{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T17:53:46Z","timestamp":1791309226561,"version":"4.3.1"},"reference-count":36,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T00:00:00Z","timestamp":1791244800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["32260154"],"award-info":[{"award-number":["32260154"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62562041"],"award-info":[{"award-number":["62562041"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Gene Regulatory Networks (GRNs) are crucial for understanding cell fate decisions and disease mechanisms. However, inferring GRNs from single-cell RNA sequencing remains challenging due to noise, data sparsity, and species-specific distribution variations. We propose GP-DHT (Gene Pair\u2013Dual Head Transformer), a single-cell GRN inference framework that pre-trains gene representations on a unified shared gene space using the BEELINE datasets of humans and mice. It models gene-cell expression relationships through a multi-relational heterograph and predicts transcription factor (TF)-target regulatory links using a dual-head Transformer. These two heads correspond to complementary task objectives: a classification head for regulatory link prediction and a projection head for gene pair-level supervised contrastive learning. This design enables the model to optimize both prediction accuracy and representation separability in a sparse single-cell setting. The human-mouse shared-gene stage is used to initialize expression-derived gene representations in a harmonized gene space. During dataset-wise supervised training, GenePairSupCon is applied within each training fold to regularize gene-pair representations, followed by fine-tuning and evaluation within each BEELINE dataset. To enhance reproducibility and interpretability, we further clarify the mapping of orthologous genes and gene symbol unification, TF-level data partitioning and leakage control, negative label sensitivity, and quantitative embedding quality assessment. Hard negative example experiments show that negative label construction has a substantial impact on task difficulty, while supervised contrastive learning significantly improves the separability of embeddings, as measured by Silhouette score, Davies-Bouldin index, Calinski-Harabasz score, and inter\/intra distance ratio. GP-DHT can also recover prior-supported candidate regulatory modules and provide additional interpretable support through post-hoc pair-level attribution analysis on frozen predictors. In summary, these results support GP-DHT as a robust shared gene representation learning framework for dataset-level single-cell GRN prediction.<\/jats:p>","DOI":"10.1371\/journal.pone.0359164","type":"journal-article","created":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T17:27:26Z","timestamp":1791307646000},"page":"e0359164","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["GP-DHT: A dual-head transformer with supervised contrastive learning for shared-gene pretraining and single-cell GRN prediction"],"prefix":"10.1371","volume":"21","author":[{"given":"Shuai","family":"Yan","sequence":"first","affiliation":[{"name":"School of Information Engineering, Jingdezhen Ceramic University","place":["Jingdezhen, Jiangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qingzhi","family":"Yu","sequence":"additional","affiliation":[{"name":"School of Information Engineering, Jingdezhen Ceramic University","place":["Jingdezhen, Jiangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenfeng","family":"Dai","sequence":"additional","affiliation":[{"name":"School of Information Engineering, Jingdezhen Ceramic University","place":["Jingdezhen, Jiangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-1919-3414","authenticated-orcid":true,"given":"Xiang","family":"Cheng","sequence":"additional","affiliation":[{"name":"School of Information Engineering, Jingdezhen Ceramic University","place":["Jingdezhen, Jiangxi, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,10,6]]},"reference":[{"key":"pone.0359164.ref001","doi-asserted-by":"crossref","first-page":"87","DOI":"10.1016\/j.coisb.2021.04.007","article-title":"Gene regulatory network inference in single-cell biology","volume":"26","author":"K Akers","year":"2021","journal-title":"Curr Opin Syst Biol"},{"issue":"2","key":"pone.0359164.ref002","doi-asserted-by":"crossref","first-page":"147","DOI":"10.1038\/s41592-019-0690-6","article-title":"Benchmarking algorithms for gene regulatory network inference from single-cell transcriptomic data","volume":"17","author":"A Pratapa","year":"2020","journal-title":"Nat Methods"},{"issue":"5","key":"pone.0359164.ref003","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbaf470","article-title":"Machine learning methods for gene regulatory network inference","volume":"26","author":"A Hegde","year":"2025","journal-title":"Brief Bioinform"},{"issue":"2","key":"pone.0359164.ref004","article-title":"Graph neural networks for single-cell omics data: a review of approaches and applications","volume":"26","author":"S Li","year":"2025","journal-title":"Brief Bioinform"},{"issue":"1","key":"pone.0359164.ref005","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1038\/s41540-023-00312-6","article-title":"Gene regulatory network reconstruction: harnessing the power of single-cell multi-omic data","volume":"9","author":"D Kim","year":"2023","journal-title":"NPJ Syst Biol Appl"},{"issue":"9","key":"pone.0359164.ref006","doi-asserted-by":"crossref","first-page":"1368","DOI":"10.1038\/s41592-023-01971-3","article-title":"Dictys: dynamic gene regulatory network dissects developmental continuum with single-cell multiomics","volume":"20","author":"L Wang","year":"2023","journal-title":"Nat Methods"},{"issue":"2","key":"pone.0359164.ref007","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbab568","article-title":"A hybrid deep learning framework for gene regulatory network inference from single-cell transcriptomic data","volume":"23","author":"M Zhao","year":"2022","journal-title":"Brief Bioinform"},{"issue":"5","key":"pone.0359164.ref008","article-title":"Deep learning-based cell-specific gene regulatory networks inferred from single-cell multiome data","volume":"53","author":"J Xu","year":"2025","journal-title":"Nucleic Acids Res"},{"issue":"7","key":"pone.0359164.ref009","doi-asserted-by":"crossref","first-page":"491","DOI":"10.1038\/s43588-021-00099-8","article-title":"Modeling gene regulatory networks using neural network architectures","volume":"1","author":"H Shu","year":"2021","journal-title":"Nat Comput Sci"},{"issue":"29","key":"pone.0359164.ref010","doi-asserted-by":"crossref","first-page":"10967","DOI":"10.1073\/pnas.0602294103","article-title":"Cross-species analysis of biological networks by Bayesian alignment","volume":"103","author":"J Berg","year":"2006","journal-title":"Proc Natl Acad Sci U S A"},{"issue":"4","key":"pone.0359164.ref011","doi-asserted-by":"crossref","first-page":"618","DOI":"10.1016\/j.molcel.2016.04.030","article-title":"Global Mapping of Human RNA-RNA Interactions","volume":"62","author":"E Sharma","year":"2016","journal-title":"Mol Cell"},{"key":"pone.0359164.ref012","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1186\/s12935-020-01245-4","article-title":"circRNA-miRNA-mRNA regulatory network in human lung cancer: an update","volume":"20","author":"Z-Z Liang","year":"2020","journal-title":"Cancer Cell Int"},{"issue":"6","key":"pone.0359164.ref013","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbab325","article-title":"DeepDRIM: a deep neural network to reconstruct cell-type-specific gene regulatory network using single-cell RNA-seq data","volume":"22","author":"J Chen","year":"2021","journal-title":"Brief Bioinform"},{"issue":"7","key":"pone.0359164.ref014","doi-asserted-by":"crossref","first-page":"491","DOI":"10.1038\/s43588-021-00099-8","article-title":"Modeling gene regulatory networks using neural network architectures","volume":"1","author":"H Shu","year":"2021","journal-title":"Nat Comput Sci"},{"issue":"1","key":"pone.0359164.ref015","doi-asserted-by":"crossref","first-page":"110","DOI":"10.1186\/s13059-019-1713-4","article-title":"Single-cell transcriptomics unveils gene regulatory network plasticity","volume":"20","author":"G Iacono","year":"2019","journal-title":"Genome Biol"},{"issue":"9","key":"pone.0359164.ref016","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pgen.1010942","article-title":"Inferring gene regulatory network from single-cell transcriptomes with graph autoencoder model","volume":"19","author":"J Wang","year":"2023","journal-title":"PLoS Genet"},{"issue":"1","key":"pone.0359164.ref017","doi-asserted-by":"crossref","first-page":"1882","DOI":"10.1038\/s41467-021-22197-x","article-title":"scGNN is a novel graph neural network framework for single-cell RNA-Seq analyses","volume":"12","author":"J Wang","year":"2021","journal-title":"Nat Commun"},{"issue":"2","key":"pone.0359164.ref018","article-title":"GMFGRN: a matrix factorization and graph neural network approach for gene regulatory network inference","volume":"25","author":"S Li","year":"2024","journal-title":"Brief Bioinform"},{"issue":"1","key":"pone.0359164.ref019","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1186\/s12859-025-06071-x","article-title":"HGATLink: single-cell gene regulatory network inference via the fusion of heterogeneous graph attention networks and transformer","volume":"26","author":"Y Sun","year":"2025","journal-title":"BMC Bioinform"},{"key":"pone.0359164.ref020","article-title":"Inferring gene regulatory networks by hypergraph variational autoencoder","author":"G Su","year":"2024","journal-title":"bioRxiv"},{"issue":"1","key":"pone.0359164.ref021","doi-asserted-by":"crossref","first-page":"100382","DOI":"10.1016\/j.crmeth.2022.100382","article-title":"Graph embedding and Gaussian mixture variational autoencoder network for end-to-end analysis of single-cell RNA sequencing data","volume":"3","author":"J Xu","year":"2023","journal-title":"Cell Rep Methods"},{"issue":"1","key":"pone.0359164.ref022","doi-asserted-by":"crossref","first-page":"88","DOI":"10.1186\/s13059-024-03226-6","article-title":"PMF-GRN: a variational inference approach to single-cell gene regulatory network inference using probabilistic matrix factorization","volume":"25","author":"C Skok Gibbs","year":"2024","journal-title":"Genome Biol"},{"issue":"1","key":"pone.0359164.ref023","doi-asserted-by":"crossref","first-page":"294","DOI":"10.1186\/s13059-025-03780-7","article-title":"KEGNI: knowledge graph enhanced framework for gene regulatory network inference","volume":"26","author":"P Li","year":"2025","journal-title":"Genome Biol"},{"issue":"10","key":"pone.0359164.ref024","doi-asserted-by":"crossref","first-page":"1458","DOI":"10.1038\/s41587-022-01284-4","article-title":"Multi-omics single-cell data integration and regulatory inference with graph-linked embedding","volume":"40","author":"Z-J Cao","year":"2022","journal-title":"Nat Biotechnol"},{"issue":"9","key":"pone.0359164.ref025","doi-asserted-by":"crossref","first-page":"1355","DOI":"10.1038\/s41592-023-01938-4","article-title":"SCENIC+: single-cell multiomic inference of enhancers and gene regulatory networks","volume":"20","author":"C Bravo Gonz\u00e1lez-Blas","year":"2023","journal-title":"Nat Methods"},{"issue":"1","key":"pone.0359164.ref026","doi-asserted-by":"crossref","first-page":"1337","DOI":"10.1038\/s41467-021-21583-9","article-title":"Comprehensive analysis of single cell ATAC-seq data with SnapATAC","volume":"12","author":"R Fang","year":"2021","journal-title":"Nat Commun"},{"issue":"7818","key":"pone.0359164.ref027","doi-asserted-by":"crossref","first-page":"699","DOI":"10.1038\/s41586-020-2493-4","article-title":"Expanded encyclopaedias of DNA elements in the human and mouse genomes","volume":"583","author":"Consortium ENCODE Project","year":"2020","journal-title":"Nature"},{"issue":"1","key":"pone.0359164.ref028","doi-asserted-by":"crossref","first-page":"6843","DOI":"10.1038\/s41467-021-27232-5","article-title":"Single-cell analysis identifies dynamic gene expression networks that govern B cell development and transformation","volume":"12","author":"RD Lee","year":"2021","journal-title":"Nat Commun"},{"issue":"12","key":"pone.0359164.ref029","doi-asserted-by":"crossref","first-page":"830","DOI":"10.1038\/s41422-024-01034-y","article-title":"GeneCompass: deciphering universal gene regulatory mechanisms with a knowledge-informed cross-species foundation model","volume":"34","author":"X Yang","year":"2024","journal-title":"Cell Res"},{"issue":"8","key":"pone.0359164.ref030","doi-asserted-by":"crossref","first-page":"1492","DOI":"10.1038\/s41592-024-02191-z","article-title":"Toward universal cell embeddings: integrating single-cell RNA-seq datasets across species with SATURN","volume":"21","author":"Y Rosen","year":"2024","journal-title":"Nat Methods"},{"issue":"1","key":"pone.0359164.ref031","doi-asserted-by":"crossref","first-page":"3607","DOI":"10.1038\/s41467-025-58699-1","article-title":"scPRINT: pre-training on 50 million cells allows robust gene network predictions","volume":"16","author":"J Kalfon","year":"2025","journal-title":"Nat Commun"},{"issue":"6","key":"pone.0359164.ref032","first-page":"6264","article-title":"Region or global? A principle for negative sampling in graph-based recommendation","volume":"35","author":"Z Yang","year":"2022","journal-title":"IEEE Trans Knowl Data Eng"},{"key":"pone.0359164.ref033","doi-asserted-by":"crossref","first-page":"3853","DOI":"10.52202\/075280-0169","article-title":"Evaluating graph neural networks for link prediction: Current pitfalls and new benchmarking","volume":"36","author":"J Li","year":"2023","journal-title":"Adv Neural Inform Process Syst"},{"issue":"4","key":"pone.0359164.ref034","doi-asserted-by":"crossref","first-page":"109352","DOI":"10.1016\/j.isci.2024.109352","article-title":"scGREAT: Transformer-based deep-language model for gene regulatory network inference from single-cell transcriptomics","volume":"27","author":"Y Wang","year":"2024","journal-title":"iScience"},{"issue":"6","key":"pone.0359164.ref035","article-title":"Predicting gene regulatory links from single-cell RNA-seq data using graph neural networks","volume":"24","author":"G Mao","year":"2023","journal-title":"Brief Bioinform"},{"key":"pone.0359164.ref036","doi-asserted-by":"crossref","first-page":"108938","DOI":"10.1016\/j.engappai.2024.108938","article-title":"Graph attention network with convolutional layer for predicting gene regulations from single-cell ribonucleic acid sequence data","volume":"136","author":"J Liu","year":"2024","journal-title":"Eng Appl Artif Intell"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0359164","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T17:27:35Z","timestamp":1791307655000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0359164"}},"subtitle":[],"editor":[{"given":"Aniruddha","family":"Datta","sequence":"first","affiliation":[{"name":"Texas A&M University","place":["UNITED STATES OF AMERICA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,10,6]]},"references-count":36,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,10,6]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0359164","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,6]]}}}