JuliaCon 2020 (times are in UTC)

Inference of Bifurcations with Differentiable Continuation
07-30, 19:00–19:30 (UTC), Red Track

In this talk I will demonstrate a gradient-based semi-supervised approach for matching target codimension one bifurcations with parameterised differential equation models. This work has been applied in synthetic biology settings, where experiments generate qualitative observations: locations of fixed points, limit cycles and bistability. Future outlooks include a view towards designing patterns and limit cycles in partial differential equations.


This project uses functionality of parameter continuation library BifurcationKit.jl to work with automatic differentiation library Zygote.jl. This talk would be interesting to anyone who infers parameters of differential equation models, users of DifferentialEquations.jl, Flux.jl and FluxDiffEq.jl. Repository: github.com/gszep/FluxContinuation

See also: slides (2.0 MB)

Together with Station B at Microsoft Research Cambridge and the Randall Division of Cell and Molecular Biophysics at King's College London we aim to create tools that can make biochemical manufacturing more efficient, with efforts towards combating Eroom's law of drug discovery and addressing the reproducibility crisis in biotechnology. My current research focus is the development of novel parameter inference procedures for differential equation models where data had been obtained by flow cytometry or fluorescence microscopy.