Machine-learning model selection and parameter estimation from kinetic data of complex first-order reaction systems
Open Access
- 9 August 2021
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16 (8) , e0255675
Abstract
Dealing with a system of first-order reactions is a recurrent issue in chemometrics, especially in the analysis of data obtained by spectroscopic methods applied on complex biological systems. We argue that global multiexponential fitting, the still common way to solve such problems, has serious weaknesses compared to contemporary methods of sparse modeling. Combining the advantages of group lasso and elastic net—the statistical methods proven to be very powerful in other areas—we created an optimization problem tunable from very sparse to very dense distribution over a large pre-defined grid of time constants, fitting both simulated and experimental multiwavelength spectroscopic data with high computational efficiency. We found that the optimal values of the tuning hyperparameters can be selected by a machine-learning algorithm based on a Bayesian optimization procedure, utilizing widely used or novel versions of cross-validation. The derived algorithm accurately recovered the true sparse kinetic parameters of an extremely complex simulated model of the bacteriorhodopsin photocycle, as well as the wide peak of hypothetical distributed kinetics in the presence of different noise levels. It also performed well in the analysis of the ultrafast experimental fluorescence kinetics data detected on the coenzyme FAD in a very wide logarithmic time window. We conclude that the primary application of the presented algorithms—implemented in available software—covers a wide area of studies on light-induced physical, chemical, and biological processes carried out with different spectroscopic methods. The demand for this kind of analysis is expected to soar due to the emerging ultrafast multidimensional infrared and electronic spectroscopic techniques that provide very large and complex datasets. In addition, simulations based on our methods could help in designing the technical parameters of future experiments for the verification of particular hypothetical models.Keywords
All Related Versions
Funding Information
- Economic Development and Innovation Operative Programme of Hungary (GINOP-2.3.2-15-2016-00001)
- Hungarian Scientific Research Fund (K-124922)
- Hungarian Scientific Research Fund (PD-121170)
- Nemzeti Kutatási Fejlesztési és Innovációs Hivatal (2018-1.2.1-NKP-2018-00009)
This publication has 61 references indexed in Scilit:
- Laplace inversion of low‐resolution NMR relaxometry data using sparse representation methodsConcepts in Magnetic Resonance Part A, 2013
- Optimization with Sparsity-Inducing PenaltiesFoundations and Trends in Machine Learning, 2012
- Modelling Multi-Pulse Population Dynamics from Ultrafast SpectroscopyPLOS ONE, 2011
- Global fitting without a global model: Regularization based on the continuity of the evolution of parameter distributionsThe Journal of Chemical Physics, 2008
- Regularization and Variable Selection Via the Elastic NetJournal of the Royal Statistical Society Series B: Statistical Methodology, 2005
- Analysis of the Bacteriorhodopsin Photocycle by Singular Value Decomposition with Self-Modeling: A Critical Evaluation Using Realistic Simulated DataThe Journal of Physical Chemistry B, 2004
- The Proton Release Group of Bacteriorhodopsin Controls the Rate of the Final Step of Its Photocycle at Low pHBiochemistry, 1999
- Application of Bayesian approach to numerical methods of global and stochastic optimizationJournal of Global Optimization, 1994
- Linear Model Selection by Cross-validationJournal of the American Statistical Association, 1993
- The Relationship Between Variable Selection and Data Agumentation and a Method for PredictionTechnometrics, 1974