The DOI matches its Crossref record, with a small difference in title or year; 95 of the 95 references it cites were found in a registry or an index; 6 of the 11 sources its claims cite were checked for support.
GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics
Open accessAbstract
The integrative analysis of gene sets, networks, and pathways is pivotal for deciphering omics data in translational biomedical research. To significantly increase gene coverage and enhance the utility of gene sets from diverse sources, we introduced pathways, annotated gene lists, and gene signatures (PAGs) enriched with metadata to represent biological functions. Furthermore, we established PAG-PAG networks by leveraging gene member similarity and gene regulations. However, in practice, high similarity in descriptions and gene membership often produces redundant, lengthy PAG lists, leading to gene set enrichment results that are difficult to interpret. We present Graph-Oriented Learning with Domain-Embedding Network (GOLDEN) fusion, an integrative framework that jointly leverages (i) connection-based embeddings derived from PAG-PAG relationships and (ii) semantic-based embeddings learned from PAG descriptions with a large language model (LLM). The two representations are combined via early fusion with a tunable weighting to produce a unified embedding on which clustering identifies concise, higher-level super-PAG. To assess when clustering is appropriate, we introduce a Connection Disparity Index, trained on synthetic stochastic block models, as a proxy for network "clusterability." We further optimize the number of clusters with consensus clustering. On Gene Ontology Annotation biological process benchmarks, GOLDEN fusion recovers is-a structure more accurately than either connection-only or semantic-only baselines, demonstrating consistent gains in Adjusted Rand Index and Normalized Mutual Information. Finally, we generate summaries for each super-PAG by synthesizing its member PAG descriptions using a comparative analysis of LLMs. GOLDEN fusion provides an integrated framework for interpreting omics results.
Overview
CoChat Summary
TL;DR
GOLDEN fusion combines graph embeddings from PAG–PAG networks with LLM-derived semantic embeddings to generate higher-level super-PAGs, using a Connection Disparity Index (CDI) to assess clusterability. On Gene Ontology Annotation benchmarks it outperforms graph-only and semantic-only baselines and provides LLM-generated summaries for super-PAGs.
Key findings
- GOLDEN fusion (weighted graph + semantic fusion) achieved higher ARI than either single-source baseline on GOA SPAGs 5–10 and SPAGs 10–100; the best configurations favored the semantic channel at α ≈ 0.8, and 256-d Node2Vec + 512-d OpenAI embeddings were recommended for the accuracy–efficiency trade-off.
- Graph-only clustering was best with Agglomerative Clustering (mean ARI 0.51 for SPAGs 5–10; 0.29 for SPAGs 10–100), while semantic-only clustering was also best with Agglomerative Clustering (mean ARI 0.66 and 0.55, respectively); HDBSCAN was lowest (0.01–0.12). GOLDEN fusion outperformed REVIGO, GOSemSim, ReCiPa, and clusterProfiler across both benchmarks.
- The CDI predictive model classified synthetic networks into unclusterable and clusterable categories with accuracy 0.88 and F1 score 0.9; random forest achieved F1 = 0.90. In SPAG 5–10 pairs, weak functional relevance accounted for 40.0% of CDI-high pairs versus 9.1% of CDI-low pairs, while moderate relevance was 54.5% versus 30.0%.
- GPT-4.1 gave the highest cosine similarity for PAG summarization: 0.71 ± 0.10 on SPAGs 5–10 and 0.72 ± 0.13 on SPAGs 10–100. In a CML TKI-resistant versus sensitive case study, GOLDEN fusion stabilized at k = 2 with CDI = 0.997.
Methods
- Inputs are PAG IDs or gene lists processed through PAGER 3.0; the method builds an m-type PAG–PAG graph weighted by hypergeometric gene-overlap significance, learns Node2Vec graph embeddings, generates semantic embeddings from PAG descriptions using OpenAI text-embedding-3-small, aligns them by PCA, L2-normalizes, and fuses them with a tunable weight α. Clustering is performed on the fused representation, with consensus clustering to choose k, followed by LLM summarization of each super-PAG.
- Benchmarks include the GOA biological process m-type PAG–PAG network with GOA is-a relations as ground truth: SPAGs 5–10 (300 two-member groups) and SPAGs 10–100 (127 two-member groups). KEGG and Reactome pathway datasets were also used with repeated random sampling (300 trials, 2 categories, up to 20 pathways per category). Synthetic SBMs and DC-SBMs were used to train CDI models.
- Performance was evaluated mainly with ARI and NMI against ground-truth labels; CDI was trained on clustering coefficient distribution features using classifiers including logistic regression, SVM, KNN, random forest, and XGBoost for CDI-β.
Limitations
- Stated: LLM-generated summaries lack human curation and are intended to support interpretation/navigation, not replace curated ontology definitions or expert annotation; super-PAGs are currently flattened one-level structures; higher edge thresholds can improve pattern representation but reduce network coverage; integrating multi-modality relationships increases modeling complexity; and beyond CML, additional experimental or literature-based validation of identified super-PAGs is needed.
- Evident from the provided text: several method and result sections are empty or fragmented, so details of CDI-based separation and PAG summarization are incomplete; benchmark ground truth relies on GOA is-a and pathway hierarchies rather than independent experimental validation.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures






Verification
Claims the paper makes
3 Valid · 0 Not valid · 1 Not checked · 74 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“One cluster summarized the concepts related to the development of resistance to chemotherapy drugs, rendering the treatment ineffective in inhibiting cancer cell proliferation or inducing cell death. This discovery has been repeatedly validated through experimental results from BCR-ABL kinase-independent pathways that promote CML cell survival and proliferation, bypassing TKI inhibition [84, 85].”
§ GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics > Results > Case study of chronic myeloid leukemia tyrosine kinase inhibitor resistant versus sensitive
What the sources say
- Supportedcites [84] Alves, 2021, Resistance to tyrosine kinase inhibitors in chronic myeloid leukemia-from molecular mechanisms to clinical relevance, Cancers (Basel) source
Section 2.4, “Alternative Signaling Pathways” (page unavailable), states that CML cells can activate alternative pathways and continue to proliferate and survive despite effective BCR-ABL1 inhibition. This supports the BCR-ABL1-independent TKI-resistance mechanism attributed to the source; the description of the cluster is the citing authors’ own finding. (LLM, conf=0.90)
- Partly supportedcites [85] Jabbour, 2013, Resistance to tyrosine kinase inhibition therapy for chronic myelogenous leukemia: a clinical perspective and emerging treatment options, Clin Lymphoma Myeloma Leuk sourceExperimental — accuracy not yet measured
The sections on alternative signaling pathways and stem-cell persistence report BCR-ABL-independent mechanisms by which CML cells survive imatinib treatment, with cytokine support permitting growth. The paper describes the Src pathway as one that “seems” critical and says the clinical significance of some resistance mechanisms remains under investigation; it does not establish the citing sentence’s stronger claim of repeated experimental validation. (LLM, conf=0.82)
boostedstates 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
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“we found that Spectral Clustering consistently outperformed the other two algorithms in accurately detecting the underlying cluster structures.”
§ GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics > Results > Exploring clusterability: the performance of clusterability predictive model
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.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“On Gene Ontology Annotation biological process benchmarks, GOLDEN fusion recovers is-a structure more accurately than either connection-only or semantic-only baselines, demonstrating consistent gains in Adjusted Rand Index and Normalized Mutual Information.”
- Also stated in: § Introduction · boosted
“We tested GOLDEN fusion on the Gene Ontology (GO) network and demonstrated that GOLDEN fusion outperforms baseline models that utilize any individual information in generating super-PAG.”
§ GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics > Abstract
boostedstates a cause- 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.
- Rests on an assumption not shown here.
- Also stated in: § Introduction · boosted
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The clusterability predictive model was designed to classify synthetic networks into unclusterable and clusterable categories with an accuracy of 0.88 and an F1 score of 0.9.”
§ GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics > Results > Exploring clusterability: the performance of clusterability predictive model
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.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“As shown in the ARI measurement heatmap in Fig. 2, a clear boundary emerges that separates networks that can be reliably clustered from those that cannot.”
§ GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics > Results > Exploring clusterability: the performance of clusterability predictive model
boostedstates an association- 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.
40 more claims
Create a free account to see all 78 claims and every source verdict.
- It may state two separate claims as one. 26
- Probably not a substantial claim. 13
- The quote only partly states this claim. 2
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 major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics > Methods > Overview design of GOLDEN fusion workflow
“We performed early fusion of this information by concatenating their embeddings and applying a hyperparameter α to adjust the weight of each data source, resulting in a new embedding matrix Df.”
§ GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics > Methods > Fusion and consensus clustering
“We form an embedding fusion for each PAG by taking a weighted linear blend of its Node2Vec vector and its description vector.”
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
§ GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics > Methods > Fusion and consensus clustering
“${D}_{f,i}=\left(1-\alpha \right){D}_{g,i}+\alpha{D}_{g,i}$”
§ GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics > Methods > Fusion and consensus clustering
“α = 0 corresponds to using only the graph embeddings, α = 1 uses only description embeddings, and intermediate values blend the two.”
Citations
Cites
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 · cited 1×
“…redundancy control, e.g. ReCiPa [ 63] (max overlapping is 0.7, and low overlapping is 0.5), and overlap-based clustering, e.g. clusterProfiler [ 74] (default cutoff is 0.7), we applied all methods to the same GOA-derived benchmark datasets (SPAGs 5–10 and SPAGs 10–100) and evaluated using ARI…”
InfluentialEnrichr: a comprehensive gene set enrichment analysis web server 2016 update
Kuleshov, M. V., Jones, M. R., Rouillard, A. D., Fernandez, N. F., Duan, Q., Wang, Z., Koplev, S., Jenkins, S. L., Jagodnik, K. M., Lachmann, A., et al. (2016) · Nucleic Acids Research · doi:10.1093/nar/gkw377 · Cited by 10,799 · cited 1×
“These include Gorilla [ 31], DAVID [ 32], WebGestalt [ 33], EnrichR [ 34], PANTHER Gene List Analysis [ 35], TermMapper [ 36], and g:profiler [ 37], which are frequently used to provide valuable interpretations into high-throughput omics.”
REVIGO summarizes and visualizes long lists of gene ontology terms
Supek, F., Bošnjak, M., Škunca, N., & Šmuc, T. (2011) · PLoS ONE · doi:10.1371/journal.pone.0021800 · Cited by 6,264 · cited 1×
“…GOLDEN fusion against the established similarity-based redundancy reduction and clustering methods, including semantic similarity, e.g. REVIGO [ 71] (default similarity cutoff is 0.7) and the Lin measure via GOSemSim [ 72, 73] (default similarity cutoff is 0.7), Jaccard-based redundancy…”
InfluentialSherman, B. T., Hao, M., Qiu, J., Jiao, X., Baseler, M. W., Lane, H. C., Imamichi, T., & Chang, W. (2022) · Nucleic Acids Research · doi:10.1093/nar/gkac194 · Cited by 6,142 · cited 1×
“These include Gorilla [ 31], DAVID [ 32], WebGestalt [ 33], EnrichR [ 34], PANTHER Gene List Analysis [ 35], TermMapper [ 36], and g:profiler [ 37], which are frequently used to provide valuable interpretations into high-throughput omics.”
KEGG for taxonomy-based analysis of pathways and genomes
Kanehisa, M., Furumichi, M., Sato, Y., Kawashima, M., & Ishiguro-Watanabe, M. (2023) · Nucleic Acids Research · doi:10.1093/nar/gkac963 · Cited by 5,833 · cited 1×
“…has significantly enhanced the analytical power, enabling comprehensive analysis from various sources, including pathways (Reactome [ 44], KEGG [ 45], BioCarta, WikiPathways [ 46], DMaps [ 47], SPIKE, NCI Nature Curated pathway [ 48], and GeoMx Cancer Transcriptome Atlas [ 49]), ontology…”
Raudvere, U., Kolberg, L., Kuzmin, I., Arak, T., Adler, P., Peterson, H., & Vilo, J. (2019) · Nucleic Acids Research · doi:10.1093/nar/gkz369 · Cited by 5,045 · cited 1×
“These include Gorilla [ 31], DAVID [ 32], WebGestalt [ 33], EnrichR [ 34], PANTHER Gene List Analysis [ 35], TermMapper [ 36], and g:profiler [ 37], which are frequently used to provide valuable interpretations into high-throughput omics.”
The Gene Ontology Resource: 20 years and still GOing strong
The Gene Ontology Consortium (2019) · Nucleic Acids Research · doi:10.1093/nar/gky1055 · Cited by 4,747 · cited 1×
“…DMaps [ 47], SPIKE, NCI Nature Curated pathway [ 48], and GeoMx Cancer Transcriptome Atlas [ 49]), ontology annotations (Gene Ontology Annotation [ 50] also known as GOA and Human Phenotype Ontology), tissue-specific expressions (GTEx [ 51], NGS Catalog [ 52], and TCGA [ 53], PathologyAtlas […”
Review The Cancer Genome Atlas (TCGA): an immeasurable source of knowledge
Tomczak, K., Czerwińska, P., & Wiznerowicz, M. (2015) · Współczesna Onkologia · doi:10.5114/wo.2014.47136 · Cited by 3,492
A pathology atlas of the human cancer transcriptome
Uhlen, M., Zhang, C., Lee, S., Sjöstedt, E., Fagerberg, L., Bidkhori, G., Benfeitas, R., Arif, M., Liu, Z., Edfors, F., et al. (2017) · Science · doi:10.1126/science.aan2507 · Cited by 3,371 · cited 1×
“…[ 50] also known as GOA and Human Phenotype Ontology), tissue-specific expressions (GTEx [ 51], NGS Catalog [ 52], and TCGA [ 53], PathologyAtlas [ 54], and FANTOM5 [ 55]), cell-specific expressions (CellAtlas [ 56]), disease phenotypes (GAD, GWAS Catalog [ 57], Phewas [ 58], and Online Mendelian…”
GOrilla: a tool for discovery and visualization of enriched GO terms in ranked gene lists
Eden, E., Navon, R., Steinfeld, I., Lipson, D., & Yakhini, Z. (2009) · BMC Bioinformatics · doi:10.1186/1471-2105-10-48 · Cited by 3,356
Research works printed but not linked: 3
- Zhang, 2013, Breast cancer subtyping from plasma proteins, BMC Med Genomics, 2013, 1
- Lin, 2004, ROUGE: A Package for Automatic Evaluation of Summaries
- OpenAi, 2026, ChatGPT (April Version) [Large Language AI Model]
Metadata
- Authors
- Li, Q., Nichols, C., Welner, R. S., Chen, J. Y., Ku, W.-S., & Yue, Z.
- Affiliations
- Fisk University (US)
- Auburn University (US)
- University of Alabama at Birmingham (US)
- Corresponding author
- Chen, Jake Y
- Ku, Wei-Shinn
- Yue, Zongliang
- Venue
- Briefings in Bioinformatics
- Publisher
- Oxford University Press (OUP)
- Published
- 2026-05
- Volume
- 27
- Issue
- 3
- ISSN
- 1467-5463 (ISSN-L)
- 1477-4054
- PMC
- PMC13200546
- OpenAlex
- W7162300909
- Cited by
- 0
- References
- 81
- FWCI
- 0.00
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Briefings in Bioinformatics (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Funders
- University of Alabama at Birmingham Center for Clinical and Translational Science Pilot Award (UM1TR004771)
- Gulf Research Program of the National Academies of Sciences (SCON-10001538)
- University of Alabama at Birmingham (UC267384)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Bioinformatics and Genomic Networks
- Health Sciences › Medicine › Health Informatics › Biomedical Text Mining and Ontologies
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Genomics and Phylogenetic Studies
- Keywords
- gene expression signatures, graph embedding, domain embedding, large language models, semantic embeddings, feature-level fusion, consensus clustering, Gene Ontology biological process, omics data interpretation, functional genomics
- MeSH terms
- Large Language Models, Humans, Genomics, 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 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (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: GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics. https://app.cochat.ai/papers/10.1093/bib/bbag244
- The paper
- Li, Q., Nichols, C., Welner, R. S., Chen, J. Y., Ku, W.-S., & Yue, Z. (2026). GOLDEN fusion: a graph-oriented learning with domain-embedding network fusion for generating super gene sets in functional genomics. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag244
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.