Neural Network Differential Equations For Ion Channel Modelling
Mathematical models of cardiac ion channels have been widely used to study and predict the behaviour of ion currents. Typically models are built using biophysically-based mechanistic principles such as Hodgkin-Huxley or Markov state transitions. These models provide an abstract description of the un...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-08-01
|
| Series: | Frontiers in Physiology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphys.2021.708944/full |
| _version_ | 1818356452836769792 |
|---|---|
| author | Chon Lok Lei Chon Lok Lei Chon Lok Lei Gary R. Mirams |
| author_facet | Chon Lok Lei Chon Lok Lei Chon Lok Lei Gary R. Mirams |
| author_sort | Chon Lok Lei |
| collection | DOAJ |
| description | Mathematical models of cardiac ion channels have been widely used to study and predict the behaviour of ion currents. Typically models are built using biophysically-based mechanistic principles such as Hodgkin-Huxley or Markov state transitions. These models provide an abstract description of the underlying conformational changes of the ion channels. However, due to the abstracted conformation states and assumptions for the rates of transition between them, there are differences between the models and reality—termed model discrepancy or misspecification. In this paper, we demonstrate the feasibility of using a mechanistically-inspired neural network differential equation model, a hybrid non-parametric model, to model ion channel kinetics. We apply it to the hERG potassium ion channel as an example, with the aim of providing an alternative modelling approach that could alleviate certain limitations of the traditional approach. We compare and discuss multiple ways of using a neural network to approximate extra hidden states or alternative transition rates. In particular we assess their ability to learn the missing dynamics, and ask whether we can use these models to handle model discrepancy. Finally, we discuss the practicality and limitations of using neural networks and their potential applications. |
| first_indexed | 2024-12-13T19:57:27Z |
| format | Article |
| id | doaj.art-198e9942a10b4e96b5ce0bcbb90b117a |
| institution | Directory Open Access Journal |
| issn | 1664-042X |
| language | English |
| last_indexed | 2024-12-13T19:57:27Z |
| publishDate | 2021-08-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physiology |
| spelling | doaj.art-198e9942a10b4e96b5ce0bcbb90b117a2022-12-21T23:33:17ZengFrontiers Media S.A.Frontiers in Physiology1664-042X2021-08-011210.3389/fphys.2021.708944708944Neural Network Differential Equations For Ion Channel ModellingChon Lok Lei0Chon Lok Lei1Chon Lok Lei2Gary R. Mirams3Institute of Translational Medicine, Faculty of Health Sciences, University of Macau, Macau, ChinaDepartment of Biomedical Sciences, Faculty of Health Sciences, University of Macau, Macau, ChinaSchool of Mathematical Sciences, Faculty of Science and Engineering, University of Nottingham, Ningbo, ChinaCentre for Mathematical Medicine & Biology, School of Mathematical Sciences, University of Nottingham, Nottingham, United KingdomMathematical models of cardiac ion channels have been widely used to study and predict the behaviour of ion currents. Typically models are built using biophysically-based mechanistic principles such as Hodgkin-Huxley or Markov state transitions. These models provide an abstract description of the underlying conformational changes of the ion channels. However, due to the abstracted conformation states and assumptions for the rates of transition between them, there are differences between the models and reality—termed model discrepancy or misspecification. In this paper, we demonstrate the feasibility of using a mechanistically-inspired neural network differential equation model, a hybrid non-parametric model, to model ion channel kinetics. We apply it to the hERG potassium ion channel as an example, with the aim of providing an alternative modelling approach that could alleviate certain limitations of the traditional approach. We compare and discuss multiple ways of using a neural network to approximate extra hidden states or alternative transition rates. In particular we assess their ability to learn the missing dynamics, and ask whether we can use these models to handle model discrepancy. Finally, we discuss the practicality and limitations of using neural networks and their potential applications.https://www.frontiersin.org/articles/10.3389/fphys.2021.708944/fullneural networksdifferential equationselectrophysiologyion channelsmathematical modellingmodel discrepancy |
| spellingShingle | Chon Lok Lei Chon Lok Lei Chon Lok Lei Gary R. Mirams Neural Network Differential Equations For Ion Channel Modelling Frontiers in Physiology neural networks differential equations electrophysiology ion channels mathematical modelling model discrepancy |
| title | Neural Network Differential Equations For Ion Channel Modelling |
| title_full | Neural Network Differential Equations For Ion Channel Modelling |
| title_fullStr | Neural Network Differential Equations For Ion Channel Modelling |
| title_full_unstemmed | Neural Network Differential Equations For Ion Channel Modelling |
| title_short | Neural Network Differential Equations For Ion Channel Modelling |
| title_sort | neural network differential equations for ion channel modelling |
| topic | neural networks differential equations electrophysiology ion channels mathematical modelling model discrepancy |
| url | https://www.frontiersin.org/articles/10.3389/fphys.2021.708944/full |
| work_keys_str_mv | AT chonloklei neuralnetworkdifferentialequationsforionchannelmodelling AT chonloklei neuralnetworkdifferentialequationsforionchannelmodelling AT chonloklei neuralnetworkdifferentialequationsforionchannelmodelling AT garyrmirams neuralnetworkdifferentialequationsforionchannelmodelling |