A programmable architecture and compiler for microfluidics
Abstract
Advances in microfluidic research has enabled lab-on-a-chip (LoC) technology to achieve miniaturization and integration of biological and chemical analyses to a single chip comprising channels, valves, mixers, heaters, separators, and sensors. These miniature instruments offer the rare combination of faster, cheaper, and higher-precision analyses in comparison to conventional bench-scale methods. LoCs have been applied to diverse domains such as proteomics, genomics, biochemistry, and chemical synthesis. However, to date LoCs have been designed as application-specific chips which incurs significant design effort, turn-around time, and cost, and degrades designer and user productivity. To address these limitations, I envision a suite of technologies to enable programmable LoC (PLoC). They include a comprehensive fluidic instruction set, called AquaCore Instruction Set (AIS), a fluidic microarchitecture, called AquaCore, and a high-level language and compiler. In this work I present five key contributions. First, I present my design decisions for AIS and AquaCore made on the basis of key design aspects in which AIS and AquaCore differ from their computer counterparts. Second, I identify and address a practical issue, namely, fluid volume management. Volume management addresses the problem that the use of a fluid depletes it and unless the given volume of a fluid is distributed carefully among all its uses, execution may run out of the fluid before all its uses are complete. Third, I identify and address another practical issue, that of fluid contamination. Contamination arises because fluid flow is not perfect and leaves residue in the PLoC's components (e.g., channels and reservoirs) despite efforts to reduce such residue. This residue may contaminate the next fluid that uses the same components. Fourth, I establish a benchmark suite of real-world assays that cover a wide range of domains, and use them to evaluate the PLoC concept and proposed solutions to the fluid volume management and contamination issues. Finally, I develop a working PLoC prototype, compiler suite, runtime system and control platform and use it to run two real-world diverse assays as a proof-of-concept to the PLoC approach.
Degree
Ph.D.
Advisors
Thottethodi, Purdue University.
Subject Area
Computer Engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.