PhD Defense: "Toward a CMOS Integrated Biological Nanopore DNA Sequencer"

Danielle Morton

March 9th (Wednesday), 1:00pm
Elings Hall (CNSI), Room 1601

Ion channels and porins, collectively termed nanopores, are central to wide-range of biosensors capable of detecting proteins, viruses, therapeutic agent screening and DNA sequencing . The deceptively simple principle of ion current blockage by the analyte in the pore transduces physical and chemical structure of the analyte obstructing the channel into an electronic signature. Furthermore, the ion channel allows for single molecule spectroscopy unlike most biosensing techniques, especially DNA sequencing.

Nanopore sequencing is based on measuring the ion current while DNA is drawn sequentially through a pore, more specifically the protein pore Mycobacterium smegmatis porin A (MspA). The ion current amplitude is modulated by the size and electrostatic properties of the nucleotides within the narrowest section of the pore. Ideally, while DNA is passing through the pore, the ion current record can be directly translated into the DNA’s sequence. Nanopore sequencing possesses desirable intrinsic properties and has the potential to deliver information that complements and possibly exceeds that of other sequencing technologies. Although great progress has been made in the past 10 years, there are still critical obstacles to the full commercialization of these nanodevices, both on the sensor optimization and the control and readout circuitry.

The key reason why biological nanopore sequencers have not found wide acceptance is poor lifetime of the lipid membrane in which protein pores are embedded. This coupled with the need for minimum residence time of the DNA in the nanopore to maximize read lengths poses a tradeoff between the noise and the bandwidth. An associated issue is the area of the readout circuitry since large arrays are needed for effective sequencing with low error rates. In this talk I will discuss my efforts in overcoming some of these key challenges.

Block copolymer membranes are known to be robust alternatives to lipid membranes. However, very little characterization on the effect of the polymer membrane on the electronic properties of the protein sensor (MspA), such as noise, has been performed. I will discuss how the hydrophobic/hydrophilic block length of the block copolymer affects the noise properties of the MspA nanopore. By careful analysis of the noise spectrum I have shown that the increase is primarily in the flicker noise arising from membrane-protein interactions and an optimal polymer composition can be found to minimize these interactions.

On the electronics side I will address the challenges of realizing these sensors in a compact footprint (largely due to the low-noise front-end amplifier required at the input stage). Besides the large area consumption, another issue is the ability of the front-end to handle the wide common mode range required for implementation of the desired control protocols. To circumvent these challenges, a four-transistor switched amplifier used in a unique way will be introduced which allows the desired control and still maintains low-noise operation in a reduced footprint. The amplifier is 20 times smaller in its core area (compared to a traditional extended common-mode range amplifier) and is implemented in a low-noise (0.03 pA/√Hz ) capacitive integrating front-end architecture. Fabricated in IBM 130nm CMOS process, the architecture is used as part of a 12X12 biosensor array.

About Danielle Morton:

photo of danielle mortonDanielle Morton received the B.S. and M.S. degrees in electrical engineering from the University of California at Santa Barbara, Santa Barbara, CA, USA, in 2010 and 2014. Her research interests include a focus on system-level and IC-level electronic control between multiple physical domains such as photonic and biological interfacing. She is currently involved interfacing with protein-based bionanotechnology for the purpose of efficient DNA/RNA sequencing.

Hosted by: Professor Luke Theogarajan