Computational Approaches for Systems and Synthetic Biology
Systems biology seeks to “reverse engineer” how cells work. It often uses the “omic” technologies (for example, genomics, proteomics or metabolomics) to paint a holistic molecular portrait of a cell. For example, to understand how a fungus synthesizes a certain molecule, we can apply sequencing technologies to its entire complement of DNA and RNA. By subjecting cells to different stimuli, environmental conditions or genetic mutations, it is possible to collect diverse portraits of cellular behavior. The goal is to assemble the puzzle pieces into a coherent model to see how the biological system works.
Dr. Michael Hallett, Canada Research Chair in Bioinformatics Algorithms, uses tools from machine learning to self-assemble such models. He presents a computer with molecular profiles from a vast array of different stimuli, conditions and mutations, and the computer learns to predict the cell’s behavior. Hallett and his research team can then conceptually “lift the hood” of an accurate model and use it to figure out how the cells themselves are wired.
This ability to predict a biological system is also central to the “forward engineering” of biological systems. For example, synthetic biology aims to modify cells to produce important molecules, such as a biofuel or a drug. Hallett and his team can play a “what if” game, computationally exploring many cellular perturbations to see which one best produces the target product. This is a much faster way to get answers versus experimental approaches.
This research is expanding our ability to explore and understand biological systems.