Towards ’Next-Next’ Generation Sequencing with Quantum Tunnelling - Experimentation and Deep Learning

This project revolves around an entirely new approach to biomolecular analysis and sequencing of DNA, RNA and other biopolymers, namely based on quantum mechanical tunnelling. Such a sequencing technology could be faster than established techniques, it could be label-free and applicable across a whole range of analytes (rather than being specific to DNA or RNA). More generally, it is intriguing to consider whether ’quantum technology’ could indeed form the basis for a new generation of bioanalytical devices in the future, even though it may still be a long way.
Conceptually, the idea is rather simple: a voltage is applied across an electrode nanojunction, either on a chip or in a Scanning Tunnelling Microscope configuration, and the resulting tunnelling current is measured. This can also be done in solution. The magnitude of the current depends on the composition and electronic properties of the junction. When different DNA nucleotides diffuse in and out of junction, characteristic changes in the tunnelling current can be recorded and used to identify the nucleotide. This has been achieved with an accuracy of about 90% for the four DNA bases individually so far - which is unfortunately not good enough for sequencing an entire strand.[1]
However, further and significant improvements are within reach and this project is designed to explore some of these systematically. This includes rigorous application of electrochemical potential control, which allows for careful ’fine-tuning’ of the junction energetics, and the application of novel methods for data analysis. To this end, we have recently demonstrated that state-of-the-art Deep Learning techniques, such as Convolutional Neural Networks, can in fact improve the detection performance, compared to conventional Support Vector Machines.[2]
Hence, the project will combine sophisticated experiments in single-molecule and nanoscience, with electrochemistry and Deep Learning and other advanced data analysis methods.[3] It is embedded in state-of-the-art infrastructure in the Albrecht group at the School of Chemistry and close collaborations with other research groups in the UK and abroad.
Please direct any enquires related to this opening to Prof. Tim Albrecht, [Email Address Removed]