April 13, 2007
EE Conference Room, 1312 Mudd
Hosted by: Columbia Integrated System Laboratory (CISL)
Speaker: Arjang Hassibi, University of Texas at Austin
The biotechnology industry has greatly matured in the in the past decade, a credit to recent scientific discoveries enabled by the new detection platforms. However, the performance of current detection platforms in biotechnology is still far from the ideal and thus there is a huge room for improvement. Their specificity (SNR), throughput, and dynamic range are even unacceptable for demanding applications such as point-of-care molecular diagnostics. Today, many of the researchers in electrical engineering and its related fields find this as a unique opportunity. Accordingly, we see a significant growth in the number of collaborative research projects between electrical engineers and biotechnologists to address the fundamental challenges of high-performance biological detection, i.e., biosensors.
The recent efforts in engineering, broadly defined, to address challenges of biotechnology have been mostly technology-driven. Unfortunately, it is very common to find many biotechnology-related engineering projects which attempt to force an existing engineering solution onto a biological application. Our goal in this talk is to look at the problem of detection (and biosensing) by using an application-driven approach, and examine the imperative and performance-limiting aspects of existing biosensor platforms. Our main focus will be on the affinity-based biosensor array technology (i.e., microarray platform) which is among the most powerful and widely used detection technologies in Genomics and Proteomics.
Initially in this talk we will examine the underlying physics of bio-molecular interactions which result in measurement uncertainty in affinity-based biosensors. Subsequently, we will introduce the concept of biological shot-noise and formulate the quantum-limited SNR of biosensors and microarrays, followed by the affects of non-specific binding (interference) and probe saturation (nonlinearity) on the limits of detection. The rest of this talk involves the methods which we have developed to increase the performance of microarrays. On the biochemical side, we demonstrate how real-time detection significantly increases the minimum-detection-level (MDL), while making probe saturation irrelevant. One the sensor implementation side, we demonstrate how standard CMOS processes can be used to integrate the microarray platform into a true system-on-a-chip (SoC) real-time integrated microarray system.