Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 48 of the 48 references it cites were found in a registry or an index.

Robust network inference using response logic

Open access

Published version of a preprint

Abstract

Motivation: A major challenge in molecular and cellular biology is to map out the regulatory networks of cells. As regulatory interactions can typically not be directly observed experimentally, various computational methods have been proposed to disentangling direct and indirect effects. Most of these rely on assumptions that are rarely met or cannot be adapted to a given context. Results: We present a network inference method that is based on a simple response logic with minimal presumptions. It requires that we can experimentally observe whether or not some of the system’s components respond to perturbations of some other components, and then identifies the directed networks that most accurately account for the observed propagation of the signal. To cope with the intractable number of possible networks, we developed a logic programming approach that can infer networks of hundreds of nodes, while being robust to noisy, heterogeneous or missing data. This allows to directly integrate prior network knowledge and additional constraints such as sparsity. We systematically benchmark our method on KEGG pathways, and show that it outperforms existing approaches in DREAM3 and DREAM4 challenges. Applied to a novel perturbation dataset on PI3K and MAPK pathways in isogenic models of a colon cancer cell line, it generates plausible network hypotheses that explain distinct sensitivities toward various targeted inhibitors due to different PI3K mutants. Availability and implementation: A Python/Answer Set Programming implementation can be accessed at github.com/GrossTor/response-logic. Data and analysis scripts are available at github.com/GrossTor/response-logic-projects. 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/btz326

Cites

48 of 48 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1. The steps of the response logic approach. The response logic and all additional prior network knowledge are formulated as a logic program. It is first used to rectify the experimentally determined response pattern, and second, takes the resulting (potentially incomplete) transitive closure as input to infer either all individual conforming networks or the union thereof
Fig. 1. The steps of the response logic approach. The response logic and all additional prior network knowledge are formulated as a logic program. It is first used to rectify the experimentally determined response pattern, and second, takes the resulting (potentially incomplete) transitive closure as input to infer either all individual conforming networks or the union thereofCC BY-NC · PubMed Central
Fig. 2. Response logic inference of a toy network. (A) From top to bottom: an example network of five nodes, where flashes indicate which nodes were perturbed; the full transitive closure; the response pattern that captures parts of the network’s transitive closure, with missing or misclassified data and confidence scores. (B) Three instants during data rectification: data points are added sequentially from high to low confidence (stars in top row), and increasingly constrain the inferred network and the (rectified) response pattern (red and blue fields in top row). Bottom row shows the inferred network at the given instant during rectification
Fig. 2. Response logic inference of a toy network. (A) From top to bottom: an example network of five nodes, where flashes indicate which nodes were perturbed; the full transitive closure; the response pattern that captures parts of the network’s transitive closure, with missing or misclassified data and confidence scores. (B) Three instants during data rectification: data points are added sequentially from high to low confidence (stars in top row), and increasingly constrain the inferred network and the (rectified) response pattern (red and blue fields in top row). Bottom row shows the inferred network at the given instant during rectificationCC BY-NC · PubMed Central
Fig. 3. The performance of the response logic approach on synthetic data generated from 270 human KEGG pathways (Kanehisa et al., 2017). N denotes network size, ϵM quantifies the fraction of missing and ϵC the fraction of misclassified data points. (A) Data generation and scoring scheme. (B) Each dot per colour represents a different pathway, colours represent different parameters for misclassification (ϵC) and missing data (ϵM). (C) Precision and recall for three particular signalling pathways as a function of the fraction of misclassified or missing data
Fig. 3. The performance of the response logic approach on synthetic data generated from 270 human KEGG pathways (Kanehisa et al., 2017). N denotes network size, ϵM quantifies the fraction of missing and ϵC the fraction of misclassified data points. (A) Data generation and scoring scheme. (B) Each dot per colour represents a different pathway, colours represent different parameters for misclassification (ϵC) and missing data (ϵM). (C) Precision and recall for three particular signalling pathways as a function of the fraction of misclassified or missing dataCC BY-NC · PubMed Central
Fig. 4. (A) Performance of the response logic approach for the gene-network reverse engineering challenges DREAM3 and DREAM4 (Greenfield et al., 2010; Marbach et al., 2010) (green bars), compared with a ‘naïve’ scoring approach (orange bars) and the 10 best approaches that took part in the respective challenges (grey bars). Scores are calculated as in the original challenge, with higher scores indicating better performance. (B) Relative changes in performance when excluding data points with confidence below a certain threshold. N: network size
Fig. 4. (A) Performance of the response logic approach for the gene-network reverse engineering challenges DREAM3 and DREAM4 (Greenfield et al., 2010; Marbach et al., 2010) (green bars), compared with a ‘naïve’ scoring approach (orange bars) and the 10 best approaches that took part in the respective challenges (grey bars). Scores are calculated as in the original challenge, with higher scores indicating better performance. (B) Relative changes in performance when excluding data points with confidence below a certain threshold. N: network sizeCC BY-NC · PubMed Central
Fig. 5. (A) Response pattern of the SW-48 cell line of selected phospho-proteins after perturbations affecting EGFR, PDGFR, ERK1 and AKT. (B) Performance of response logic network inference under various (combinations of) heuristics, as explained in text, compared to a random classifier (shaded colours). (C) Literature network (filled arrows) and final network prediction (finer arrows, only links with consensus ratio ≥0.4 are shown)
Fig. 5. (A) Response pattern of the SW-48 cell line of selected phospho-proteins after perturbations affecting EGFR, PDGFR, ERK1 and AKT. (B) Performance of response logic network inference under various (combinations of) heuristics, as explained in text, compared to a random classifier (shaded colours). (C) Literature network (filled arrows) and final network prediction (finer arrows, only links with consensus ratio ≥0.4 are shown)CC BY-NC · PubMed Central
Fig. 6. (A) Response pattern for two PI3K-mutant cell lines derived from SW-48, carrying either the E545K or H1047R mutations in PIK3CA, as in Figure 5. (B) Mutant-specific networks derived from the response data arrows in blue (parental cell line), orange and green (two mutant cell lines), with links constrained due to literature knowledge shown with large arrows. (C) Dose–response curves for different inhibitors targeting the inferred networks in the parental cell line (blue) and the two clones with PI3K mutations (orange and green), and the area under the curve displayed as bar graphs
Fig. 6. (A) Response pattern for two PI3K-mutant cell lines derived from SW-48, carrying either the E545K or H1047R mutations in PIK3CA, as in Figure 5. (B) Mutant-specific networks derived from the response data arrows in blue (parental cell line), orange and green (two mutant cell lines), with links constrained due to literature knowledge shown with large arrows. (C) Dose–response curves for different inhibitors targeting the inferred networks in the parental cell line (blue) and the two clones with PI3K mutations (orange and green), and the area under the curve displayed as bar graphsCC 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 48 references

Research works printed but not linked: 2

  1. Bockmayr, 2002, Logic Programming, Lecture Notes in Computer Science, 85
  2. Hagberg, 2008, Proceedings of the 7th Python in Science Conference, Pasadena, CA, 11

Cited by

See all 7

Metadata

Authors
Gross, T., Wongchenko, M. J., Yan, Y., & Blüthgen, N.
Affiliations
  • Humboldt-Universität zu Berlin (DE)
  • Genentech (US)
Corresponding author
Blüthgen, Nils
Venue
Bioinformatics
Published
2019-07
Volume
35
Issue
14
Pages
i634-i642
ISSN
  • 1367-4803 (ISSN-L)
  • 1367-4811
OpenAlex
W2953607540
Cited by
7
References
48
Citations by year
2024: 2 · 2022: 1 · 2021: 3 · 2020: 1
FWCI
0.41
Type
journal article
Language
en
Open access
Locations
  • Bioinformatics (publishedVersion, cc-by-nc)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by-nc)
  • 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
  • Berlin Institute of Health
  • German Research Foundation (GRK1772, 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
network inference, gene regulatory networks, signal propagation, logic programming, answer set programming, robust inference, sparsity constraints, PI3K signaling, MAPK signaling, colon cancer cell lines, targeted inhibitors, DREAM challenge
MeSH terms
Algorithms, Logic, Computational 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.

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: Robust network inference using response logic. https://app.cochat.ai/papers/10.1093/bioinformatics/btz326
The paper
Gross, T., Wongchenko, M. J., Yan, Y., & Blüthgen, N. (2019). Robust network inference using response logic. Bioinformatics. https://doi.org/10.1093/bioinformatics/btz326

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.