{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T09:05:11Z","timestamp":1784797511188,"version":"3.55.0"},"reference-count":34,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T00:00:00Z","timestamp":1784764800000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["31460297"],"award-info":[{"award-number":["31460297"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Yunnan Province Key Laboratory of Digital-Intelligent Ethnic Computing","award":["20254916CE340003"],"award-info":[{"award-number":["20254916CE340003"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Spatial transcriptomics provides unprecedented insights into gene expression in the spatial context of tissues; however, some mainstream techniques lack quantitative analysis of uncertainty.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      In this study, we propose a computational framework\u2014Spatial Deconvolution\n                      <jats:italic>via<\/jats:italic>\n                      Deep Gaussian Processes (SDDGP)\u2014that integrates deep Gaussian processes with Bayesian inference for spatially aware deconvolution of spatial transcriptomics data. The framework consists of three tightly coupled components. A multi-layer deep Gaussian process (DGP) captures nonlinear spatial dependencies in cell-type composition by propagating spatial coordinates through successive Gaussian process (GP) layers, each equipped with sparse variational inducing points and a radial basis function (RBF) kernel that can be combined with an external Matern spatial kernel. The DGP output serves as an adaptive, spatially informed prior for per-spot cell-type proportions, which are then estimated\n                      <jats:italic>via<\/jats:italic>\n                      a Negative-Binomial Markov chain Monte Carlo (MCMC) sampler with automatic proposal tuning, thinning, and split-chain convergence diagnostics.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We compared our method with existing approaches on two datasets: a pancreatic ductal adenocarcinoma (PDAC) dataset with matched single-cell RNA sequencing and spatial transcriptomics, and a human thymus dataset. Across both datasets, our method estimated cell type proportions more accurately than existing methods, particularly for rare cell types, and recovered biologically meaningful spatial patterns such as the distribution of T cells and plasma cells in PDAC. The Bayesian formulation also provides a quantification of uncertainty for each estimate, which aids the interpretation of spatial heterogeneity in complex tissues.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.7717\/peerj.21548","type":"journal-article","created":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T08:42:27Z","timestamp":1784796147000},"page":"e21548","source":"Crossref","is-referenced-by-count":0,"title":["A deep learning representation and spatial Bayesian cell-type deconvolution for spatial transcriptomics"],"prefix":"10.7717","volume":"14","author":[{"given":"Xiao","family":"Yang","sequence":"first","affiliation":[{"name":"School of Mathematics and Computer Science, Yunnan Minzu University, Kunming, Yunnan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanbin","family":"Feng","sequence":"additional","affiliation":[{"name":"School of Mathematics and Computer Science, Yunnan Minzu University, Kunming, Yunnan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanfang","family":"Zhao","sequence":"additional","affiliation":[{"name":"School of Mathematics and Computer Science, Yunnan Minzu University, Kunming, Yunnan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Huamei","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Rheumatology and Immunology, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, Jiangsu, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaozhou","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Mathematics and Computer Science, Yunnan Minzu University, Kunming, Yunnan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"4443","published-online":{"date-parts":[[2026,7,23]]},"reference":[{"key":"10.7717\/peerj.21548\/ref-1","doi-asserted-by":"publisher","first-page":"565","DOI":"10.1038\/s42003-020-01247-y","article-title":"Single-cell and spatial transcriptomics enables probabilistic inference of cell type topography","volume":"3","author":"Andersson","year":"2020","journal-title":"Communications Biology"},{"key":"10.7717\/peerj.21548\/ref-2","doi-asserted-by":"publisher","first-page":"1352","DOI":"10.1038\/s41592-021-01264-7","article-title":"Deep learning and alignment of spatially resolved single-cell transcriptomes with tangram","volume":"18","author":"Biancalani","year":"2021","journal-title":"Nature Methods"},{"key":"10.7717\/peerj.21548\/ref-3","doi-asserted-by":"publisher","first-page":"317","DOI":"10.1038\/s41576-019-0129-z","article-title":"Spatial transcriptomics coming of age","volume":"20","author":"Burgess","year":"2019","journal-title":"Nature Reviews Genetics"},{"key":"10.7717\/peerj.21548\/ref-4","doi-asserted-by":"publisher","first-page":"517","DOI":"10.1038\/s41587-021-00830-w","article-title":"Robust decomposition of cell type mixtures in spatial transcriptomics","volume":"40","author":"Cable","year":"2022","journal-title":"Nature Biotechnology"},{"issue":"1","key":"10.7717\/peerj.21548\/ref-5","doi-asserted-by":"publisher","first-page":"btae747","DOI":"10.1093\/bioinformatics\/btae747","article-title":"NLSDeconv: an efficient cell-type deconvolution method for spatial transcriptomics data","volume":"41","author":"Chen","year":"2024","journal-title":"Bioinformatics"},{"key":"10.7717\/peerj.21548\/ref-6","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1186\/s13059-021-02286-2","article-title":"Giotto: a toolbox for integrative analysis and visualization of spatial expression data","volume":"22","author":"Dries","year":"2021","journal-title":"Genome Biology"},{"key":"10.7717\/peerj.21548\/ref-7","doi-asserted-by":"publisher","first-page":"e50","DOI":"10.1093\/nar\/gkab043","article-title":"SPOTlight: seeded NMF regression to deconvolute spatial transcriptomics spots with single-cell transcriptomes","volume":"49","author":"Elosua-Bayes","year":"2021","journal-title":"Nucleic Acids Research"},{"issue":"5","key":"10.7717\/peerj.21548\/ref-8","doi-asserted-by":"publisher","first-page":"439","DOI":"10.1093\/chemse\/bju015","article-title":"Activity-dependent genes in mouse olfactory sensory neurons","volume":"39","author":"Fischl","year":"2014","journal-title":"Chemical Senses"},{"key":"10.7717\/peerj.21548\/ref-9","doi-asserted-by":"publisher","first-page":"103","DOI":"10.1159\/000237126","article-title":"lnterleukin-12","volume":"108","author":"Germann","year":"2009","journal-title":"International Archives of Allergy and Immunology"},{"key":"10.7717\/peerj.21548\/ref-10","doi-asserted-by":"publisher","first-page":"1729","DOI":"10.1016\/j.cell.2023.02.014","article-title":"Pancreatic cancer: advances and challenges","volume":"186","author":"Halbrook","year":"2023","journal-title":"Cell"},{"key":"10.7717\/peerj.21548\/ref-11","doi-asserted-by":"publisher","first-page":"97","DOI":"10.1093\/biomet\/57.1.97","article-title":"Monte Carlo sampling methods using Markov chains and their applications","volume":"57","author":"Hastings","year":"1970","journal-title":"Biometrika"},{"key":"10.7717\/peerj.21548\/ref-12","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1038\/s41416-019-0479-5","article-title":"Tumour microenvironment of pancreatic cancer: immune landscape is dictated by molecular and histopathological features","volume":"121","author":"Karamitopoulou","year":"2019","journal-title":"British Journal of Cancer"},{"key":"10.7717\/peerj.21548\/ref-13","doi-asserted-by":"publisher","first-page":"377","DOI":"10.1038\/nri3667","article-title":"Positive and negative selection of the T cell repertoire: what thymocytes see (and don\u2019t see)","volume":"14","author":"Klein","year":"2014","journal-title":"Nature Reviews Immunology"},{"key":"10.7717\/peerj.21548\/ref-14","doi-asserted-by":"publisher","first-page":"661","DOI":"10.1038\/s41587-021-01139-4","article-title":"Cell2location maps fine-grained cell types in spatial transcriptomics","volume":"40","author":"Kleshchevnikov","year":"2022","journal-title":"Nature Biotechnology"},{"key":"10.7717\/peerj.21548\/ref-15","doi-asserted-by":"publisher","first-page":"7784","DOI":"10.1038\/s41467-024-51767-y","article-title":"Unraveling the spatial organization and development of human thymocytes through integration of spatial transcriptomics and single-cell multi-omics profiling","volume":"15","author":"Li","year":"2024","journal-title":"Nature Communications"},{"key":"10.7717\/peerj.21548\/ref-16","doi-asserted-by":"publisher","first-page":"49","DOI":"10.1186\/s13073-021-00861-7","article-title":"Development of double-positive thymocytes at single-cell resolution","volume":"13","author":"Li","year":"2021","journal-title":"Genome Medicine"},{"key":"10.7717\/peerj.21548\/ref-17","doi-asserted-by":"publisher","first-page":"662","DOI":"10.1038\/s41592-022-01480-9","article-title":"Benchmarking spatial and single-cell transcriptomics integration methods for transcript distribution prediction and cell type deconvolution","volume":"19","author":"Li","year":"2022","journal-title":"Nature Methods"},{"key":"10.7717\/peerj.21548\/ref-18","doi-asserted-by":"publisher","first-page":"1349","DOI":"10.1038\/s41587-022-01273-7","article-title":"Spatially informed cell-type deconvolution for spatial transcriptomics","volume":"40","author":"Ma","year":"2022","journal-title":"Nature Biotechnology"},{"key":"10.7717\/peerj.21548\/ref-19","doi-asserted-by":"publisher","first-page":"2339","DOI":"10.1038\/s41467-022-30033-z","article-title":"Reference-free cell type deconvolution of multi-cellular pixel-resolution spatially resolved transcriptomics data","volume":"13","author":"Miller","year":"2022","journal-title":"Nature Communications"},{"key":"10.7717\/peerj.21548\/ref-20","doi-asserted-by":"publisher","first-page":"333","DOI":"10.1038\/s41587-019-0392-8","article-title":"Integrating microarray-based spatial transcriptomics and single-cell RNA-seq reveals tissue architecture in pancreatic ductal adenocarcinomas","volume":"38","author":"Moncada","year":"2020","journal-title":"Nature Biotechnology"},{"key":"10.7717\/peerj.21548\/ref-21","doi-asserted-by":"publisher","first-page":"534","DOI":"10.1038\/s41592-022-01409-2","article-title":"Museum of spatial transcriptomics","volume":"19","author":"Moses","year":"2022","journal-title":"Nature Methods"},{"key":"10.7717\/peerj.21548\/ref-22","doi-asserted-by":"publisher","DOI":"10.1101\/2020.05.31.125658","article-title":"stLearn: integrating spatial location, tissue morphology and gene expression to find cell types, cell\u2013cell interactions and spatial trajectories within undissociated tissues","author":"Pham","year":"2020","journal-title":"bioRxiv"},{"key":"10.7717\/peerj.21548\/ref-23","doi-asserted-by":"publisher","first-page":"1227","DOI":"10.1038\/nn.2923","article-title":"Heterogeneity of CNS myeloid cells and their roles in neurodegeneration","volume":"14","author":"Prinz","year":"2011","journal-title":"Nature Neuroscience"},{"key":"10.7717\/peerj.21548\/ref-24","doi-asserted-by":"publisher","first-page":"37","DOI":"10.1016\/j.immuni.2018.12.027","article-title":"Plasmacytoid dendritic cells: development, regulation, and function","volume":"50","author":"Reizis","year":"2019","journal-title":"Immunity"},{"key":"10.7717\/peerj.21548\/ref-25","doi-asserted-by":"publisher","first-page":"351","DOI":"10.1214\/ss\/1015346320","article-title":"Optimal scaling for various Metropolis\u2013Hastings algorithms","volume":"16","author":"Roberts","year":"2001","journal-title":"Statistical Science"},{"key":"10.7717\/peerj.21548\/ref-26","doi-asserted-by":"publisher","first-page":"1463","DOI":"10.1126\/science.aaw1219","article-title":"Slide-seq: a scalable technology for measuring genome-wide expression at high spatial resolution","volume":"363","author":"Rodriques","year":"2019","journal-title":"Science"},{"issue":"6444","key":"10.7717\/peerj.21548\/ref-27","doi-asserted-by":"publisher","first-page":"eaas9536","DOI":"10.1126\/science.aas9536","article-title":"Spatiotemporal structure of cell fate decisions in murine neural crest","volume":"364","author":"Soldatov","year":"2019","journal-title":"Science"},{"key":"10.7717\/peerj.21548\/ref-28","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1126\/science.aaf2403","article-title":"Visualization and analysis of gene expression in tissue sections by spatial transcriptomics","volume":"353","author":"St\u00e5hl","year":"2016","journal-title":"Science"},{"key":"10.7717\/peerj.21548\/ref-29","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1146\/annurev.immunol.26.021607.090248","article-title":"Mechanism and regulation of class switch recombination","volume":"26","author":"Stavnezer","year":"2008","journal-title":"Annual Review of Immunology"},{"key":"10.7717\/peerj.21548\/ref-30","doi-asserted-by":"publisher","first-page":"1888","DOI":"10.1016\/j.cell.2019.05.031","article-title":"Comprehensive integration of single-cell data","volume":"177","author":"Stuart","year":"2019","journal-title":"Cell"},{"key":"10.7717\/peerj.21548\/ref-31","doi-asserted-by":"publisher","first-page":"343","DOI":"10.1038\/nmeth.4636","article-title":"SpatialDE: identification of spatially variable genes","volume":"15","author":"Svensson","year":"2018","journal-title":"Nature Methods"},{"key":"10.7717\/peerj.21548\/ref-32","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1186\/s13073-024-01283-x","article-title":"Unsupervised spatially embedded deep representation of spatial transcriptomics","volume":"16","author":"Xu","year":"2024","journal-title":"Genome Medicine"},{"key":"10.7717\/peerj.21548\/ref-33","doi-asserted-by":"publisher","first-page":"7561","DOI":"10.1038\/s41467-024-51891-9","article-title":"scCAD: Cluster decomposition-based anomaly detection for rare cell identification in single-cell expression data","volume":"15","author":"Xu","year":"2024","journal-title":"Nature Communications"},{"key":"10.7717\/peerj.21548\/ref-34","doi-asserted-by":"publisher","first-page":"7930","DOI":"10.1038\/s41467-023-43600-9","article-title":"Spatial transcriptomics deconvolution at single-cell resolution using Redeconve","volume":"14","author":"Zhou","year":"2023","journal-title":"Nature Communications"}],"container-title":["PeerJ"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/21548.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/21548.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/21548.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/21548.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T08:42:30Z","timestamp":1784796150000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/21548"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,23]]},"references-count":34,"alternative-id":["10.7717\/peerj.21548"],"URL":"https:\/\/doi.org\/10.7717\/peerj.21548","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2167-8359"],"issn-type":[{"value":"2167-8359","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,23]]},"article-number":"e21548"}}