CoChat

The DOI matches its Crossref record; 44 of the 44 references it cites were found in a registry or an index.

Identifiability and experimental design in perturbation studies

Open access
Gross, Torsten, Blüthgen, Nils · 2020 · Bioinformatics

Published version of a preprint

Abstract

Motivation: A common strategy to infer and quantify interactions between components of a biological system is to deduce them from the network’s response to targeted perturbations. Such perturbation experiments are often challenging and costly. Therefore, optimizing the experimental design is essential to achieve a meaningful characterization of biological networks. However, it remains difficult to predict which combination of perturbations allows to infer specific interaction strengths in a given network topology. Yet, such a description of identifiability is necessary to select perturbations that maximize the number of inferable parameters. Results: We show analytically that the identifiability of network parameters can be determined by an intuitive maximum-flow problem. Furthermore, we used the theory of matroids to describe identifiability relationships between sets of parameters in order to build identifiable effective network models. Collectively, these results allowed to device strategies for an optimal design of the perturbation experiments. We benchmarked these strategies on a database of human pathways. Remarkably, full network identifiability was achieved, on average, with less than a third of the perturbations that are needed in a random experimental design. Moreover, we determined perturbation combinations that additionally decreased experimental effort compared to single-target perturbations. In summary, we provide a framework that allows to infer a maximal number of interaction strengths with a minimal number of perturbation experiments. Availability and implementation: IdentiFlow is available at github.com/GrossTor/IdentiFlow. Supplementary information: Supplementary data are available at Bioinformatics online.

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 · 10.1093/bioinformatics/btaa404

Cites

44 of 44 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1. A maximum-flow problem determines the identifiability of interaction strengths and perturbation sensitivities when reconstructing a network from perturbation data. (A) Example network with three perturbations (yellow squares) to illustrate the algorithm. (B) The corresponding flow network to determine the identifiability of the edges into node 3 and the sensitivity of node 3 to perturbations. The flow passing through any node (besides source and sink) must not exceed one. A path carrying the resulting maximal flow of one is denoted in red (note that it is not unique). (C) The interaction strength between a given node and node 3 is identifiable if and only if the maximum flow is reduced after removing that node’s edge to the sink node. In this example, there are alternative max-flow paths that re-establish a unit-flow after removal of the according edges. Thus, the respective interaction strengths are non-identifiable. (D) Similarly, the sensitivity of node 3 to perturbation 3 is identifiable, if and only if the depicted extension of the flow network does not increase the maximum flow. In this example, the maximum flow is increased by one, again revealing non-identifiability. Note that such flow representations provide an intuitive understanding on how alterations in the network or perturbation setting affect identifiability. For example, it is obvious that if the toy model would not contain an edge from node 3 to 4, the edge from 2 to 3 would become identifiable
Fig. 1. A maximum-flow problem determines the identifiability of interaction strengths and perturbation sensitivities when reconstructing a network from perturbation data. (A) Example network with three perturbations (yellow squares) to illustrate the algorithm. (B) The corresponding flow network to determine the identifiability of the edges into node 3 and the sensitivity of node 3 to perturbations. The flow passing through any node (besides source and sink) must not exceed one. A path carrying the resulting maximal flow of one is denoted in red (note that it is not unique). (C) The interaction strength between a given node and node 3 is identifiable if and only if the maximum flow is reduced after removing that node’s edge to the sink node. In this example, there are alternative max-flow paths that re-establish a unit-flow after removal of the according edges. Thus, the respective interaction strengths are non-identifiable. (D) Similarly, the sensitivity of node 3 to perturbation 3 is identifiable, if and only if the depicted extension of the flow network does not increase the maximum flow. In this example, the maximum flow is increased by one, again revealing non-identifiability. Note that such flow representations provide an intuitive understanding on how alterations in the network or perturbation setting affect identifiability. For example, it is obvious that if the toy model would not contain an edge from node 3 to 4, the edge from 2 to 3 would become identifiableCC BY-NC · PubMed Central
Fig. 2. (A) An example network with three perturbations (yellow squares), where nodes 4 and 5 are associated with non-identifiable parameters (grey). (B) Their identifiability relationships are represented by the lattices of cyclic flats of rank r. Each cyclic flat consists of the annotated elements in addition to elements from its preceding cyclic flats. All parameters of a cyclic flat with rank r become identifiable if at least r independent flat parameters are fixed
Fig. 2. (A) An example network with three perturbations (yellow squares), where nodes 4 and 5 are associated with non-identifiable parameters (grey). (B) Their identifiability relationships are represented by the lattices of cyclic flats of rank r. Each cyclic flat consists of the annotated elements in addition to elements from its preceding cyclic flats. All parameters of a cyclic flat with rank r become identifiable if at least r independent flat parameters are fixedCC BY-NC · PubMed Central
Fig. 3. (A) The same network topology as in Figure 2 was subjected to a set of perturbations that target each node individually. Shown are distributions of numbers of identifiable edges for different experimental design strategies and an increasing number of perturbations. (B) All perturbation sequences associated with the single-target strategy and (C) one sequence associated with the multi-target sequence
Fig. 3. (A) The same network topology as in Figure 2 was subjected to a set of perturbations that target each node individually. Shown are distributions of numbers of identifiable edges for different experimental design strategies and an increasing number of perturbations. (B) All perturbation sequences associated with the single-target strategy and (C) one sequence associated with the multi-target sequenceCC BY-NC · PubMed Central
Fig. 4. Performance of different experimental design strategies on 267 human KEGG pathways. (A) Identifiability AUC, defined as area under the number of identified nodes versus number of perturbation curve, see Equation (13). (B) For each network and strategy, the average number of perturbations required for full identifiability is shown relative to the average number required for a random strategy. (C) The fraction of required perturbations correlated against the isolation score of a network [Equation (14)], r: Spearman’s rank correlation. (D) The fraction between multi-target perturbations with a specific number of targets and all multi-target perturbations (experiments) in KEGG networks of the annotated size range
Fig. 4. Performance of different experimental design strategies on 267 human KEGG pathways. (A) Identifiability AUC, defined as area under the number of identified nodes versus number of perturbation curve, see Equation (13). (B) For each network and strategy, the average number of perturbations required for full identifiability is shown relative to the average number required for a random strategy. (C) The fraction of required perturbations correlated against the isolation score of a network [Equation (14)], r: Spearman’s rank correlation. (D) The fraction between multi-target perturbations with a specific number of targets and all multi-target perturbations (experiments) in KEGG networks of the annotated size rangeCC BY-NC · 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 44 references

Research works printed but not linked: 1

  1. Minkenberg, 2017, Progress in Molecular Biology and Translational Science. Gene Editing in Plants, 111

Cited by

See all 11

Metadata

Authors
Gross, T., & Blüthgen, N.
Affiliations
Humboldt-Universität zu Berlin (DE)
Corresponding author
Blüthgen, Nils
Venue
Bioinformatics
Published
2020-07-01
Volume
36
Issue
Supplement_1
Pages
i482-i489
ISSN
  • 1367-4803 (ISSN-L)
  • 1367-4811
OpenAlex
W3042055820
Cited by
11
References
44
Citations by year
2026: 1 · 2024: 3 · 2023: 1 · 2022: 1 · 2021: 4 · 2020: 1
FWCI
0.57
Type
journal article
Language
en
Open access
Locations
  • Bioinformatics (publishedVersion, cc-by-nc)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc)
  • PubMed (publishedVersion)
  • edoc Publication server (Humboldt University of Berlin) (submittedVersion, cc-by-nc)
  • edoc Publication server (Humboldt University of Berlin)
Funders
Deutsche Forschungsgemeinschaft (RTG2424)
Topics
  • 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
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Microbial Metabolic Engineering and Bioproduction
Keywords
identifiability, experimental design, perturbation experiments, biological networks, maximum flow, matroid theory, network parameter estimation
MeSH terms
Humans, Models, Biological, Research Design

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: Identifiability and experimental design in perturbation studies. https://app.cochat.ai/papers/10.1093/bioinformatics/btaa404
The paper
Gross, T., & Blüthgen, N. (2020). Identifiability and experimental design in perturbation studies. Bioinformatics. https://doi.org/10.1093/bioinformatics/btaa404

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.