CoChat

Structural Identifiability of Systems Biology Models: A Critical Comparison of Methods

Open access
Chis, Oana-Teodora, Banga, Julio R., Balsa-Canto, Eva · 2011 · PLoS ONE

The DOI matches its Crossref record; 64 of the 64 references it cites were found in a registry or an index; none of the 22 sources its claims cite is checked for support yet.

Abstract

Analysing the properties of a biological system through in silico experimentation requires a satisfactory mathematical representation of the system including accurate values of the model parameters. Fortunately, modern experimental techniques allow obtaining time-series data of appropriate quality which may then be used to estimate unknown parameters. However, in many cases, a subset of those parameters may not be uniquely estimated, independently of the experimental data available or the numerical techniques used for estimation. This lack of identifiability is related to the structure of the model, i.e. the system dynamics plus the observation function. Despite the interest in knowing a priori whether there is any chance of uniquely estimating all model unknown parameters, the structural identifiability analysis for general non-linear dynamic models is still an open question. There is no method amenable to every model, thus at some point we have to face the selection of one of the possibilities. This work presents a critical comparison of the currently available techniques. To this end, we perform the structural identifiability analysis of a collection of biological models. The results reveal that the generating series approach, in combination with identifiability tableaus, offers the most advantageous compromise among range of applicability, computational complexity and information provided.

Overview

CoChat Summary

TL;DR

This paper critically compares structural identifiability methods for nonlinear systems-biology models across six case studies. It concludes that the generating series approach combined with identifiability tableaus offers the best compromise among applicability, computational complexity, and information provided.

Key findings

  • Across six models, generating series plus tableaus was most advantageous; for the glycolysis-inspired pathway it concluded global identifiability with a unique solution for all parameters, whereas the Taylor series approach gave a rank-5 tableau and multiple solutions.
  • The high-dimensional model (20 differential equations, 22 parameters, all states measured) was found structurally globally identifiable by the Taylor series approach in a few seconds and by the generating series approach; DAISY also reported structural global identifiability in prior work.
  • For the pharmacokinetics model, Taylor/generating series tableaus had full rank, guaranteeing at least structural local identifiability even with one observable; DAISY found the six parameters structurally globally identifiable for the two-observable scenario.
  • The Arabidopsis clock was not structurally globally identifiable under constant light, with a subset of parameters non-identifiable; pulse-wise stimulation and polynomial reformulation gave only partial results. For NFκB, generating series and reaction-network plus generating series concluded global/local identifiability of the parameters considered.

Methods

  • Design: critical comparison of structural identifiability methods, including Taylor series, generating series, similarity transformation, differential algebra, direct test, implicit function theorem, and reaction-network-based analysis, with identifiability tableaus.
  • Data/sample: six biological models of increasing size and complexity—Goodwin oscillator, pharmacokinetics model of receptor-mediated glucose oxidase uptake, glycolysis-inspired metabolic pathway, high-dimensional biochemical reaction model, Arabidopsis thaliana central clock, and NFκB signalling module.
  • Analysis: symbolic computations using DAISY, GenSSI, MAPLE, MATHEMATICA, and MATLAB; methods were evaluated by range of applicability, computational complexity, and information provided.

Limitations

  • Stated: no single method is amenable to every model; Taylor and generating series methods require an unknown number of derivatives; differential algebra and implicit function theorem struggle with low observable-to-parameter ratios and Michaelis-Menten/Hill kinetics; similarity transformation requires controllability/observability and can be computationally prohibitive; direct test applies only to uncontrolled/autonomous systems.
  • Stated: the selected examples are small and medium-size models; apparent conflicting local/global conclusions can arise from initial conditions, symbolic solvability, and DAISY sometimes not using initial conditions.
  • Evident from the text: several methods failed due to memory or computational errors, and some results were partial or inconclusive; the comparison is limited to the chosen six models and the specific software implementations used.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1371/journal.pone.0027755

Cites

64 of 64 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Goodwin oscillator: Identifiability tableaus. (a) Identifiability tableau obtained by means of the power series methods for the case of full observation, (b) Identifiability tableau obtained by means of the power series methods for the case of pure polynomial form and full observation. and regard the different generating series coefficients, H is used for zero order coefficients whereas V correspond to the successive Lie derivatives of along , for example, . A black square in the coordinates indicates that the corresponding non-zero generating series coefficient depends on the parameter .
Figure 1 Goodwin oscillator: Identifiability tableaus. (a) Identifiability tableau obtained by means of the power series methods for the case of full observation, (b) Identifiability tableau obtained by means of the power series methods for the case of pure polynomial form and full observation. and regard the different generating series coefficients, H is used for zero order coefficients whereas V correspond to the successive Lie derivatives of along , for example, . A black square in the coordinates indicates that the corresponding non-zero generating series coefficient depends on the parameter .CC BY · PubMed Central
Figure 2 Pharmacokinetics model [33]. Identifiability tableau obtained by means of the Taylor/generating series method
Figure 2 Pharmacokinetics model [33]. Identifiability tableau obtained by means of the Taylor/generating series methodCC BY · PubMed Central
Figure 3 Glycolysis metabolic pathway: Identifiability tableaus. (a) Identifiability tableau obtained by means of the Taylor series method (, regards the component of the order coefficients of the Taylor series, (b) Identifiability tableau obtained by means of the generating series method.
Figure 3 Glycolysis metabolic pathway: Identifiability tableaus. (a) Identifiability tableau obtained by means of the Taylor series method (, regards the component of the order coefficients of the Taylor series, (b) Identifiability tableau obtained by means of the generating series method.CC BY · PubMed Central
Figure 4 High dimensional nonlinear model: Identifiability tableaus. (a) Identifiability tableau obtained by means of the Taylor series method, (b) Identifiability tableau obtained by means of the generating series method.
Figure 4 High dimensional nonlinear model: Identifiability tableaus. (a) Identifiability tableau obtained by means of the Taylor series method, (b) Identifiability tableau obtained by means of the generating series method.CC BY · PubMed Central
Figure 5 Arabidopsis Thaliana model: Reduced identifiability tableaus. Reduced identifiability tableau obtained by means of the (a) Taylor series and (b) generating series methods applied to the polynomial form of the model.
Figure 5 Arabidopsis Thaliana model: Reduced identifiability tableaus. Reduced identifiability tableau obtained by means of the (a) Taylor series and (b) generating series methods applied to the polynomial form of the model.CC BY · PubMed Central
Figure 6 Arabidopsis Thaliana model: Full identifiability tableau. Identifiability tableau obtained by means of the generating series method applied to the polynomial form of the model. Despite the large number of terms included in the tableau some parameters are not appearing. The analysis may be complemented with global sensitivity analysis.
Figure 6 Arabidopsis Thaliana model: Full identifiability tableau. Identifiability tableau obtained by means of the generating series method applied to the polynomial form of the model. Despite the large number of terms included in the tableau some parameters are not appearing. The analysis may be complemented with global sensitivity analysis.CC BY · PubMed Central

Verification

Claims the paper makes

0 Valid · 3 Not valid · 21 Not checked · 115 Own results and interpretations

64 more claims

Create a free account to see all 139 claims and to verify the 22 references still open.

See all claims — free account
  • Probably not a substantial claim. 29
  • It may state two separate claims as one. 49
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 64 references

Research works printed but not linked: 1

  1. [43]Denis-Vidal L Joly-Blanchard G 1996 Identifiability of some nonlinear kinetics. Proceedings of the Third Workshop on Modelling of Chemical Reaction Systems, Heidelberg

Cited by

See all 448

Metadata

Authors
Chis, O.-T., Banga, J. R., & Balsa-Canto, E.
Affiliations
Consejo Superior de Investigaciones Científicas (ES)
Venue
PLoS ONE
Published
2011-11-22
Volume
6
Issue
11
Pages
e27755
ISSN
1932-6203 (ISSN-L)
OpenAlex
W2033783472
Cited by
448
References
64
Citations by year
2026: 29 · 2025: 35 · 2024: 38 · 2023: 35 · 2022: 40 · 2021: 41 · 2020: 29 · 2019: 37 · 2018: 45 · 2017: 21 · 2016: 26 · 2015: 24 · 2014: 24 · 2013: 17 · 2012: 7
FWCI
6.23
Type
journal article
Language
en
Open access
Locations
  • PLoS ONE (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DIGITAL.CSIC (Spanish National Research Council (CSIC)) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Ministerio de Ciencia e Innovación
  • Xunta de Galicia
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
  • Physical Sciences › Engineering › Control and Systems Engineering › Control Systems and Identification
  • Physical Sciences › Mathematics › Applied Mathematics › Statistical and numerical algorithms
Keywords
structural identifiability, systems biology models, parameter estimation, nonlinear dynamic model, generating series, model structure analysis, time series, in silico experimentation
MeSH terms
Glycolysis, Models, Biological, NF-kappa B, Arabidopsis, Nonlinear Dynamics, Systems Biology

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-05, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-05 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Structural Identifiability of Systems Biology Models: A Critical Comparison of Methods. https://app.cochat.ai/papers/10.1371/journal.pone.0027755
The paper
Chis, O.-T., Banga, J. R., & Balsa-Canto, E. (2011). Structural Identifiability of Systems Biology Models: A Critical Comparison of Methods. PLoS ONE. https://doi.org/10.1371/journal.pone.0027755

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.