Abstract:
AI is transforming our approach to molecular engineering. Driven by the goal of accelerating drug development, our aim is to develop AI-driven molecular engineering methods which will enhance our approach to biomolecular discovery, such as drug discovery, drug repurposing, and chemical probe identification. This entails the development of generative and predictive tools that can learn from biochemical data, such as molecular structures, chemical reactions, and biomedical data. While AI can be applied to a range of molecular engineering tasks, one ideal area is de novo molecular design. De novo design is the concept of designing molecules with desired properties from scratch so as to minimize experimental screening, and is poised to allow scientists to more efficiently traverse chemical space in search of optimal molecules, and delegate error-prone decisions to computers via the use of predictive and generative models. In drug development, de novo design methods can aid medicinal chemists in the design and selection of drug candidates, with the added advantage that they can learn from datasets of billions of molecules in minutes and be constantly updated with new data. Deep molecular generative models are a particular approach to de novo design which uses deep neural networks to generate new molecules in silico, and works by proposing atom-by-atom (or fragment-by-fragment) modifications to an initial graph structure to generate compounds predicted to achieve a certain property profile. Such models can be applied to a range of therapeutic modalities.

In this talk, I will discuss the development of deep generative models for various molecular engineering tasks relevant to early-stage drug discovery. These include a model for synthesizability-constrained molecular generation, a reinforcement learning framework for molecular graph optimization, and recent applications from our group to the design of large modalities for targeted protein degradation. While the methods I discuss are focused on biological applications in this talk, they can be applied in a range of domains, including solid materials, and I will draw connections to applications outside drug discovery realm where relevant.

Bio:
Prof. Mercado is a tenure-track assistant professor in the Data Science and AI division at Chalmers University in Gothenburg, Sweden. She is the head of the AI Laboratory for Biomolecular Engineering (AIBE). Prof. Mercado is a California native. She did her BS degree in chemistry at CalTech. She obtained her PhD in chemical engineering at UC Berkeley. She did also complete a postdoc at MIT and an industrial postdoc in the AI team of AstraZeneca.