Machine learning informed nucleic acid origami design

The creation of devices operating at nanoscale has been a focal point of nanotechnology, recently leading to interesting applications, e.g. nanorobots delivering drug compounds or bio-computing interfaces. The breakthrough point was the advent of scaffolded DNA origami, a technique to fold a nucleic acid sequence into a desired shape, with the use of "programmable" staples (short oligonucleotides). Despite the initial success of this technique, several challenges to its widespread use still remain, including limits to its scalability and the complexity of created structures.

In this talk, I will discuss the aforementioned challenges and potential ways to address them. Then, I will describe machine learning strategies we are developing, to assist in the design of experimental lab protocols of well-folded origami nanostructures.