Biologically-informed neural networks guide mechanistic modeling from sparse experimental data.
Biologically-informed neural networks (BINNs), an extension of physics-informed neural networks [1], are introduced and used to discover the underlying dynamics of biological systems from sparse experimental data. In the present work, BINNs are trained in a supervised learning framework to approxima...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2020-12-01
|
| Series: | PLoS Computational Biology |
| Online Access: | https://doi.org/10.1371/journal.pcbi.1008462 |
| _version_ | 1818579241421242368 |
|---|---|
| author | John H Lagergren John T Nardini Ruth E Baker Matthew J Simpson Kevin B Flores |
| author_facet | John H Lagergren John T Nardini Ruth E Baker Matthew J Simpson Kevin B Flores |
| author_sort | John H Lagergren |
| collection | DOAJ |
| description | Biologically-informed neural networks (BINNs), an extension of physics-informed neural networks [1], are introduced and used to discover the underlying dynamics of biological systems from sparse experimental data. In the present work, BINNs are trained in a supervised learning framework to approximate in vitro cell biology assay experiments while respecting a generalized form of the governing reaction-diffusion partial differential equation (PDE). By allowing the diffusion and reaction terms to be multilayer perceptrons (MLPs), the nonlinear forms of these terms can be learned while simultaneously converging to the solution of the governing PDE. Further, the trained MLPs are used to guide the selection of biologically interpretable mechanistic forms of the PDE terms which provides new insights into the biological and physical mechanisms that govern the dynamics of the observed system. The method is evaluated on sparse real-world data from wound healing assays with varying initial cell densities [2]. |
| first_indexed | 2024-12-16T06:58:34Z |
| format | Article |
| id | doaj.art-c5335f6d2e5c4065bc1ff58b82467b4f |
| institution | Directory Open Access Journal |
| issn | 1553-734X 1553-7358 |
| language | English |
| last_indexed | 2024-12-16T06:58:34Z |
| publishDate | 2020-12-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS Computational Biology |
| spelling | doaj.art-c5335f6d2e5c4065bc1ff58b82467b4f2022-12-21T22:40:14ZengPublic Library of Science (PLoS)PLoS Computational Biology1553-734X1553-73582020-12-011612e100846210.1371/journal.pcbi.1008462Biologically-informed neural networks guide mechanistic modeling from sparse experimental data.John H LagergrenJohn T NardiniRuth E BakerMatthew J SimpsonKevin B FloresBiologically-informed neural networks (BINNs), an extension of physics-informed neural networks [1], are introduced and used to discover the underlying dynamics of biological systems from sparse experimental data. In the present work, BINNs are trained in a supervised learning framework to approximate in vitro cell biology assay experiments while respecting a generalized form of the governing reaction-diffusion partial differential equation (PDE). By allowing the diffusion and reaction terms to be multilayer perceptrons (MLPs), the nonlinear forms of these terms can be learned while simultaneously converging to the solution of the governing PDE. Further, the trained MLPs are used to guide the selection of biologically interpretable mechanistic forms of the PDE terms which provides new insights into the biological and physical mechanisms that govern the dynamics of the observed system. The method is evaluated on sparse real-world data from wound healing assays with varying initial cell densities [2].https://doi.org/10.1371/journal.pcbi.1008462 |
| spellingShingle | John H Lagergren John T Nardini Ruth E Baker Matthew J Simpson Kevin B Flores Biologically-informed neural networks guide mechanistic modeling from sparse experimental data. PLoS Computational Biology |
| title | Biologically-informed neural networks guide mechanistic modeling from sparse experimental data. |
| title_full | Biologically-informed neural networks guide mechanistic modeling from sparse experimental data. |
| title_fullStr | Biologically-informed neural networks guide mechanistic modeling from sparse experimental data. |
| title_full_unstemmed | Biologically-informed neural networks guide mechanistic modeling from sparse experimental data. |
| title_short | Biologically-informed neural networks guide mechanistic modeling from sparse experimental data. |
| title_sort | biologically informed neural networks guide mechanistic modeling from sparse experimental data |
| url | https://doi.org/10.1371/journal.pcbi.1008462 |
| work_keys_str_mv | AT johnhlagergren biologicallyinformedneuralnetworksguidemechanisticmodelingfromsparseexperimentaldata AT johntnardini biologicallyinformedneuralnetworksguidemechanisticmodelingfromsparseexperimentaldata AT ruthebaker biologicallyinformedneuralnetworksguidemechanisticmodelingfromsparseexperimentaldata AT matthewjsimpson biologicallyinformedneuralnetworksguidemechanisticmodelingfromsparseexperimentaldata AT kevinbflores biologicallyinformedneuralnetworksguidemechanisticmodelingfromsparseexperimentaldata |