{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T08:13:02Z","timestamp":1781597582883,"version":"3.54.5"},"update-to":[{"DOI":"10.1371\/journal.pcbi.1014008","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,3,26]],"date-time":"2026-03-26T00:00:00Z","timestamp":1774483200000}}],"reference-count":45,"publisher":"Public Library of Science (PLoS)","issue":"3","license":[{"start":{"date-parts":[[2026,3,16]],"date-time":"2026-03-16T00:00:00Z","timestamp":1773619200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"NSF","award":["DMS2310836"],"award-info":[{"award-number":["DMS2310836"]}]},{"name":"NIH","award":["5U01AI167892"],"award-info":[{"award-number":["5U01AI167892"]}]}],"content-domain":{"domain":["www.ploscompbiol.org"],"crossmark-restriction":false},"short-container-title":["PLoS Comput Biol"],"abstract":"<jats:p>Semi-supervised methods for single-cell RNA-seq integration promise improved batch correction and preservation of biological signal by leveraging cell-type labels. However, reported benefits and robustness of them towards imperfect cell type labels often come from overly idealized settings. Here we present, to our knowledge, the first systematic benchmark comparing leading semi-supervised methods with widely used unsupervised approaches across six diverse datasets under realistic conditions. Beyond randomly missing or erroneous labels, we examine four additional scenarios (boundary-mixed labels, batch-specific annotations, auto-generated labels and varied-granularity labels) and evaluate performance using nine established metrics. We find that although semi-supervised methods can provide benefits under perfect annotations, their robustness often degrades substantially under realistic imperfections. Only scANVI and ssSTACAS maintain stable but modest improvements over their unsupervised counterparts, and none consistently outperform the strongest unsupervised approach. These results indicate that current semi-supervised strategies offer limited practical advantage when label quality is modest uncertain.<\/jats:p>","DOI":"10.1371\/journal.pcbi.1014008","type":"journal-article","created":{"date-parts":[[2026,3,16]],"date-time":"2026-03-16T17:36:22Z","timestamp":1773682582000},"page":"e1014008","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":1,"title":["A benchmark of semi-supervised scRNA-seq integration methods in real-world scenarios"],"prefix":"10.1371","volume":"22","author":[{"given":"Xiaoyu","family":"Shen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6412-450X","authenticated-orcid":true,"given":"Chuan","family":"He","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Leying","family":"Guan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,3,16]]},"reference":[{"issue":"3","key":"pcbi.1014008.ref001","doi-asserted-by":"crossref","first-page":"328","DOI":"10.1038\/s41588-022-01018-x","article-title":"Systematic reconstruction of cellular trajectories across mouse embryogenesis","volume":"54","author":"C Qiu","year":"2022","journal-title":"Nat Genet"},{"issue":"3","key":"pcbi.1014008.ref002","doi-asserted-by":"crossref","first-page":"546","DOI":"10.1038\/s41591-020-01227-z","article-title":"Single-cell meta-analysis of SARS-CoV-2 entry genes across tissues and demographics","volume":"27","author":"C Muus","year":"2021","journal-title":"Nat Med"},{"key":"pcbi.1014008.ref003","doi-asserted-by":"crossref","first-page":"101595","DOI":"10.1016\/j.molmet.2022.101595","article-title":"A transcriptional cross species map of pancreatic islet cells","volume":"66","author":"S Tritschler","year":"2022","journal-title":"Mol Metabol"},{"key":"pcbi.1014008.ref004","doi-asserted-by":"crossref","DOI":"10.7554\/eLife.27041","article-title":"The Human Cell Atlas","volume":"6","author":"A Regev","year":"2017","journal-title":"Elife"},{"issue":"7777","key":"pcbi.1014008.ref005","doi-asserted-by":"crossref","first-page":"187","DOI":"10.1038\/s41586-019-1629-x","article-title":"The human body at cellular resolution: the NIH Human Biomolecular Atlas Program","volume":"574","author":"H Consortium","year":"2019","journal-title":"Nature"},{"issue":"5","key":"pcbi.1014008.ref006","doi-asserted-by":"crossref","first-page":"421","DOI":"10.1038\/nbt.4091","article-title":"Batch effects in single-cell RNA-sequencing data are corrected by matching mutual nearest neighbors","volume":"36","author":"L Haghverdi","year":"2018","journal-title":"Nat Biotechnol"},{"issue":"6","key":"pcbi.1014008.ref007","doi-asserted-by":"crossref","DOI":"10.15252\/msb.20188746","article-title":"Current best practices in single\u2010cell RNA\u2010seq analysis: a tutorial","volume":"15","author":"MD Luecken","year":"2019","journal-title":"Mol Syst Biol"},{"issue":"1","key":"pcbi.1014008.ref008","doi-asserted-by":"crossref","first-page":"12","DOI":"10.1186\/s13059-019-1850-9","article-title":"A benchmark of batch-effect correction methods for single-cell RNA sequencing data","volume":"21","author":"HTN Tran","year":"2020","journal-title":"Genome Biol"},{"issue":"4","key":"pcbi.1014008.ref009","doi-asserted-by":"crossref","first-page":"562","DOI":"10.1093\/biostatistics\/kxx053","article-title":"Missing data and technical variability in single-cell RNA-sequencing experiments","volume":"19","author":"SC Hicks","year":"2018","journal-title":"Biostatistics"},{"issue":"1","key":"pcbi.1014008.ref010","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1038\/s41592-021-01336-8","article-title":"Benchmarking atlas-level data integration in single-cell genomics","volume":"19","author":"MD Luecken","year":"2022","journal-title":"Nat Methods"},{"issue":"1","key":"pcbi.1014008.ref011","article-title":"Probabilistic harmonization and annotation of single-cell transcriptomics data with deep generative models","volume":"17","author":"C Xu","year":"2021","journal-title":"Mol Syst Biol"},{"issue":"1","key":"pcbi.1014008.ref012","doi-asserted-by":"crossref","first-page":"7781","DOI":"10.1038\/s41467-023-43590-8","article-title":"scDREAMER for atlas-level integration of single-cell datasets using deep generative model paired with adversarial classifier","volume":"14","author":"A Shree","year":"2023","journal-title":"Nat Commun"},{"issue":"8","key":"pcbi.1014008.ref013","doi-asserted-by":"crossref","first-page":"715","DOI":"10.1038\/s41592-019-0494-8","article-title":"scGen predicts single-cell perturbation responses","volume":"16","author":"M Lotfollahi","year":"2019","journal-title":"Nat Methods"},{"issue":"10","key":"pcbi.1014008.ref014","doi-asserted-by":"crossref","first-page":"607","DOI":"10.1038\/s42256-020-00233-7","article-title":"Iterative transfer learning with neural network for clustering and cell type classification in single-cell RNA-seq analysis","volume":"2","author":"J Hu","year":"2020","journal-title":"Nat Mach Intell"},{"issue":"1","key":"pcbi.1014008.ref015","doi-asserted-by":"crossref","first-page":"872","DOI":"10.1038\/s41467-024-45240-z","article-title":"Semi-supervised integration of single-cell transcriptomics data","volume":"15","author":"M Andreatta","year":"2024","journal-title":"Nat Commun"},{"key":"pcbi.1014008.ref016","doi-asserted-by":"crossref","first-page":"961","DOI":"10.1016\/j.csbj.2021.01.015","article-title":"Automated methods for cell type annotation on scRNA-seq data","volume":"19","author":"G Pasquini","year":"2021","journal-title":"Comput Struct Biotechnol J"},{"issue":"1","key":"pcbi.1014008.ref017","doi-asserted-by":"crossref","first-page":"194","DOI":"10.1186\/s13059-019-1795-z","article-title":"A comparison of automatic cell identification methods for single-cell RNA sequencing data","volume":"20","author":"T Abdelaal","year":"2019","journal-title":"Genome Biol"},{"key":"pcbi.1014008.ref018","doi-asserted-by":"crossref","first-page":"3796","DOI":"10.1016\/j.csbj.2021.06.043","article-title":"UCell: Robust and scalable single-cell gene signature scoring","volume":"19","author":"M Andreatta","year":"2021","journal-title":"Comput Struct Biotechnol J"},{"issue":"7","key":"pcbi.1014008.ref019","doi-asserted-by":"crossref","first-page":"1888","DOI":"10.1016\/j.cell.2019.05.031","article-title":"Comprehensive Integration of Single-Cell Data","volume":"177","author":"T Stuart","year":"2019","journal-title":"Cell"},{"issue":"12","key":"pcbi.1014008.ref020","doi-asserted-by":"crossref","first-page":"1053","DOI":"10.1038\/s41592-018-0229-2","article-title":"Deep generative modeling for single-cell transcriptomics","volume":"15","author":"R Lopez","year":"2018","journal-title":"Nat Methods"},{"issue":"12","key":"pcbi.1014008.ref021","doi-asserted-by":"crossref","first-page":"1289","DOI":"10.1038\/s41592-019-0619-0","article-title":"Fast, sensitive and accurate integration of single-cell data with Harmony","volume":"16","author":"I Korsunsky","year":"2019","journal-title":"Nat Methods"},{"issue":"6","key":"pcbi.1014008.ref022","doi-asserted-by":"crossref","first-page":"685","DOI":"10.1038\/s41587-019-0113-3","article-title":"Efficient integration of heterogeneous single-cell transcriptomes using Scanorama","volume":"37","author":"B Hie","year":"2019","journal-title":"Nat Biotechnol"},{"key":"pcbi.1014008.ref023","first-page":"2024","article-title":"Partially characterized topology guides reliable anchor-free scRNA integration","author":"C He","year":"2024","journal-title":"bioRxiv"},{"issue":"6","key":"pcbi.1014008.ref024","doi-asserted-by":"crossref","first-page":"106877","DOI":"10.1016\/j.isci.2023.106877","article-title":"A comprehensive mouse kidney atlas enables rare cell population characterization and robust marker discovery","volume":"26","author":"C Novella-Rausell","year":"2023","journal-title":"iScience"},{"issue":"6","key":"pcbi.1014008.ref025","doi-asserted-by":"crossref","first-page":"1563","DOI":"10.1038\/s41591-023-02327-2","article-title":"An integrated cell atlas of the lung in health and disease","volume":"29","author":"L Sikkema","year":"2023","journal-title":"Nature Medicine"},{"issue":"15","key":"pcbi.1014008.ref026","doi-asserted-by":"crossref","first-page":"1970","DOI":"10.3390\/cells12151970","article-title":"A Review of Single-Cell RNA-Seq Annotation, Integration, and Cell-Cell Communication","volume":"12","author":"C Cheng","year":"2023","journal-title":"Cells"},{"issue":"6","key":"pcbi.1014008.ref027","doi-asserted-by":"crossref","first-page":"2749","DOI":"10.1038\/s41596-021-00534-0","article-title":"Tutorial: guidelines for annotating single-cell transcriptomic maps using automated and manual methods","volume":"16","author":"ZA Clarke","year":"2021","journal-title":"Nat Protoc"},{"issue":"9","key":"pcbi.1014008.ref028","doi-asserted-by":"crossref","first-page":"2642","DOI":"10.1093\/bioinformatics\/btac141","article-title":"scGate: marker-based purification of cell types from heterogeneous single-cell RNA-seq datasets","volume":"38","author":"M Andreatta","year":"2022","journal-title":"Bioinformatics"},{"issue":"13","key":"pcbi.1014008.ref029","doi-asserted-by":"crossref","first-page":"3573","DOI":"10.1016\/j.cell.2021.04.048","article-title":"Integrated analysis of multimodal single-cell data","volume":"184","author":"Y Hao","year":"2021","journal-title":"Cell"},{"issue":"10","key":"pcbi.1014008.ref030","doi-asserted-by":"crossref","first-page":"1007","DOI":"10.1038\/s41592-019-0529-1","article-title":"Probabilistic cell-type assignment of single-cell RNA-seq for tumor microenvironment profiling","volume":"16","author":"AW Zhang","year":"2019","journal-title":"Nat Methods"},{"issue":"2","key":"pcbi.1014008.ref031","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1038\/s41587-023-01767-y","article-title":"Dictionary learning for integrative, multimodal and scalable single-cell analysis","volume":"42","author":"Y Hao","year":"2024","journal-title":"Nat Biotechnol"},{"issue":"6","key":"pcbi.1014008.ref032","doi-asserted-by":"crossref","first-page":"1653","DOI":"10.1016\/j.celrep.2018.07.025","article-title":"Single-Cell Transcriptomes Distinguish Stem Cell State Changes and Lineage Specification Programs in Early Mammary Gland Development","volume":"24","author":"RR Giraddi","year":"2018","journal-title":"Cell Rep"},{"issue":"1","key":"pcbi.1014008.ref033","doi-asserted-by":"crossref","first-page":"1627","DOI":"10.1038\/s41467-017-01560-x","article-title":"Construction of developmental lineage relationships in the mouse mammary gland by single-cell RNA profiling","volume":"8","author":"B Pal","year":"2017","journal-title":"Nat Commun"},{"issue":"1","key":"pcbi.1014008.ref034","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41467-017-02001-5","article-title":"Differentiation dynamics of mammary epithelial cells revealed by single-cell RNA sequencing","volume":"8","author":"K Bach","year":"2017","journal-title":"Nature Communications"},{"issue":"5","key":"pcbi.1014008.ref035","article-title":"Molecular Classification and Comparative Taxonomics of Foveal and Peripheral Cells in Primate Retina","volume":"176","author":"Y-R Peng","year":"2019","journal-title":"Cell"},{"issue":"1","key":"pcbi.1014008.ref036","doi-asserted-by":"crossref","first-page":"960","DOI":"10.1038\/s41467-023-36635-5","article-title":"Batch alignment of single-cell transcriptomics data using deep metric learning","volume":"14","author":"X Yu","year":"2023","journal-title":"Nat Commun"},{"issue":"7","key":"pcbi.1014008.ref037","doi-asserted-by":"crossref","first-page":"1153","DOI":"10.1038\/s41591-019-0468-5","article-title":"A cellular census of human lungs identifies novel cell states in health and in asthma","volume":"25","author":"FA Vieira Braga","year":"2019","journal-title":"Nat Med"},{"issue":"1","key":"pcbi.1014008.ref038","doi-asserted-by":"crossref","first-page":"121","DOI":"10.1038\/s41587-021-01001-7","article-title":"Mapping single-cell data to reference atlases by transfer learning","volume":"40","author":"M Lotfollahi","year":"2022","journal-title":"Nat Biotechnol"},{"issue":"1","key":"pcbi.1014008.ref039","doi-asserted-by":"crossref","first-page":"193","DOI":"10.1007\/BF01908075","article-title":"Comparing partitions","volume":"2","author":"L Hubert","year":"1985","journal-title":"J Classif"},{"key":"pcbi.1014008.ref040","first-page":"2825","article-title":"Scikit-learn: Machine Learning in Python","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"issue":"1","key":"pcbi.1014008.ref041","doi-asserted-by":"crossref","first-page":"43","DOI":"10.1038\/s41592-018-0254-1","article-title":"A test metric for assessing single-cell RNA-seq batch correction","volume":"16","author":"M B\u00fcttner","year":"2019","journal-title":"Nat Methods"},{"issue":"16","key":"pcbi.1014008.ref042","doi-asserted-by":"crossref","first-page":"3942","DOI":"10.1093\/bioinformatics\/btac427","article-title":"ResPAN: a powerful batch correction model for scRNA-seq data through residual adversarial networks","volume":"38","author":"Y Wang","year":"2022","journal-title":"Bioinformatics"},{"issue":"1","key":"pcbi.1014008.ref043","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1186\/s13059-021-02280-8","article-title":"iMAP: integration of multiple single-cell datasets by adversarial paired transfer networks","volume":"22","author":"D Wang","year":"2021","journal-title":"Genome Biol"},{"issue":"2","key":"pcbi.1014008.ref044","doi-asserted-by":"crossref","first-page":"163","DOI":"10.1038\/s41590-018-0276-y","article-title":"Reference-based analysis of lung single-cell sequencing reveals a transitional profibrotic macrophage","volume":"20","author":"D Aran","year":"2019","journal-title":"Nat Immunol"},{"issue":"1","key":"pcbi.1014008.ref045","article-title":"SCANPY: large-scale single-cell gene expression data analysis","volume":"19","author":"FA Wolf","year":"2018","journal-title":"Genome Biol"}],"updated-by":[{"DOI":"10.1371\/journal.pcbi.1014008","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,3,26]],"date-time":"2026-03-26T00:00:00Z","timestamp":1774483200000}}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1014008","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,26]],"date-time":"2026-03-26T17:53:01Z","timestamp":1774547581000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1014008"}},"subtitle":[],"editor":[{"given":"Tao","family":"Wang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,3,16]]},"references-count":45,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3,16]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1014008","relation":{},"ISSN":["1553-7358"],"issn-type":[{"value":"1553-7358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,16]]}}}