SigFormer: interpretable modeling of transcellular signaling and microenvironment interactions from multimodal single-cell data | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article SigFormer: interpretable modeling of transcellular signaling and microenvironment interactions from multimodal single-cell data Yunping Zhu, Mingfei Han, Xiao Li, Yu Zhao, Tongpeng Yue, Liya Wang, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9489991/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Signaling pathways integrate extracellular cues with intracellular signal transduction and gene regulation, providing a mechanistic foundation for interpretable perturbation analysis and drug discovery. Here we present SigFormer, a graph Transformer-based multi-task framework that integrates cell classification, network reconstruction, and in silico knockout to infer mechanistically interpretable sender–receiver pathways spanning ligands, receptors, mediators, transcription factors, and target genes. SigFormer does not require large training cohorts, enabling accurate pathway reconstruction from both single-sample single-cell RNA sequencing and multimodal datasets. Benchmarking demonstrates that SigFormer consistently outperforms nine existing methods across five signaling stages, improving F1 scores by 8–46% over the second-best method at each stage. By applying SigFormer to cancer multi-omics data, we constructed the Cancer Cell Pathway Atlas—spanning 28 cancers and 23 tumor microenvironment cell types—facilitating cross-tumor pathway analysis, drug repurposing, and future tissue-level "white-box" modeling of signaling programs. Biological sciences/Computational biology and bioinformatics/Cellular signalling networks Biological sciences/Cancer/Cancer microenvironment Biological sciences/Systems biology/Genetic interaction/Epistasis Biological sciences/Computational biology and bioinformatics/Machine learning Full Text Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryTables.xlsx Supplementary Tables SupplementaryMaterials.pdf Supplementary Materials Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9489991","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":631974356,"identity":"e8b44fc1-0d1d-49d4-8ad3-06eb496452d0","order_by":0,"name":"Yunping Zhu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9ElEQVRIiWNgGAWjYDACCQjFw8DAfADESCBFC1sCaVpAugyI0yI/u8dMmnfHHRlz/jXfHvzcUZfHwN77+AVDzR2cWhjnnAFqOfOMx3LG2+2GvWcOFzPwHDezYDj2DKcWZokcoJa2wzwGN85uk2ZsO5DYIJHGZsDYcBinFjaEljPPgFrqCGvhgWs538MG1MIM0sL8AJ8WCYm0Ysu5bc+AtrCZSYL8wsZzjI0h4RhuLfIzkjfeeNt2x97g/OFnEqAQ42dvY/7woQa3FiBgAUbNAaB9CcDwawD6DuTBBHwagIH2AayF/wBEC1RkFIyCUTAKRgEcAAAyw1ORyVUZfAAAAABJRU5ErkJggg==","orcid":"","institution":"State Key Laboratory of Medical Proteomics, National Center for Protein Sciences (Beijing)","correspondingAuthor":true,"prefix":"","firstName":"Yunping","middleName":"","lastName":"Zhu","suffix":""},{"id":631974357,"identity":"cb6b9d9d-9775-455d-bb5b-73628b117641","order_by":1,"name":"Mingfei Han","email":"","orcid":"https://orcid.org/0000-0002-6036-6309","institution":"National Center for Protein Sciences (Beijing)","correspondingAuthor":false,"prefix":"","firstName":"Mingfei","middleName":"","lastName":"Han","suffix":""},{"id":631974358,"identity":"86fa0cfa-2125-4ad4-8089-31ff705a498d","order_by":2,"name":"Xiao Li","email":"","orcid":"","institution":"National Center for Protein Sciences (Beijing)","correspondingAuthor":false,"prefix":"","firstName":"Xiao","middleName":"","lastName":"Li","suffix":""},{"id":631974359,"identity":"3722c418-bed8-49fd-9d97-74e61a0f0cdf","order_by":3,"name":"Yu Zhao","email":"","orcid":"","institution":"College of Life Sciences, Hebei University","correspondingAuthor":false,"prefix":"","firstName":"Yu","middleName":"","lastName":"Zhao","suffix":""},{"id":631974360,"identity":"22dc4c86-88ed-42b2-a84e-2410e9efcca4","order_by":4,"name":"Tongpeng Yue","email":"","orcid":"","institution":"Peking Union Medical College Hospital","correspondingAuthor":false,"prefix":"","firstName":"Tongpeng","middleName":"","lastName":"Yue","suffix":""},{"id":631974361,"identity":"3a1ac01b-720e-46f8-9120-453d5f9bb313","order_by":5,"name":"Liya Wang","email":"","orcid":"","institution":"College of Life Sciences, Hebei University","correspondingAuthor":false,"prefix":"","firstName":"Liya","middleName":"","lastName":"Wang","suffix":""},{"id":631974362,"identity":"6f654f16-e81d-4ce4-b01e-211bcf51aa35","order_by":6,"name":"Juan Wang","email":"","orcid":"","institution":"State Key Laboratory of Medical Proteomics, National Center for Protein Sciences (Beijing)","correspondingAuthor":false,"prefix":"","firstName":"Juan","middleName":"","lastName":"Wang","suffix":""},{"id":631974363,"identity":"ee0c4cc3-972d-4782-9032-7029feac1d82","order_by":7,"name":"Jie Ma","email":"","orcid":"https://orcid.org/0000-0002-8934-922X","institution":"National Center for Protein Sciences (Beijing)","correspondingAuthor":false,"prefix":"","firstName":"Jie","middleName":"","lastName":"Ma","suffix":""},{"id":631974364,"identity":"94e6df0e-41c9-47bb-b980-33f31166a6fd","order_by":8,"name":"Hebing Chen","email":"","orcid":"","institution":"Academy of Military Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Hebing","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2026-04-22 03:00:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9489991/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9489991/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108492373,"identity":"2a5aef61-c8c6-4e8e-90c7-5640768f17fc","added_by":"auto","created_at":"2026-05-05 09:57:36","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2214696,"visible":true,"origin":"","legend":"Article File","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9489991/v1_covered_963128de-13f5-4e5f-829c-c5ed3cd5b195.pdf"},{"id":108379944,"identity":"d7bb6b24-3274-4891-adfb-cf5c4a208202","added_by":"auto","created_at":"2026-05-04 04:34:41","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1227420,"visible":true,"origin":"","legend":"Supplementary Tables","description":"","filename":"SupplementaryTables.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-9489991/v1/ab97a1be6dbb9b072d24678f.xlsx"},{"id":108379945,"identity":"ff35f4c4-e61b-4eae-a544-edca91201b25","added_by":"auto","created_at":"2026-05-04 04:34:41","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":2328437,"visible":true,"origin":"","legend":"Supplementary Materials","description":"","filename":"SupplementaryMaterials.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9489991/v1/fbe1164d176dd1df0a331935.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"SigFormer: interpretable modeling of transcellular signaling and microenvironment interactions from multimodal single-cell data","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"
[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9489991/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9489991/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Signaling pathways integrate extracellular cues with intracellular signal transduction and gene regulation, providing a mechanistic foundation for interpretable perturbation analysis and drug discovery. Here we present SigFormer, a graph Transformer-based multi-task framework that integrates cell classification, network reconstruction, and in silico knockout to infer mechanistically interpretable sender–receiver pathways spanning ligands, receptors, mediators, transcription factors, and target genes. SigFormer does not require large training cohorts, enabling accurate pathway reconstruction from both single-sample single-cell RNA sequencing and multimodal datasets. Benchmarking demonstrates that SigFormer consistently outperforms nine existing methods across five signaling stages, improving F1 scores by 8–46% over the second-best method at each stage. By applying SigFormer to cancer multi-omics data, we constructed the Cancer Cell Pathway Atlas—spanning 28 cancers and 23 tumor microenvironment cell types—facilitating cross-tumor pathway analysis, drug repurposing, and future tissue-level \"white-box\" modeling of signaling programs.","manuscriptTitle":"SigFormer: interpretable modeling of transcellular signaling and microenvironment interactions from multimodal single-cell data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-04 04:34:37","doi":"10.21203/rs.3.rs-9489991/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"
[email protected]","identity":"nature-biomedical-engineering","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"natbiomedeng","sideBox":"Learn more about [Nature Biomedical Engineering](http://www.nature.com/natbiomedeng/)","snPcode":"41551","submissionUrl":"https://mts-natbiomedeng.nature.com/cgi-bin/main.plex","title":"Nature Biomedical Engineering","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"ce9f5d50-3cf0-4a00-bb24-3956b542f7c2","owner":[],"postedDate":"May 4th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-05-09T09:55:16+00:00","index":4,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2026-05-09T02:35:54+00:00","index":3,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-05-06T01:06:42+00:00","index":3,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T23:55:44+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T22:33:04+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"11","date":"2026-04-29T18:59:11+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67267658,"name":"Biological sciences/Computational biology and bioinformatics/Cellular signalling networks"},{"id":67267659,"name":"Biological sciences/Cancer/Cancer microenvironment"},{"id":67267660,"name":"Biological sciences/Systems biology/Genetic interaction/Epistasis"},{"id":67267661,"name":"Biological sciences/Computational biology and bioinformatics/Machine learning"}],"tags":[],"updatedAt":"2026-05-04T04:34:37+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-04 04:34:37","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9489991","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9489991","identity":"rs-9489991","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.