stereochemistry and organometallic compounds, and an Android version of the app is also in the works. “We plan to include dynamic movements, chemical reactions and the calculation of quantum mechanical properties. This will enable the visualization of molecular orbitals and the calculation of spectra. ‘


Please enter your comment!
Please enter your name here