The dynamic behavior of synthetic gene circuits plays a key role in ensuring their correct function. Although there has been substantial work on modeling dynamic behavior after circuit construction, the forward engineering of dynamic behavior remains …
Conference Paper (ICML): Introduces the methodology of using variational autoencoders to inference hierarchically assigned parameters of ordinary differential equation (ODE) models
Patent: The present disclosure relates to the modelling of ordinary differential equations (ODEs) based on a machine learning algorithm in the form of a variational auto encoder (VAE). For instance, this may be used to predict a temporal profile of the cell density in a growing population of cells engineered to produce a desired product such as a protein.
Patent: The present invention relates to a method for determining one or more intrinsic properties of a DNA component from a plurality of measurements obtained over a time period from a cell culture, with each cell comprising the DNA component, wherein the DNA component is involved in transcription of one or more target signals, wherein the plurality of measurements comprises measurements relating to the density of the cell culture over the time period and measurements relating to the amount of the one or more target signals in the cell culture over the time period.
Variational autoencoders can be used to infer hierarchical (global, group-level and individual-level) parameters of dynamical systems models (ordinary differential equations), and leverages gradient-based optimization