Automated Acquisition and Mining of Biomedical Datasets
We are developing and applying computational technologies that could enhance rapid scientific discovery and enable large-scale deployment of medical interventions using a diverse array of technologies such as artificial intelligence, natural language processing, computer vision, audio processing and small unmanned aerial vehicles (drones). Examples of our projects in this area include using computer vision and deep learning to monitor clinical parameters in thermal images of newborns, using natural language processing to extract scientific insights from gene editing literature and using deep learning to infer unique sequence signatures for predicting viral outbreaks from DNA/RNA sequences. We also develop and apply computational methods for discovering fundamental processes in biology and emergence of complex processes for learning and evolution.
Programmable Biotechnologies & Gene Editing
We are interested in the precise (programmable) control of biological and chemical molecules. Our work in this area includes research on enhancing the efficiency and robustness of gene editing technologies in a range of applications including editing of human genes across genetically diverse individuals and development of anti-viral therapeutics directly targeting viral DNA or RNA. We are also developing computational methods for identifying small molecule interactions with gene editing technologies (CRISPR/Cas) and more broadly gene therapies.
Open and Equitable Innovation
We are leveraging various frameworks for enhancing large-scale scientific collaboration: i) Open Innovation approaches bringing together a global network of data scientists to solve computational challenges for malaria drug discovery under the DREAM of Malaria project; ii) Computational technologies that address technical and social barriers to collaborations such as data privacy, commercial interests and competition.