文献类型：期刊文献

英文题名：Neural network aided approximation and parameter inference of non-Markovian models of gene expression

作者：Jiang, Qingchao[1];Fu, Xiaoming[1,2];Yan, Shifu[1];Li, Runlai[3];Du, Wenli[1];Cao, Zhixing[1,4];Qian, Feng[1];Grima, Ramon[2]

机构：[1]East China Univ Sci & Technol, Key Lab Smart Mfg Energy Chem Proc, Minist Educ, Shanghai, Peoples R China;[2]Univ Edinburgh, Sch Biol Sci, Edinburgh, Midlothian, Scotland;[3]Natl Univ Singapore, Dept Chem, Singapore, Singapore;[4]East China Univ Sci & Technol, State Key Lab Bioreactor Engn, Shanghai, Peoples R China

年份：2021

卷号：12

期号：1

外文期刊名：NATURE COMMUNICATIONS

收录：;WOS:【SCI-EXPANDED(收录号:WOS:000658722500002)】；

基金：Z.C., W.D. and F.Q. acknowledge the support from Natural Science Foundation of China (NSFC No. 61988101); Z.C. acknowledges the support from NSFC No. 62073137; W.D. acknowledges the support from NSFC No. 61725301; Q.J. and S.Y. acknowledge the support from NSFC No. 61973119, National Key Research and Development Program of China (2020YFA0908303) and Shanghai Rising-Star Program (20QA1402600); R.G. thanks the support from the Leverhulme Trust Grant (RPG-2018-423). We thank James Holehouse, Kaan ocal and Guido Sanguinetti for useful discussions and insightful feedback.

语种：英文

摘要：Non-Markovian models of stochastic biochemical kinetics often incorporate explicit time delays to effectively model large numbers of intermediate biochemical processes. Analysis and simulation of these models, as well as the inference of their parameters from data, are fraught with difficulties because the dynamics depends on the system's history. Here we use an artificial neural network to approximate the time-dependent distributions of non-Markovian models by the solutions of much simpler time-inhomogeneous Markovian models; the approximation does not increase the dimensionality of the model and simultaneously leads to inference of the kinetic parameters. The training of the neural network uses a relatively small set of noisy measurements generated by experimental data or stochastic simulations of the non-Markovian model. We show using a variety of models, where the delays stem from transcriptional processes and feedback control, that the Markovian models learnt by the neural network accurately reflect the stochastic dynamics across parameter space. Cells are complex systems that make decisions biologists struggle to understand. Here, the authors use neural networks to approximate the solution of mathematical models that capture the history and randomness of biochemical processes in order to understand the principles of transcription control.