{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,2]],"date-time":"2026-10-02T18:17:23Z","timestamp":1790965043626,"version":"4.1.0"},"reference-count":66,"publisher":"Oxford University Press (OUP)","issue":"Supplement_1","license":[{"start":{"date-parts":[[2022,6,27]],"date-time":"2022-06-27T00:00:00Z","timestamp":1656288000000},"content-version":"vor","delay-in-days":3,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"NSERC Alexander Graham Bell Canadian Graduate Scholarship","award":["CZF2019-002436"],"award-info":[{"award-number":["CZF2019-002436"]}]},{"name":"Chan Zuckerberg Foundation"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,6,24]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Motivation<\/jats:title><jats:p>Advances in bioimaging now permit in situ proteomic characterization of cell\u2013cell interactions in complex tissues, with important applications across a spectrum of biological problems from development to disease. These methods depend on selection of antibodies targeting proteins that are expressed specifically in particular cell types. Candidate marker proteins are often identified from single-cell transcriptomic data, with variable rates of success, in part due to divergence between expression levels of proteins and the genes that encode them. In principle, marker identification could be improved by using existing databases of immunohistochemistry for thousands of antibodies in human tissue, such as the Human Protein Atlas. However, these data lack detailed annotations of the types of cells in each image.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>We develop a method to predict cell type specificity of protein markers from unlabeled images. We train a convolutional neural network with a self-supervised objective to generate embeddings of the images. Using non-linear dimensionality reduction, we observe that the model clusters images according to cell types and anatomical regions for which the stained proteins are specific. We then use estimates of cell type specificity derived from an independent single-cell transcriptomics dataset to train an image classifier, without requiring any human labelling of images. Our scheme demonstrates superior classification of known proteomic markers in kidney compared to selection via single-cell transcriptomics.<\/jats:p><\/jats:sec><jats:sec><jats:title>Availability and implementation<\/jats:title><jats:p>Code and trained model are available at www.github.com\/murphy17\/HPA-SimCLR.<\/jats:p><\/jats:sec><jats:sec><jats:title>Supplementary information<\/jats:title><jats:p>Supplementary data are available at Bioinformatics online.<\/jats:p><\/jats:sec>","DOI":"10.1093\/bioinformatics\/btac263","type":"journal-article","created":{"date-parts":[[2022,4,14]],"date-time":"2022-04-14T11:10:15Z","timestamp":1649934615000},"page":"i395-i403","source":"Crossref","is-referenced-by-count":6,"title":["Self-supervised learning of cell type specificity from immunohistochemical images"],"prefix":"10.1093","volume":"38","author":[{"given":"Michael","family":"Murphy","sequence":"first","affiliation":[{"name":"Department of Biological Engineering, Massachusetts Institute of Technology , Cambridge, MA, USA"},{"name":"Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology , Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stefanie","family":"Jegelka","sequence":"additional","affiliation":[{"name":"Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology , Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ernest","family":"Fraenkel","sequence":"additional","affiliation":[{"name":"Department of Biological Engineering, Massachusetts Institute of Technology , Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2022,6,27]]},"reference":[{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"1609","DOI":"10.1038\/s41467-021-21727-x","article-title":"Joint analysis of expression levels and histological images identifies genes associated with tissue morphology","volume":"12","author":"Ash","year":"2021","journal-title":"Nat. Commun"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"e0242858","DOI":"10.1371\/journal.pone.0242858","article-title":"Identifying transcriptomic correlates of histology using deep learning","volume":"15","author":"Badea","year":"2020","journal-title":"PLoS One"},{"key":"2023041407562153900_","author":"Biohub","year":"2022"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"111","DOI":"10.4103\/0973-029X.141346","article-title":"Artefacts in histopathology","volume":"18","author":"Chatterjee","year":"2014","journal-title":"J. Oral Maxillofac. Pathol"},{"key":"2023041407562153900_","volume-title":"Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6\u201312, 2020, Vancouver, BC, Canada","author":"Chen","year":"2020"},{"key":"2023041407562153900_","first-page":"1597","author":"Chen","year":"2020"},{"key":"2023041407562153900_","volume-title":"Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6\u201312, 2020, Vancouver, BC, Canada","author":"Chuang","year":"2020"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"e9005","DOI":"10.15252\/msb.20199005","article-title":"Combinatorial prediction of marker panels from single-cell transcriptomic data","volume":"15","author":"Delaney","year":"2019","journal-title":"Mol. Syst. Biol"},{"key":"2023041407562153900_","volume-title":"2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA.","author":"Deng","year":"2009"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"130","DOI":"10.1186\/s13059-020-02048-6","article-title":"Systematic assessment of tissue dissociation and storage biases in single-cell and single-nucleus RNA-seq workflows","volume":"21","author":"Denisenko","year":"2020","journal-title":"Genome Biol"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"737","DOI":"10.1038\/s41587-020-0465-8","article-title":"Systematic comparison of single-cell and single-nucleus RNA-sequencing methods","volume":"38","author":"Ding","year":"2020","journal-title":"Nat. Biotechnol"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"1734","DOI":"10.1109\/TPAMI.2015.2496141","article-title":"Discriminative unsupervised feature learning with exemplar convolutional neural networks","volume":"38","author":"Dosovitskiy","year":"2016","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"1186","DOI":"10.1038\/s41467-021-21453-4","article-title":"Optimal marker gene selection for cell type discrimination in single cell analyses","volume":"12","author":"Dumitrascu","year":"2021","journal-title":"Nat. Commun"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"100140","DOI":"10.1016\/j.mcpro.2021.100140","article-title":"DeepHistoClass: a novel strategy for confident classification of immunohistochemistry images using deep learning","volume":"20","author":"Ghoshal","year":"2021","journal-title":"Mol. Cell. Proteomics"},{"key":"2023041407562153900_","author":"Gidaris","year":"2018"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"417","DOI":"10.1038\/nmeth.2869","article-title":"Highly multiplexed imaging of tumor tissues with subcellular resolution by mass cytometry","volume":"11","author":"Giesen","year":"2014","journal-title":"Nat. Methods"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"968","DOI":"10.1016\/j.cell.2018.07.010","article-title":"Deep profiling of mouse splenic architecture with CODEX multiplexed imaging","volume":"174","author":"Goltsev","year":"2018","journal-title":"Cell"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"2776","DOI":"10.1038\/s41598-017-03057-5","article-title":"Single-cell protein-mRNA correlation analysis enabled by multiplexed dual-analyte co-detection","volume":"7","author":"Gong","year":"2017","journal-title":"Sci. Rep"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"eaar7042","DOI":"10.1126\/science.aar7042","article-title":"Multiplexed protein maps link subcellular organization to cellular states","volume":"361","author":"Gut","year":"2018","journal-title":"Science"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"e116125","DOI":"10.1371\/journal.pone.0116125","article-title":"The kidney transcriptome and proteome defined by transcriptomics and antibody-based profiling","volume":"9","author":"Habuka","year":"2014","journal-title":"PLoS One"},{"key":"2023041407562153900_","first-page":"9726","volume-title":"2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, WA, USA, June 13\u201319, 2020, Seattle, WA, USA","author":"He","year":"2020"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"493","DOI":"10.1002\/prot.26244","article-title":"Incorporating label correlations into deep neural networks to classify protein subcellular location patterns in immunohistochemistry images","volume":"90","author":"Hu","year":"2022","journal-title":"Proteins Struct. Funct. Bioinf"},{"key":"2023041407562153900_","first-page":"2261","volume-title":"2017 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, Honolulu, HI, USA, July 21\u201326, 2017, Honolulu, HI, USA","author":"Huang","year":"2017"},{"key":"2023041407562153900_","first-page":"299","volume-title":"Methods in Molecular Biology","author":"Im","year":"2018"},{"key":"2023041407562153900_","first-page":"3620","article-title":"Production of tissue microarrays, immunohistochemistry staining and digitalization within the human protein atlas","volume":"63","author":"Kampf","year":"2012","journal-title":"J. Vis. Exp"},{"key":"2023041407562153900_","author":"Kingma","year":"2015"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"483","DOI":"10.1038\/nmeth.4236","article-title":"SC3: consensus clustering of single-cell RNA-seq data","volume":"14","author":"Kiselev","year":"2017","journal-title":"Nat. Methods"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"1664","DOI":"10.2215\/CJN.13791215","article-title":"The players: cells involved in glomerular disease","volume":"11","author":"Kitching","year":"2016","journal-title":"Clin. J. Am. Soc. Nephrol"},{"key":"2023041407562153900_","article-title":"Self-supervised deep learning encodes high-resolution features of protein subcellular localization","author":"Kobayashi","year":"2021","journal-title":"bioRxiv"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"34","DOI":"10.1016\/j.csbj.2018.01.001","article-title":"Machine learning methods for histopathological image analysis","volume":"16","author":"Komura","year":"2018","journal-title":"Comput. Struct. Biotechnol. J"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"765","DOI":"10.1109\/JBHI.2013.2276766","article-title":"Removing batch effects from histopathological images for enhanced cancer diagnosis","volume":"18","author":"Kothari","year":"2014","journal-title":"IEEE J. Biomed. Health Inform"},{"key":"2023041407562153900_","year":"2021"},{"key":"2023041407562153900_","article-title":"An atlas of healthy and injured cell states and niches in the human kidney","author":"Lake","year":"2021","journal-title":"bioRxiv"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"e50514","DOI":"10.1371\/journal.pone.0050514","article-title":"Automated analysis and reannotation of subcellular locations in confocal images from the human protein atlas","volume":"7","author":"Li","year":"2012","journal-title":"PLoS One"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"8390","DOI":"10.1038\/ncomms9390","article-title":"Highly multiplexed imaging of single cells using a high-throughput cyclic immunofluorescence method","volume":"6","author":"Lin","year":"2015","journal-title":"Nat. Commun"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"2244","DOI":"10.1093\/bioinformatics\/btz909","article-title":"ImPLoc: a multi-instance deep learning model for the prediction of protein subcellular localization based on immunohistochemistry images","volume":"36","author":"Long","year":"2020","journal-title":"Bioinformatics"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"e1007348","DOI":"10.1371\/journal.pcbi.1007348","article-title":"Learning unsupervised feature representations for single cell microscopy images with paired cell inpainting","volume":"15","author":"Lu","year":"2019","journal-title":"PLoS Comput. Biol"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"e8746","DOI":"10.15252\/msb.20188746","article-title":"Current best practices in single-cell RNA-seq analysis: a tutorial","volume":"15","author":"Luecken","year":"2019","journal-title":"Mol. Syst. Biol"},{"key":"2023041407562153900_","author":"McInnes","year":"2018"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"6","DOI":"10.1186\/1471-2105-12-6","article-title":"Logical development of the cell ontology","volume":"12","author":"Meehan","year":"2011","journal-title":"BMC Bioinformatics"},{"key":"2023041407562153900_","first-page":"6706","volume-title":"2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, WA, USA, June 13\u201319, 2020, Seattle, WA, USA","author":"Misra","year":"2020"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"2190","DOI":"10.1038\/s41467-021-22368-w","article-title":"Single cell transcriptional and chromatin accessibility profiling redefine cellular heterogeneity in the adult human kidney","volume":"12","author":"Muto","year":"2021","journal-title":"Nat. Commun"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"2300","DOI":"10.1021\/pr7007626","article-title":"A framework for the automated analysis of subcellular patterns in human protein atlas images","volume":"7","author":"Newberg","year":"2008","journal-title":"J. Proteome Res"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"1254","DOI":"10.1038\/s41592-019-0658-6","article-title":"Analysis of the human protein atlas image classification competition","volume":"16","author":"Ouyang","year":"2019","journal-title":"Nat. Methods"},{"key":"2023041407562153900_","first-page":"8024","author":"Paszke","year":"2019"},{"key":"2023041407562153900_","first-page":"2536","volume-title":"2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, NV, USA, June 27-30, 2016, Las Vegas, NV, USA","author":"Pathak","year":"2016"},{"key":"2023041407562153900_","first-page":"2825","article-title":"Scikit-learn: machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"9","DOI":"10.1016\/j.cbpa.2021.04.001","article-title":"Image-based cell phenotyping with deep learning","volume":"65","author":"Pratapa","year":"2021","journal-title":"Curr. Opin. Chem. Biol"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"3228","DOI":"10.1093\/bioinformatics\/btab257","article-title":"Identification of cell-type-specific marker genes from co-expression patterns in tissue samples","volume":"37","author":"Qiu","year":"2021","journal-title":"Bioinformatics"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"e27041","DOI":"10.7554\/eLife.27041","article-title":"The human cell atlas","volume":"6","author":"Regev","year":"2017","journal-title":"eLife"},{"key":"2023041407562153900_","first-page":"3663","volume-title":"IEEE Winter Conference on Applications of Computer Vision, WACV 2020, Snowmass Village, CO, USA, March 1-5, 2020, Snowglass Village, CO, USA","author":"Riba","year":"2020"},{"key":"2023041407562153900_","author":"Robinson","year":"2021"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41536-021-00156-w","article-title":"Defining the variety of cell types in developing and adult human kidneys by single-cell RNA sequencing","volume":"6","author":"Schumacher","year":"2021","journal-title":"NPJ Regen. Med"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"1341","DOI":"10.1016\/j.cell.2020.07.005","article-title":"Coordinated cellular neighborhoods orchestrate antitumoral immunity at the colorectal cancer invasive front","volume":"182","author":"Sch\u00fcrch","year":"2020","journal-title":"Cell"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"1791","DOI":"10.1172\/JCI42344","article-title":"Uric acid transport and disease","volume":"120","author":"So","year":"2010","journal-title":"J. Clin. Invest"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"865","DOI":"10.1038\/nmeth.4380","article-title":"Simultaneous epitope and transcriptome measurement in single cells","volume":"14","author":"Stoeckius","year":"2017","journal-title":"Nat. Methods"},{"key":"2023041407562153900_","article-title":"Hierarchical multi-scale attention for semantic segmentation","author":"Tao","year":"2020","journal-title":"CoRR"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"101544","DOI":"10.1016\/j.media.2019.101544","article-title":"Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology","volume":"58","author":"Tellez","year":"2019","journal-title":"Med. Image Anal"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"820","DOI":"10.1126\/science.aal3321","article-title":"A subcellular map of the human proteome","volume":"356","author":"Thul","year":"2017","journal-title":"Science"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"1260419","DOI":"10.1126\/science.1260419","article-title":"Tissue-based map of the human proteome","volume":"347","author":"Uhl\u00e9n","year":"2015","journal-title":"Science"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"823","DOI":"10.1038\/nmeth.3995","article-title":"A proposal for validation of antibodies","volume":"13","author":"Uhlen","year":"2016","journal-title":"Nat. Methods"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"4285","DOI":"10.1210\/endo.137.10.8828488","article-title":"Distribution of parathyroid hormone-2 receptor messenger ribonucleic acid in rat","volume":"137","author":"Usdin","year":"1996","journal-title":"Endocrinology"},{"key":"2023041407562153900_","article-title":"Representation learning with contrastive predictive coding","author":"van den Oord","year":"2018","journal-title":"CoRR"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1186\/s13059-017-1382-0","article-title":"SCANPY: large-scale single-cell gene expression data analysis","volume":"19","author":"Wolf","year":"2018","journal-title":"Genome Biol"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"649","DOI":"10.1007\/978-3-319-46487-9_40","volume-title":"Computer Vision \u2013 ECCV 2016, Amsterdam, Netherlands","author":"Zhang","year":"2016"},{"key":"2023041407562153900_","doi-asserted-by":"crossref","first-page":"651","DOI":"10.1038\/s41467-020-14391-0","article-title":"Surface protein imputation from single cell transcriptomes by deep neural networks","volume":"11","author":"Zhou","year":"2020","journal-title":"Nat. Commun"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/38\/Supplement_1\/i395\/49886954\/btac263.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/38\/Supplement_1\/i395\/49886954\/btac263.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,20]],"date-time":"2023-11-20T00:28:53Z","timestamp":1700440133000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/38\/Supplement_1\/i395\/6617512"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,6,24]]},"references-count":66,"journal-issue":{"issue":"Supplement_1","published-print":{"date-parts":[[2022,6,24]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btac263","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022,7,1]]},"published":{"date-parts":[[2022,6,24]]}}}