Department of Institut Polytechnique de Paris

CS in biotechnologies and health

Up to two decades ago, Bioinformatics mainly developed with algorithmic tools to reason on the sequences and structure branches of molecular cell biology, ranging from DNA word problems, phylogenetic tree reconstruction, spatial structure abstractions of RNA and proteins, to prediction and optimisation of molecular interactions.

Since then, Bioinformatics is also developing on the great challenge of Systems Biology to gain system-level understanding of molecular interaction networks responsible for the high-level functions of the cell. This involves the development of boolean abstractions and model-checkers for molecular reaction networks, a theory of analog computability and complexity, the verification and synthesis of cyber-physical systems, up to the view of chemical reaction networks as a programming language for Synthetic Biology.

Besides, in the last years, artificial intelligence has revolutionized the biomedical domain, especially in the medical imaging field. The development of robust, automated and trustworthy methods for diagnosis, therapeutic or surgery planning has completely changed the clinical procedures in hospitals and laboratories. The final goal would be the creation of a digital twin of each patient combining several kinds of data at different scales, from micro-data, such as the DNA, to anatomical images, like MR or CT scans. The development of computational methods to integrate all these data will constitute an important effort towards personalized medicine and therefore towards clinical or pharmaceutical solutions better tailored to each patient.

These efforts, tightly coupled with research on data analytics, multimodal data fusion, stability and robustness of classifiers, geometric representations, and model-based image analysis, augmented reality and medical robotics, are currently participating altogether in a scientific revolution in Health, Medicine and the Pharma industry.

This vast application domain is in fact challenging computational methods, theories, concepts and tools, and drives a lot of research in computer science.