Extensible Semantics for Fluidics

Abstract

Lab automation technology automatically manipulates chemical or biological samples at smaller scales than ever before, saving time and reagents. Scientists are now writing programs instead of informal, textbook-style protocols.

These programs mix computation with fluidic manipulation, introducing the difficulties of traditional programming plus liquid handling problems and domain-specific complexities. A successful programming solution must manage these complexities and leverage scientists’ domain knowledge. Importantly, the system must be accessible by users who are not programming language experts (or even programmers!).

We believe the programming language community is well-suited to address these challenges. We picture a two-fold solution: a core fluidic semantics to manage the complexities of liquid handling, and an extensible layer that ensures programs respect user-provided domain-specific properties. This talk will present how some of the problems in this area line up with—and sometimes challenge—well understood PL techniques.