Constraint-driven design automation frameworks for microfluidics and scalable biological networks
Embargo Date
2028-01-29
OA Version
Citation
Abstract
Microfluidic devices enable precise, programmable manipulation of fluids at the microscale, forming the basis of many modern biological and chemical workflows. Despite their powerful capabilities, the design and operation of these devices remain highly manual, requiring extensive domain expertise to ensure correct execution. This limits scalability, slows iteration, and restricts broader adoption. This dissertation develops a set of computational techniques that automate critical stages of the microfluidic design and execution pipeline, providing a path toward more reliable, scalable, and accessible experimentation. First, the work introduces constraint-driven verification methods that evaluate whether an experimental protocol is compatible with a given microfluidic device. By identifying conflicts between biological intent, device topology, and operational constraints before execution, these tools reduce trial-and-error and improve experimental robustness. Second, the dissertation presents algorithms for organizing and partitioning large biological networks, enabling complex multicellular or multi-component behaviors to be mapped onto physically realizable experimental systems. Finally, it introduces a web-based design automation platform that generates microfluidic layouts directly from structured, high-level specifications, replacing manual drafting with automated synthesis and significantly accelerating the design–build–test process. Together, these contributions form a unified framework for constraint-driven microfluidic automation. By linking high-level biological intent to verified protocols and manufacturable device designs, this work lays the foundation for next-generation experimental platforms in synthetic biology and related fields—platforms in which sophisticated biological workflows can be rapidly translated into reliable, reproducible, and scalable microfluidic implementations.
Description
2026