When Proteins Become Robots - A Roboticists Perspective on Structural Molecular Biology

Life depends on proteins. For example, oxygen transport, digestion of nutrients, contraction of muscles, fighting off a flu, or simply providing a stable structure to skin, tendons, and other tissue: none of it would be possible without proteins. Even plants cannot exist without them: when they perform photosynthesis, for example, proteins get it all started. Knowing how exactly proteins perform their functions would enable us to increase food production, clean up toxic waste, and cure or even eradicate many diseases. But there are many challenges on the way towards this objective. With the sequencing of the human genome (and of many other species), we have gained knowledge, at least in principle, of the "biological recipe" for most biologically relevant proteins. To understand their function, however, we must either observe physical phenomena at tiny temporal and spatial scales, or solve intractable computational problems. Both are, generally speaking, beyond our reach today. In this presentation, I will talk about our attempts to work towards the latter option. To do so, we leverage insights, methods, and algorithms from robotics and apply them to problems such as proteins structure prediction, protein motion simulation, and protein loop modeling. I will introduce all relevant biological concepts; no prior knowledge is required.