Predicting Composition of Genetic Circuits with Resource Competition: Demand and Sensitivity

The
design of genetic circuits typically relies on characterization
of constituent modules in isolation to predict the behavior of modules’
composition. However, it has been shown that the behavior of a genetic
module changes when other modules are in the cell due to competition
for shared resources. In order to engineer multimodule circuits that
behave as intended, it is thus necessary to predict changes in the
behavior of a genetic module when other modules load cellular resources.
Here, we introduce two characteristics of circuit modules: the demand
for cellular resources and the sensitivity to resource loading. When
both are known for every genetic module in a circuit library, they
can be used to predict any module’s behavior upon addition
of any other module to the cell. We develop an experimental approach
to measure both characteristics for any circuit module using a resource
sensor module. Using the measured resource demand and sensitivity
for each module in a library, the outputs of the modules can be accurately
predicted when they are inserted in the cell in arbitrary combinations.
These resource competition characteristics may be used to inform the
design of genetic circuits that perform as predicted despite resource
competition.