Physics-based machine learning for modeling stochastic IP3-dependent calcium dynamics
This work addresses the challenge of efficiently simulating complex biological systems, such as calcium oscillations in non-excitable cells, with an incremental approach that integrates domain-specific physics into machine learning.
The authors tackled the problem of modeling stochastic IP3-dependent calcium dynamics by developing a physics-based machine learning method for model reduction, which improved generalization and allowed a large reduction in network size for a classic model.
We present a machine learning method for model reduction which incorporates domain-specific physics through candidate functions. Our method estimates an effective probability distribution and differential equation model from stochastic simulations of a reaction network. The close connection between reduced and fine scale descriptions allows approximations derived from the master equation to be introduced into the learning problem. This representation is shown to improve generalization and allows a large reduction in network size for a classic model of inositol trisphosphate (IP3) dependent calcium oscillations in non-excitable cells.