 |
Center for Nonlinear Studies |
Biomolecular computing devices can potentially be interfaced with living systems to allow for unprecedented control and interaction of processes in the cell. For example, a biomolecular circuit could be designed to sense a particular combination of mRNAs and produce a disruptive siRNA only in the correct context. One of the challenges facing designers of biomolecular devices is the variety of fragilities that arise when devices are deployed in uncertain environments, i.e. when the buffer is not perfect, when unknown strands are present, when downstream devices create varying loads on upstream ones, etc. One way to address these problems is through the use of feedback control, which allows for the construction of reliable systems from unreliable parts. That is, feedback allows for the management of uncertainty in both components and the environment. In our lab, we have designed several DNA devices that operate using feedback. In particular, I will discuss two systems: The first is a feedback controlled transcriptional circuit, using a short DNA "genelet" that produces an RNA product that inhibits itself. I show this system can be tuned to be insensitive to downstream loads. Second, is a general framework for implementing linear I/O systems from DNA gates. In particular, I show how to implement proportional derivative controllers and filters with DNA. These two systems demonstrate that feedback can be engineered in biomolecular devices, and could eventually help make computation at the molecular scale more robust and more broadly useful.
Host: Brian Munsky