Abstract

We are fortunate to witness and participate in the moment when biological science is becoming mathematical and computable, as the physical sciences already are. New challenges being met along the way include the intrinsic complexity and multiscale nature, as well as the stochasticity, of most biological systems and questions. Symbolic computing has a salutary role to play in addressing these challenges. Symbolic computing augments numerical computing with algebra, calculus, symbolic logic, probabilistic models, and meta-programming, among other benefits of formalized abstraction. I will illustrate symbolic modeling languages with plant science examples from the “Computable Plant” collaboration, and also stochastic dynamics in synapses. The work can be organized by mapping biological hierarchies (eg. of scale) into mathematical hierarchies (of abstraction). A key idea in doing so is to find computational methods for radical model reduction, in which the number of degrees of freedom can be substantially reduced while preserving some aspects of dynamical behavior.

Joint work with many, including my hosts at the Sainsbury Laboratory Cambridge University (SLCU).