Machine Intelligence Techniques for Nucleic Acid Origami

PhD Topic: Machine Learning for Synthetic Biology

Project Aim: To develop strategies driven by machine intelligence techniques, to assist the design of Nucleic Acid Origami sequences and experimental protocols.

Project Objectives:
1) Create automated quantification of well-formed nucleic acid origami.
2) Attempt to determine causes of well-formed nucleic acid origami using machine intelligence techniques with microscopy data.
3) To use Machine Intelligence to identify and overcome bottlenecks in Nucleic Acid Origami Scaffold and Staple Generation.

Approaches:
1) To create correct, accurate and automated quantification measures for use in experiments. A computer vision pipeline will be built and validated with lab data.
2) To attempt to determine causes of well-formed nucleic acid origami using machine intelligence techniques with microscopy data. This would allow increased scalability of the technique and complexity of designs through better informed nucleic acid origami creation. UV Visual Spectroscopy data can be gathered on how strands of complementary DNA hybridize under certain conditions or parameters.
3) To use Machine Intelligence to identify and overcome bottlenecks in Nucleic Acid Origami Scaffold and Staple Generation. This would allow creation of Nucleic Acid Origami with favourable properties for being well-formed using optimal experimental conditions. The approach is to try and create an all-encompassing predictive model of well-formed origami or multiple predictive models. Proceed to test and validate them and produce a generative model that re-iterates over the predictive model(s).