IBM’s newly launched Molecule Generation Experience (MolGX) artificial intelligence platform is a cloud-based molecular reverse design platform that can quickly and diversify the automatic design of new molecular structures.
MolGX is part of IBM’s accelerated exploration strategy, which aims to use artificial intelligence, hybrid cloud and automation, and even quantum computing technology to strengthen scientific methods and increase the exploration speed of new materials by 10 to 100 times.
Also Read: IBM Rumored To Consider Selling Watson Health Unit
IBM mentioned that the discovery of new materials is the driving force for the development of industrial products. From tissue engineering to drug discovery, and the discovery of sustainable materials, there are still thousands of compounds waiting to be discovered by humans. But the chemistry of material structure space is too huge. It is beyond the ability of human scientists to explore. So with the help of artificial intelligence, people can accelerate the pace of exploration.
The purpose of IBM’s development of MolGX is to use new molecular generative artificial intelligence model technology to accelerate molecular reverse engineering. Researchers mentioned that it can draw non-existent landscape maps or portraits. It establishes the molecular structure, according to the required chemical properties, such as water solubility and heatability. So the generation model can generate molecular structures with specific chemical properties according to the parameters.
Three Key Features Of MolGX
MolGX mainly provides three features. First, MolGX uses algorithm-based coding and structure generation procedures. Unlike other data-driven methods, MolGX does not require pre-training with large data sets, nor does it require task-related pre-training. Furthermore, MolGX’s space and structure generation program is completely interpretable for human chemical scientists. So it is possible to perform chemical analysis of molecular structure simply. The third feature of MolGX can provide a hierarchical data structure and a clear user interface, allowing users to operate flexibly and intuitively.
The properties of materials depend on the shape of molecules. But molecular shapes are almost infinite. Therefore, people used to spend a lot of time developing new materials. The growth of artificial intelligence and hybrid cloud technology has promoted the development of technologies such as MolGX. So it can develop materials in a more efficient and economical way. MolGX provides a new way of material development, which not only reduces development time, but also enables innovative development of new materials without being restricted by fixed concepts.