DMREF: DNA-Nanocarbon Hybrid Materials for Perception-Based, Analyte-Agnostic Sensing

Lehigh University, Bethlehem PA

Investigators

Abstract

Non-technical Description: Sensing specific molecules is essential for life and forms the basis of many diagnostic technologies. Often, this is achieved by one sensor-one analyte lock-and-key mechanisms, e.g., by antibody-antigen binding. This project develops materials for an alternative approach to sensing based on an artificial perception system. This system comprises a set of nanosensor elements chosen from the family of DNA/Single-Walled Carbon Nanotubes (SWCNT) and will work on bodily fluids. Each nanosensor element (a particular DNA/SWCNT combination) will have only a weakly specific response to an analyte or physiological state. However, acting in concert with machine learning (ML) techniques and high throughput experimental interrogation, a suitably designed nanosensor array can accurately detect or measure multiple analytes or physiological states in biofluids. A compelling feature of this approach will be the ability for a nanosensor array to be designed first and primarily by the choice of nanosensor elements and secondarily by the use of machine learning algorithms. Because the nanosensor array will be built as initially analyte-agnostic, the optimally designed sensor array has the potential to be a universal biofluid sensing system capable of diagnosing many diseases. Technical Description: The principal goal of this project will be to test the hypothesis that well-chosen nanosensor arrays can be found and trained to detect a variety of analytes or physiological states in aqueous bodily fluids. Three principal aims are proposed: to develop a method for selection of a functionally diverse set of nanosensor elements from the family of DNA/Single-Walled Carbon Nanotubes (SWCNT) based on a standard amino acid test; to demonstrate that an initially analyte-agnostic pool of nanosensor elements can be used as elements of a system that can be trained to sense specific analytes or physiological conditions; and to establish the structural and physical basis for analyte-DNA-SWCNT interactions using Coarse-Grained (CG) and All-Atom (AA) molecular models. The project integrates high-throughput experimentation, machine learning (ML), and physical modeling in order to accomplish the Materials Genome Initiative (MGI) mission to discover, manufacture, and deploy advanced materials much more quickly and at a much-reduced cost. The multidisciplinary team is led by scientists from three institutions, including a chemist, a materials scientist, a bioengineer, and a computer scientist. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.