ABSTRACT
Experiment design and execution is a central activity in the natural sciences. The SeqER system provides a general architecture for the integration of automated planning techniques with a variety of domain knowledge in order to plan scientific experiments. These planning techniques include rule-based methods and, especially, the use of derivational analogy. Derivational analogy allows planning experience, captured as cases, to be reused. Analogy also allows the system to function in the absence of strong domain knowledge. Cases are efficiently and flexibly retrieved from a large casebase using massively parallel methods. The SeqER system is initially configured to plan protein sequencing experiments. Planning is interleaved with experiment execution, simulated using the SequenceIt program. SeqER interacts with a human user who analyzes the data generated by laboratory procedures and supplies SeqER with new hypotheses. SeqER is a vehicle in which to test theories about how scientists reason about experimental design.
