Environmental Sustainability, Human Health and Machine Learning

WELCOME TO MY WEBSITE

The Lancet Commission on Pollution and Health reports that the economic costs of environmental pollution globally are over $4.6 trillion annually, an 18% equivalence of the USA’s annual GDP. Environmental contamination costs governments significant budgets through research to develop strategies and technologies to clean contaminated sites and treat related illnesses in populations. There is a need to create easy, quick, and less costly tools to inform us about the transformation and fate of contaminants in the environment. There has been tremendous progress in scientific advances toward analytical techniques that can generate vast amounts of data from environmental sample analyses which has spurred rapidly growing interest in the interaction of artificial intelligence and chemistry. Molecular chemoinformatics and simulation are powerful tools that can inform about the properties and complex structures. In addition to the advances in the field of analytical chemistry, the advancement in GPU computing has resulted in dramatic improvements in the computational speed of complex chemoinformatic tasks, making it possible to study larger,  and more complex systems. In the last decade, the field of deep learning has taken the world by storm.  Graph neural networks, autoencoders, and transformers are increasingly applied to study molecular chemistry and de novo generation. 

My research explores the intersection between environmental sustainability, molecular chemistry, and artificial intelligence to develop tools that can aid in informing us about the properties, fate, and transformation of contaminants in the environment. Specifically, I am interested in developing tools that can effectively use information from the molecular characterization of environmental contaminants and constituents to model their transformation and fate and predict their physicochemical and toxicological properties. This research will be focused on two research goals; research goal one will aim at the collection of data through field and laboratory experiments, and online data curation while research goal two will aim at using the collected data to produce computer-based models and solutions. This knowledge can help us predict the fate of contaminants in the environment, and design mitigation strategies and treatment processes to prevent potential risks and inform decision-making.