Covering Disruptive Technology Powering Business in The Digital Age

Written by: Rogelio Legaspi, Journalist, AOPG

Imagine a material more efficient than silicon in converting solar energy to electricity, with more than 90% efficiency. Or a material stronger than Kevlar that can withstand virtually anything, from aerospace applications to bulletproof vests. How about a new material that can rapidly absorb excess carbon dioxide in the atmosphere?

The discovery of such materials is what scientists are aiming for today, hoping to solve problems in various aspects. This may include applications in energy, environment or even in enterprise use. However, designing a material means configuring atoms so that a molecule can have the desired properties and it can be a complex task, especially for humans alone.

Knowing this, IBM recently unveiled the IBM Molecule Generation Experience (MolGX), a cloud-based, AI-driven molecular inverse-design platform, which automatically designs brand new molecular structures rapidly and diversely.

The MolGX is actually part of IBM’s accelerated discovery strategy, which aims to supercharge the scientific methodology using AI, hybrid cloud, automation and eventually quantum computing. Its goal is to speed up the discovery of new materials by 10 to 100 times.

Researchers of the AI platform, Toshiyuki Hama, Hsiang Han Hsu, Akihiro Kishimoto, and led by Seiji Takeda, had one question before they embarked on a voyage to find ways to automate the process of designing brand new molecular structures. “What if AI could help crack open the door to the infinite possibilities that chemical space has to offer”?

Takeda also said that the vastness of chemical space, which inhabits all possible design combinations of material structures, exceeds the capability of human experts to explore even a small fraction of it all. That is why they came up with the MolGX platform, hoping to provide AI capabilities for material science discoveries.

MolGX currently has two deployments – the basic web application with an easy-to-understand user interface with general users in mind and a professional version for chemistry professionals and industries, which includes additional functionalities.

Basically, MolGX works in three steps. First, a user will select the desired dataset for their experiment to train the AI, a process where the AI learns the numerical relations between a molecule’s structures (quantified structural features) and their properties.

Then, the user will tune the parameters of the AI model to accurately predict chemical properties. Finally, using AI, the user can specify their target chemical property and apply the trained model to generate new molecules.

In addition, the researchers are also aiming to let users automatically generate molecules from desired chemical properties, such as “solubility in water” and “heatability”. They call this inverse design, which aims to deliver tailored materials from the property targets of the product.

“Our platform is a perfect example of how AI and cutting-edge data processing technologies can put us on the fast track to the discovery of innovative materials that can make a significant impact on the environment and our society”, Takeda said in a blog.

You can try MolGX here for free.

(0)(0)