The DOI matches its Crossref record; 181 of the 181 references it cites were found in a registry or an index; 104 of the 126 sources its claims cite were checked for support.
Gene co-expression analysis for functional classification and gene–disease predictions
Open accessAbstract
Gene co-expression networks can be used to associate genes of unknown function with biological processes, to prioritize candidate disease genes or to discern transcriptional regulatory programmes. With recent advances in transcriptomics and next-generation sequencing, co-expression networks constructed from RNA sequencing data also enable the inference of functions and disease associations for non-coding genes and splice variants. Although gene co-expression networks typically do not provide information about causality, emerging methods for differential co-expression analysis are enabling the identification of regulatory genes underlying various phenotypes. Here, we introduce and guide researchers through a (differential) co-expression analysis. We provide an overview of methods and tools used to create and analyse co-expression networks constructed from gene expression data, and we explain how these can be used to identify genes with a regulatory role in disease. Furthermore, we discuss the integration of other data types with co-expression networks and offer future perspectives of co-expression analysis.
Overview
CoChat Summary
TL;DR
This review guides researchers through (differential) co-expression analysis of gene expression data, especially RNA-seq, for functional classification and gene–disease prediction. It surveys network construction, module and hub detection, differential co-expression, and multi-omics integration, emphasizing that co-expression alone is correlative but can prioritize regulatory candidates when integrated with other data.
Key findings
- RNA-seq-based co-expression can assign putative functions to non-coding RNAs and splice variants and identify disease candidates; RNA-seq quantifies the over 70,000 non-coding RNAs not usually measured with microarrays, and a 20-sample minimum has been suggested for co-expression networks, with 10 million reads per sample commonly used as a cut-off.
- Hub genes have limited power for identifying causal targets: one myocardial study using 1,617 samples found known foetal gene markers upregulated in disease networks were not hub genes, and Salmonella hub genes were dispensable for growth, stress adaptation and virulence. In guilt-by-association, the fraction of module genes related to the main biological function is often <20%, and module–trait correlations can be <0.5 even when statistically significant.
- Differential co-expression can uncover regulators: in mutant cattle, differential wiring identified the causal mutation in myostatin (Mstn) even though Mstn expression itself hardly changed; biclustering of single-cell RNA-seq data identified 49 different cell types and corresponding cell-type-specific gene modules.
- Integrating other data layers—GWAS hits, eQTLs, TFBSs, methylation, protein–protein interactions—increases prediction accuracy; cis-eQTL-anchored co-expression identified TYROBP as the most likely causal factor in late-onset Alzheimer disease, supported by known Nasu-Hakola disease mutations in this gene.
Methods
- Design: narrative review and practical guide; no new primary experiment or dataset is generated. It synthesizes methods and tools for constructing and analysing co-expression networks from microarray and RNA-seq data, with a focus on differential co-expression.
- Data or sample: literature and existing tools/databases, including WGCNA, DICER, DiffCoEx, GSVD, biclustering, COXPRESdb, GeneFriends, GeneMANIA, GENEVESTIGATOR and GIANT; no primary sample or dataset was collected.
- Main analysis: describes the three-step construction of co-expression networks (pairwise correlation or mutual information, network representation, clustering into modules), module/hub/guilt-by-association interpretation, differential co-expression between predefined groups or via biclustering/GSVD, and integration with multi-omics data.
Limitations
- The paper states that co-expression networks identify correlations, not causality, and normally cannot distinguish regulatory from regulated genes; hub genes are not guaranteed to be causal or essential.
- The paper states that differential co-expression methods are sensitive to noise and outliers and require high-quality data; comparison is difficult because of a lack of gold-standard gene sets, and normalization or mapping-tool choices can affect results. Single-cell RNA-seq faces sparse expression matrices, low reads per cell, and normalization methods not attuned to these data.
- Evident from the text: as a narrative review, it does not perform a systematic, unbiased comparison of all tools; the authors note that comprehensive comparison is desirable but difficult and likely situation-dependent.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures


![Figure 3 Changes in gene co-expression patterns that can occur between samples. Differential co-expression can occur as the presence of a module in only one of the sample groups (A), as differences in the structure of the module (B) or as differences in the correlation strength between members of the modules (C). Additionally, differential co-expression can be detected if one larger interconnected module splits into several smaller ones (D) or if a group of genes changes its correlation partners [‘gene hopping’ (E)]. If sample groups are not defined before the differential co-expression analysis, or are unknown, biclustering methods can identify modules unique to a subpopulation of samples by simultaneously classifying the samples into groups in which these modules exist (F).](/paper-figures/595bc70853de429d90fea1a5b9f60f0f/gene-co-expression-analysis-for-functional-classification-and-genedisease-figure-3.jpg)

Verification
Claims the paper makes
67 Valid · 8 Not valid · 6 Not checked · 39 Own results and interpretations
Some cited sources support this claim and some don’t — read each source’s verdict below.
In the paper
“The most frequently used programs for differential clustering analysis, which have also been compared with others programs, are WGCNA [54], DICER [4] and DiffCoEx [100], all of which first identify modules co-expressed across the full set of study samples.”
§ Gene co-expression analysis for functional classification and gene–disease predictions > Differential co-expression analysis > Differential co-expression analysis between sample groups
What the sources say
- Partly supportedcites [54] sourceExperimental — accuracy not yet measured
The Results section, under “Functions for module detection,” confirms that WGCNA identifies co-expression modules through unsupervised clustering. It does not establish that WGCNA is among the most frequently used programs for differential clustering or that it always begins by finding modules across the full set of samples; the same section also describes consensus module detection across separate data sets. The citations to DICER and DiffCoEx are not being judged against this source. (LLM, conf=0.86)
- Partly supportedcites [4] sourceExperimental — accuracy not yet measured
The Introduction identifies DICER as a differential co-expression method whose meta-modules contain gene sets correlated across all classes. But the Author Summary says DICER’s first step calculates pairwise differential- and consistent-correlation scores, not modules co-expressed across the pooled study samples. This source supports the identification of DICER, but not the claimed order of operations for it. (LLM, conf=0.86)
- Disputedcites [100] sourceExperimental — accuracy not yet measured
The paper identifies DiffCoEx as a differential coexpression method, but contradicts the claim about its first step. Step 1 calculates gene correlations within each condition; Steps 2–4 use changes in those correlations to identify modules, rather than first identifying modules coexpressed across the full set of samples. The indexed copy provides section labels but no page numbers. (LLM, conf=0.88)
Checked by a reasoning model — may be wrong.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“Only a handful of tools and methods are currently available to investigate multi-omics data”
§ Gene co-expression analysis for functional classification and gene–disease predictions > Future prospects
What the sources say
- Disputedcites [177] Bersanelli, 2016, Methods for the integration of multi-omics data: mathematical aspects, BMC Bioinformatics sourceExperimental — accuracy not yet measured
The Conclusions (page unavailable) say that many established multi-layer approaches exist for genomics and transcriptomics, contrary to the claim that only a handful of methods are available for multi-omics data. The paper notes narrower coverage for some other omics and gaps in software availability, but does not describe methods overall as limited to a handful. (LLM, conf=0.85)
Checked by a reasoning model — may be wrong.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Disputed
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Gene expression and regulation can be highly tissue-specific, and most disease-related genes have tissue-specific expression abnormalities [16, 17].”
§ Gene co-expression analysis for functional classification and gene–disease predictions > Introduction
What the sources say
- Partly supportedcites [16] sourceExperimental — accuracy not yet measured
The Results report that nonessential disease genes tend to be expressed in only a few tissues and that genes associated with the same disorder tend to be expressed together in specific tissues. This supports tissue-specific expression, but the paper analyzes expression in healthy tissues; it does not establish that most disease-related genes have tissue-specific expression abnormalities or address tissue-specific regulation. (LLM, conf=0.85)
- Not checked — no full text of the sourcecites [17] source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“To create co-expression networks from RNA-seq data, a 20-sample minimum has been suggested [21, 54], and increased sample sizes produce networks with a higher functional connectivity [21, 59].”
§ Gene co-expression analysis for functional classification and gene–disease predictions > Co-expression networks > RNA-seq data for co-expression networks
What the sources say
- Supportedcites [21] source
Section 3.2 reports that networks constructed using more samples were generally higher performing. Section 2.5 defines that performance measure as functional connectivity, so the reported trend supports the claim. (LLM, conf=0.93)
- Could not verifycites [59] sourceExperimental — accuracy not yet measured
The Results passages compare RNA-seq platforms, read depth, and gene detection; the Discussion addresses cross-platform reproducibility. None addresses whether increasing the number of samples improves the functional connectivity of co-expression networks. Searches for functional connectivity, co-expression networks, and larger biological sample sizes found no supporting passage in our copy. (LLM, conf=0.72)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“One of the major benefits of RNA-seq is that it quantifies the expression of the over 70 000 non-coding RNAs not usually measured with microarrays [1]”
§ Gene co-expression analysis for functional classification and gene–disease predictions > Co-expression networks > Microarrays versus RNA-seq data
What the sources say
- Partly supportedcites [1] sourceExperimental — accuracy not yet measured
The Introduction says RNA sequencing has led to the identification of many long non-coding RNAs, and Data Collection and Processing describes expression profiles for them (page numbers unavailable). The paper does not say RNA-seq quantifies the expression of over 70,000 non-coding RNAs or compare its coverage with microarrays; its database counts are not counts of RNAs whose expression RNA-seq quantified. (LLM, conf=0.82)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
53 more claims · 34 with their cited sources checked
Create a free account to see all 120 claims and every source verdict.
- It may state two separate claims as one. 30
- Probably not a substantial claim. 36
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?
The cited source says what the claim attributes to it.
In the citing paper
“A major family of methods in this category is based on co-expression analysis (reviewed in [32]), and has been broadly used on bulk gene expression data (e.g. GENIE3 [33] and WGCNA [34], see [27, 35] for benchmark comparisons).”
From Mapping gene regulatory networks from single-cell omics data (2018) · cites this paper as [32]
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.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Gene co-expression analysis for functional classification and gene–disease predictions > Clustering and network analysis > Guilt by association
“| GSCNA [103] | • A tool that tests whether a predefined defined gene set is differentially expressed between two sample groups. |”
§ Gene co-expression analysis for functional classification and gene–disease predictions > Differential co-expression analysis > Differential co-expression analysis between sample groups
“GSNCA [103] can be used to identify differentially co-expressed gene sets, which have to be defined a priori, between two sample groups.”
Citations
Cites
Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2
Love, M. I., Huber, W., & Anders, S. (2014) · Genome Biology · doi:10.1186/s13059-014-0550-8 · Cited by 95,106
Fast and accurate short read alignment with Burrows-Wheeler transform
Li, H., & Durbin, R. (2009) · Bioinformatics · doi:10.1093/bioinformatics/btp324 · Cited by 58,607
Cytoscape: A software Environment for integrated models of biomolecular interaction networks
Shannon, P., Markiel, A., Ozier, O., Baliga, N. S., Wang, J. T., Ramage, D., Amin, N., Schwikowski, B., & Ideker, T. (2003) · Genome Research · doi:10.1101/gr.1239303 · Cited by 51,888
STAR: ultrafast universal RNA-seq aligner
Dobin, A., Davis, C. A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., Batut, P., Chaisson, M., & Gingeras, T. R. (2013) · Bioinformatics · doi:10.1093/bioinformatics/bts635 · Cited by 51,709
limma powers differential expression analyses for RNA-sequencing and microarray studies
Ritchie, M. E., Phipson, B., Wu, D., Hu, Y., Law, C. W., Shi, W., & Smyth, G. K. (2015) · Nucleic Acids Research · doi:10.1093/nar/gkv007 · Cited by 40,458
clusterProfiler: an R Package for Comparing Biological Themes Among Gene Clusters
Yu, G., Wang, L.-G., Han, Y., & He, Q.-Y. (2012) · OMICS: A Journal of Integrative Biology · doi:10.1089/omi.2011.0118 · Cited by 36,259
Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources
Huang, D. W., Sherman, B. T., & Lempicki, R. A. (2009) · Nature Protocols · doi:10.1038/nprot.2008.211 · Cited by 35,890
WGCNA: an R package for weighted correlation network analysis
Langfelder, P., & Horvath, S. (2008) · BMC Bioinformatics · doi:10.1186/1471-2105-9-559 · Cited by 27,252
featureCounts: an efficient general purpose program for assigning sequence reads to genomic features
Liao, Y., Smyth, G. K., & Shi, W. (2014) · Bioinformatics · doi:10.1093/bioinformatics/btt656 · Cited by 27,219
HISAT: a fast spliced aligner with low memory requirements
Kim, D., Langmead, B., & Salzberg, S. L. (2015) · Nature Methods · doi:10.1038/nmeth.3317 · Cited by 24,112
Cited by
Müller-Dott, S., Tsirvouli, E., Vazquez, M., Ramirez Flores, R. O., Badia-i-Mompel, P., Fallegger, R., Türei, D., Lægreid, A., & Saez-Rodriguez, J. (2023) · Nucleic Acids Research · doi:10.1093/nar/gkad841 · Cited by 380
Deep learning for inferring gene relationships from single-cell expression data
Yuan, Y., & Bar-Joseph, Z. (2019) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1911536116 · Cited by 265
Mapping gene regulatory networks from single-cell omics data
Fiers, M. W. E. J., Minnoye, L., Aibar, S., Bravo González-Blas, C., Kalender Atak, Z., & Aerts, S. (2018) · Briefings in Functional Genomics · doi:10.1093/bfgp/elx046 · Cited by 235
New insights into protein–protein interaction modulators in drug discovery and therapeutic advance
Nada, H., Choi, Y., Kim, S.-D., Jeong, K. S., Meanwell, N. A., & Lee, K. (2024) · Signal Transduction and Targeted Therapy · doi:10.1038/s41392-024-02036-3 · Cited by 258
Single-cell transcriptomic analysis of allergen-specific T cells in allergy and asthma
Seumois, G., Ramírez-Suástegui, C., Schmiedel, B. J., Liang, S. Y., Peters, B., Sette, A., & Vijayanand, P. (2020) · Science Immunology · doi:10.1126/sciimmunol.aba6087 · Cited by 216
Metadata
- Authors
- van Dam, S., Võsa, U., van der Graaf, A., Franke, L., & de Magalhães, J. P.
- Affiliations
- University Medical Center Groningen (NL)
- University of Liverpool (GB)
- Corresponding author
- van Dam, Sipko
- de Magalhães, João Pedro
- Venue
- Briefings in Bioinformatics
- Publisher
- Oxford University Press (OUP)
- Published
- 2017-01-10
- Volume
- 19
- Issue
- 4
- Pages
- bbw139
- ISSN
- 1467-5463 (ISSN-L)
- 1477-4054
- PMC
- PMC6054162
- OpenAlex
- W2577490495
- Semantic Scholar
- defdda7f3a58a5c1bb228bebcbe2fffa5a9975be
- Cited by
- 995
- References
- 181
- Citations by year
- 2026: 47 · 2025: 92 · 2024: 111 · 2023: 116 · 2022: 138 · 2021: 159 · 2020: 135 · 2019: 125 · 2018: 51 · 2017: 20
- FWCI
- 25.09
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Briefings in Bioinformatics (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- University of Groningen research database (University of Groningen / Centre for Information Technology) (publishedVersion, cc-by)
- University of Groningen research database (University of Groningen / Centre for Information Technology) (submittedVersion, cc-by)
- Funders
- University of Liverpool
- Rijksuniversiteit Groningen
- Universitair Medisch Centrum Groningen
- Directorate for Biological Sciences
- Biotechnology and Biological Sciences Research Council (BB/K016741/1)
- H2020 European Research Council (637640)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Bioinformatics and Genomic Networks
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene expression and cancer classification
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
- Keywords
- gene co-expression network, gene function prediction, disease gene prioritization, gene-disease association, non-coding genes, splice variants, transcriptional regulation, multi-omics integration, RNA sequencing
- MeSH terms
- Disease, Gene Expression Regulation, Genes, Humans, Phenotype, Computational Biology, Gene Expression Profiling, Gene Regulatory Networks
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
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).
Cited sources are checked with a language model’s help (gpt-6-sol).
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: Gene co-expression analysis for functional classification and gene–disease predictions. https://app.cochat.ai/papers/10.1093/bib/bbw139
- The paper
- van Dam, S., Võsa, U., van der Graaf, A., Franke, L., & de Magalhães, J. P. (2017). Gene co-expression analysis for functional classification and gene–disease predictions. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbw139
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.