Machine Learning at the (Nano)materials-biology Interface

Machine learning has a long history of success in the pharmaceutical sector, helping discover and optimize new drugs and predicting useful physicochemical properties like aqueous solubility. Materials science has embraced similar approaches and transferred useful technologies from the pharmaceutical sector. Although materials are more complex than small organic molecules, ML approaches have shown impressive results in predicting the properties of materials for application in diverse fields like 2D photonics, porous materials for energy and environmental applications, and in the development of biomaterials and regenerative medicine therapies. Here, we summarize some of the challenges in ML modelling of materials and highlight some exciting recent applications.