CoChat

iVUN: interactive Visualization of Uncertain biochemical reaction Networks

Open access

The DOI matches its Crossref record; 46 of the 46 references it cites were found in a registry or an index.

Abstract

BACKGROUND: Mathematical models are nowadays widely used to describe biochemical reaction networks. One of the main reasons for this is that models facilitate the integration of a multitude of different data and data types using parameter estimation. Thereby, models allow for a holistic understanding of biological processes. However, due to measurement noise and the limited amount of data, uncertainties in the model parameters should be considered when conclusions are drawn from estimated model attributes, such as reaction fluxes or transient dynamics of biological species. METHODS AND RESULTS: We developed the visual analytics system iVUN that supports uncertainty-aware analysis of static and dynamic attributes of biochemical reaction networks modeled by ordinary differential equations. The multivariate graph of the network is visualized as a node-link diagram, and statistics of the attributes are mapped to the color of nodes and links of the graph. In addition, the graph view is linked with several views, such as line plots, scatter plots, and correlation matrices, to support locating uncertainties and the analysis of their time dependencies. As demonstration, we use iVUN to quantitatively analyze the dynamics of a model for Epo-induced JAK2/STAT5 signaling. CONCLUSION: Our case study showed that iVUN can be used to perform an in-depth study of biochemical reaction networks, including attribute uncertainties, correlations between these attributes and their uncertainties as well as the attribute dynamics. In particular, the linking of different visualization options turned out to be highly beneficial for the complex analysis tasks that come with the biological systems as presented here.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1186/1471-2105-14-s19-s2

Cites

46 of 46 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Workflow of model development. Answering a biological question using data-driven mechanistic modeling requires at least four essential steps: collection of measurement data (left, bottom); derivation of ODE model (left, top); estimation of the parameters of the ODE model using the measurement data (middle); and analysis of the fitted model, including the parameter and prediction uncertainties (right). Depending on the complexity of the problem, theses steps have to be iterated.
Figure 1 Workflow of model development. Answering a biological question using data-driven mechanistic modeling requires at least four essential steps: collection of measurement data (left, bottom); derivation of ODE model (left, top); estimation of the parameters of the ODE model using the measurement data (middle); and analysis of the fitted model, including the parameter and prediction uncertainties (right). Depending on the complexity of the problem, theses steps have to be iterated.CC BY · PubMed Central
Figure 2 Conceptual overview of iVUN functionality. The central component of iVUN is the graph view of the BRN, with which all static and dynamic attributes are associated. From a modeling perspective, this graph has several layers. The dynamic attributes v, x, and y are associated with the flux layer (links), state layer (nodes), and the output layer (a subset of nodes), respectively. These layers are connected, i.e., an output (e.g., y1) is associated with a set of states (here, y1 represents the sum of x1, x2, and x3) and each state depends on one or more fluxes (here, x1 depends on v1 and v2). Conversely, a flux vj influences one or more states xi, which may be part of one or more outputs yk. To obtain an overview of the attribute dynamics, iVUN can map the numerical value of attributes to the color of the respective components in the BRN and animate the time dependence (illustration: top right). In addition, line plots can be used to gain further insight into the dynamic behavior and to compare time series. iVUN supports within- experiment comparison (line plots: left) and between-experiment comparison (line plot: top center) of time series. Besides the analysis of dynamics, the analysis of statistics of single attributes (bottom center) and correlations (right) is supported at different levels of granularity. In this conceptual overview, colors are used to identify the dynamic attributes in the line plots as well as the static parameters in the statistic and correlation views with the corresponding components of the graph. In the actual visual analytics tool iVUN, correspondence of elements in the different views is shown by brushing and linking as well as labeling. In this way, color can be used for encoding of other values. Screenshot of iVUN in action can be found at www.vis.uni-stuttgart.de/iVUN.
Figure 2 Conceptual overview of iVUN functionality. The central component of iVUN is the graph view of the BRN, with which all static and dynamic attributes are associated. From a modeling perspective, this graph has several layers. The dynamic attributes v, x, and y are associated with the flux layer (links), state layer (nodes), and the output layer (a subset of nodes), respectively. These layers are connected, i.e., an output (e.g., y1) is associated with a set of states (here, y1 represents the sum of x1, x2, and x3) and each state depends on one or more fluxes (here, x1 depends on v1 and v2). Conversely, a flux vj influences one or more states xi, which may be part of one or more outputs yk. To obtain an overview of the attribute dynamics, iVUN can map the numerical value of attributes to the color of the respective components in the BRN and animate the time dependence (illustration: top right). In addition, line plots can be used to gain further insight into the dynamic behavior and to compare time series. iVUN supports within- experiment comparison (line plots: left) and between-experiment comparison (line plot: top center) of time series. Besides the analysis of dynamics, the analysis of statistics of single attributes (bottom center) and correlations (right) is supported at different levels of granularity. In this conceptual overview, colors are used to identify the dynamic attributes in the line plots as well as the static parameters in the statistic and correlation views with the corresponding components of the graph. In the actual visual analytics tool iVUN, correspondence of elements in the different views is shown by brushing and linking as well as labeling. In this way, color can be used for encoding of other values. Screenshot of iVUN in action can be found at www.vis.uni-stuttgart.de/iVUN.CC BY · PubMed Central
Figure 3 Analysis of parameter correlations in the JAK2/STAT5 signaling pathway. By linking the graph view (A) with the Pearson correlation matrix (B), scatter plots and histograms (C), our visual analytics systems iVUN allowed for a step-by-step exploration of the MCMC sample obtained for the JAK2/STAT5 signaling pathway [34]. The graph view (A) shows the reactions (links), states (nodes), and outputs (subsets of nodes surrounded by a semitransparent area). The sample mean of fluxes and states is mapped to the color of links and nodes respectively. The Pearson correlation matrix (B) facilitates the efficient search for pairs of strongly correlated parameters. The selection of individual parameter pairs in the Pearson correlation matrix followed by the automatic highlighting of associated edges in the graph view (A) allows for the identification of these parameters in the BRN. In the JAK2/STAT5 signaling pathway, pairs of strongly correlated parameters in general influence similar states and outputs of the network. In addition, there are some parameters that alter several reaction fluxes, e.g., 'SOC3Inh', and which therefore correlate with many parameters. This yields clusters of strong parameter-wise correlations, which are recognizable in the matrix view. Beyond the analysis of linear correlations using the Pearson correlation matrix, scatter plots (C) reveal nonlinear correlation structures as shown for 'SOCS3RNADelay' and 'SOCS3RNATurn'. The parameter sample for the JAK2/STAT5 signaling pathway shows only few strongly nonlinear correlations, although histograms (C) reveal that, e.g., the distribution for the parameter 'SOCS3RNATurn' is bimodal.
Figure 3 Analysis of parameter correlations in the JAK2/STAT5 signaling pathway. By linking the graph view (A) with the Pearson correlation matrix (B), scatter plots and histograms (C), our visual analytics systems iVUN allowed for a step-by-step exploration of the MCMC sample obtained for the JAK2/STAT5 signaling pathway [34]. The graph view (A) shows the reactions (links), states (nodes), and outputs (subsets of nodes surrounded by a semitransparent area). The sample mean of fluxes and states is mapped to the color of links and nodes respectively. The Pearson correlation matrix (B) facilitates the efficient search for pairs of strongly correlated parameters. The selection of individual parameter pairs in the Pearson correlation matrix followed by the automatic highlighting of associated edges in the graph view (A) allows for the identification of these parameters in the BRN. In the JAK2/STAT5 signaling pathway, pairs of strongly correlated parameters in general influence similar states and outputs of the network. In addition, there are some parameters that alter several reaction fluxes, e.g., 'SOC3Inh', and which therefore correlate with many parameters. This yields clusters of strong parameter-wise correlations, which are recognizable in the matrix view. Beyond the analysis of linear correlations using the Pearson correlation matrix, scatter plots (C) reveal nonlinear correlation structures as shown for 'SOCS3RNADelay' and 'SOCS3RNATurn'. The parameter sample for the JAK2/STAT5 signaling pathway shows only few strongly nonlinear correlations, although histograms (C) reveal that, e.g., the distribution for the parameter 'SOCS3RNATurn' is bimodal.CC BY · PubMed Central
Figure 4 Hierarchical analysis of the uncertain dynamics of the JAK2/STAT5 signaling pathway. We made use of the linking between the graph view and the line plots, as well as the linking between different line plots, to perform a hierarchical analysis of the dynamic attributes of the JAK2/STAT5 signaling pathway (A). We studied the three layers of the BRN: the output layer (B), which includes the measured quantities; the state layer (C), which includes the states of the different chemical species; and the flux layer (D), which uniquely describes the changes in the states and the dynamics of outputs. The stack of graph views (A) illustrates the animation of the graph, where the states and fluxes at a particular time point are mapped to the color of the nodes and links respectively. The line plots (B)-(E) depict the median (full line) and the 90% Bayesian confidence interval (semitransparent area). Starting from the animation, the hierarchical analysis reveals that the output trajectories (B), which are fitted to the experimental data (dots), are the most well determined properties of the systems. The uncertainties in the states (C) that determine a certain output are in general much larger. Furthermore, different combinations of fluxes can yield roughly the same state trajectories, which in general results in a further increase of the uncertainties in the fluxes (D). However, for the JAK2/STAT5 signaling pathway this increase in uncertainty cannot be observed, probably, because the fluxes are constrained by molecular conservation laws. Variations in one flux must be compensated by another flux, resulting in significant flux-flux correlation (E). Complementary to the analysis of individual experimental data, the comparison across experiments provides information about the relative importance and role of certain mechanisms.
Figure 4 Hierarchical analysis of the uncertain dynamics of the JAK2/STAT5 signaling pathway. We made use of the linking between the graph view and the line plots, as well as the linking between different line plots, to perform a hierarchical analysis of the dynamic attributes of the JAK2/STAT5 signaling pathway (A). We studied the three layers of the BRN: the output layer (B), which includes the measured quantities; the state layer (C), which includes the states of the different chemical species; and the flux layer (D), which uniquely describes the changes in the states and the dynamics of outputs. The stack of graph views (A) illustrates the animation of the graph, where the states and fluxes at a particular time point are mapped to the color of the nodes and links respectively. The line plots (B)-(E) depict the median (full line) and the 90% Bayesian confidence interval (semitransparent area). Starting from the animation, the hierarchical analysis reveals that the output trajectories (B), which are fitted to the experimental data (dots), are the most well determined properties of the systems. The uncertainties in the states (C) that determine a certain output are in general much larger. Furthermore, different combinations of fluxes can yield roughly the same state trajectories, which in general results in a further increase of the uncertainties in the fluxes (D). However, for the JAK2/STAT5 signaling pathway this increase in uncertainty cannot be observed, probably, because the fluxes are constrained by molecular conservation laws. Variations in one flux must be compensated by another flux, resulting in significant flux-flux correlation (E). Complementary to the analysis of individual experimental data, the comparison across experiments provides information about the relative importance and role of certain mechanisms.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

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.

The self-check needs the paper’s full text.

Citations

Cites

Show all 46 references

Cited by

See all 11

Metadata

Authors
Vehlow, C., Hasenauer, J., Kramer, A., Raue, A., Hug, S., Timmer, J., Radde, N., Theis, F. J., & Weiskopf, D.
Affiliations
  • University of Stuttgart (DE)
  • Helmholtz Munich (DE)
  • University of Freiburg (DE)
Corresponding author
  • Vehlow, Corinna
  • Hasenauer, Jan
Venue
BMC Bioinformatics
Published
2013-11
Volume
14
Issue
S19
Pages
S2-S2
ISSN
1471-2105 (ISSN-L)
OpenAlex
W2059272641
Cited by
11
References
46
Citations by year
2026: 1 · 2021: 1 · 2018: 4 · 2017: 1 · 2016: 3 · 2014: 1
FWCI
1.50
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • BMC Bioinformatics (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Site cant be reached (submittedVersion, cc-by)
  • OPUS (Augsburg University) (submittedVersion, cc-by)
Funders
  • Deutsches Krebsforschungszentrum
  • European Commission
  • Deutsche Forschungsgemeinschaft (EXC 3101, EXC 310, (EXC 294))
  • Bundesministerium für Bildung und Forschung (EXC 294, 0315766, 0316042G, 0315494A)
  • Universität Stuttgart (EXC310/1)
Topics
  • Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › Data Visualization and Analytics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Bioinformatics and Genomic Networks
Keywords
interactive visualization, uncertainty visualization, biochemical reaction networks, parameter uncertainty, ordinary differential equation model, visual analytics, node-link diagrams, multi-view visualization
MeSH terms
Computer Graphics, Models, Biological, Models, Chemical, Signal Transduction, Computational Biology, Uncertainty, Metabolic Networks and Pathways

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.

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: iVUN: interactive Visualization of Uncertain biochemical reaction Networks. https://app.cochat.ai/papers/10.1186/1471-2105-14-s19-s2
The paper
Vehlow, C., Hasenauer, J., Kramer, A., Raue, A., Hug, S., Timmer, J., Radde, N., Theis, F. J., & Weiskopf, D. (2013). iVUN: interactive Visualization of Uncertain biochemical reaction Networks. BMC Bioinformatics. https://doi.org/10.1186/1471-2105-14-s19-s2

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.