About

Emma Lundberg is an Associate Professor of Bioengineering and of Pathology at Stanford University, starting her appointment in January 2022. Her research focuses on bioengineering and pathology, contributing to the understanding of biological systems through engineering principles. Her work involves interdisciplinary approaches to explore cellular and molecular mechanisms, leveraging her expertise to advance biomedical research and innovation.

Research topics

• Computational biology
• Biology
• Bioinformatics
• Computer Science
• Genetics
• Cell biology
• Data science
• Anatomy
• Engineering
• Neuroscience

Selected publications

• A high-resolution spatial map of cilia-associated proteins in the human fallopian tube

  Nature Communications · 2026-04-20

  articleOpen access

  Molecular alterations in the fallopian tubes play a pivotal role in the development of cancer and reproductive disorders, yet their molecular landscape at the protein level remains poorly defined. Here, we map key fallopian tube proteins at single-cell resolution utilizing an integrated transcriptomics and proteomics approach. Based on RNA-seq analysis, we identify 310 genes with elevated expression in the fallopian tube, the majority of which are associated with motile cilia function. We spatially characterize 133 of the corresponding proteins in the fallopian tube and other human tissues with motile cilia to subcellular structures of ciliated cells, validating the findings with single-cell RNA-seq and mass-spectrometry data. Eleven proteins previously only studied on the transcript level without information in cilia databases are further analyzed in a hydrosalpinx patient, showing a thinner epithelium, lower density of FOXJ1 expression, and reduced expression of FHAD1, RIIAD1, and C2orf81. Our high-resolution spatial map aids in dissecting the pathways underlying infertility and diseases linked to cilia-specific functions.

  PublisherDOI

• Abstract 6296: Deciphering the promotional determinants of esophageal cancer in countries with varying incidence.

  Cancer Research · 2026-04-03

  article

  Abstract Esophageal squamous cell carcinoma (ESCC) incidence varies widely across the world, correlating with exposure to known risk factors like alcohol, smoking, and hot liquids. It has traditionally been assumed that such exposures lead to malignant cell transformation by directly or indirectly causing DNA mutations. However, recent evidence revealed that most ESCC risk factors do not elicit distinct mutation profiles, suggesting that these carcinogens may act as non-mutagenic tumor-promoting agents. This study aims to investigate the role of exogenous exposures in promoting the clonal expansion of pre-initiated cells in normal esophageal tissue, leading to ESCC. To address this, we are examining the clonal structure of 200 esophageal epithelium samples from ESCC patients and non-cancer donors across nine geographical regions in China, Iran, Malawi, South Africa, Russia, Brazil, and Canada, with age-standardized ESCC incidence rates ranging from 1 to 84 per 100,000. Detailed demographic, lifestyle, and exposure information, including tobacco, alcohol, and hot liquids, is available. Esophageal epithelial tissues are dissected and analyzed by NanoSeq for mutational signature and mutation burden evaluation, and by 500x whole-exome sequencing to identify mutant genes under positive selection. Spatial proteomics and transcriptomics are ongoing to profile the phenotype of premalignant cells and their microenvironment niches. Preliminary analyses of 97 cases from high- and intermediate-risk regions revealed a strong mutagenic effect of alcohol and tobacco on normal esophageal tissue. Alcohol-related mutational signatures SBS16 and ID11 were detected in 73% of drinkers (22/30), and were enriched in cases exposed to both alcohol and tobacco. In contrast, hot liquid consumption was not related to distinct mutation profiles. dN/dS analysis in 47 cases with available WES data identified 11 genes under positive selection, including NOTCH1, TP53, FAT1, and PPM1D. Cases exposed to tobacco presented higher fractions of mutated epithelia in several of these cancer genes. Notably, the fraction of NOTCH1-mutant epithelia was inversely associated with ESCC incidence, suggesting a reduction of clones harboring this protective mutation in high-risk populations. These findings shed light on the complex interplay between mutagenic and promotional processes in ESCC development and provide insights into the role of known and suspected risk factors in clonal selection. This work may ultimately guide preventive strategies targeting the earliest stages of esophageal carcinogenesis. Citation Format: Laura Torrens, Raquel Blanco, Joanna C. Fowler, Ana Carolina de Carvalho, Behnoush Abedi-Ardekani, Valérie Gaborieau, Priscilia Chopard, Christine Carreira, Abel Gonzalez, Jeffrey Reina, Anna Martinez-Casals, Augusta Jensen, Rui Manuel Reis M. Reis, Abdolreza Fazel, M. Iqbal Parker, David Zaridze, Patricia Ashton-Prolla, Maria P. Curado, Mats Nilsson, Emma Lundberg, Philip H. Jones, Nuria Lopez-Bigas, Paul Brennan, on behalf of the PROMINENT project. Deciphering the promotional determinants of esophageal cancer in countries with varying incidence [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 6296.

  PublisherDOI

• PaperSearchQA: Learning to Search and Reason over Scientific Papers with RLVR

  2026-01-01

  articleOpen access

  James Burgess, Jan N. Hansen, Duo Peng, Yuhui Zhang, Alejandro Lozano, Min Woo Sun, Emma Lundberg, Serena Yeung-Levy. Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

  PublisherOA PDFDOI

• Uncertainty Quantification for Distribution-to-Distribution Flow Matching in Scientific Imaging

  arXiv (Cornell University) · 2026-03-23

  articleOpen access

  Distribution-to-distribution generative models support scientific imaging tasks ranging from modeling cellular perturbation responses to translating medical images across conditions. Trustworthy generation requires both reliability (generalization across labs, devices, and experimental conditions) and accountability (detecting out-of-distribution cases where predictions may be unreliable). Uncertainty quantification (UQ) based approaches serve as promising candidates for these tasks, yet UQ for distribution-to-distribution generative models remains underexplored. We present a unified UQ framework, Bayesian Stochastic Flow Matching (BSFM), that disentangles aleatoric and epistemic uncertainty. The Stochastic Flow Matching (SFM) component augments deterministic flows with a diffusion term to improve model generalization to unseen scenarios. For UQ, we develop a scalable Bayesian approach -- MCD-Antithetic -- that combines Monte Carlo Dropout with sample-efficient antithetic sampling to produce effective anomaly scores for out-of-distribution detection. Experiments on cellular imaging (BBBC021, JUMP) and brain fMRI (Theory of Mind) across diverse scenarios show that SFM improves reliability while MCD-Antithetic enhances accountability.

  PublisherOA PDF

• Cell shapes decode molecular phenotypes in image-based spatial proteomics

  Cell Systems · 2026-04-01

  articleOpen accessSenior author

  Cellular and tissue structures arise from a few cell shapes, which undergo transformations based on biophysical constraints. Despite links between signaling pathways and cellular geometry, whole-proteome orchestration in association with cell shape is underexplored. In this study, over 1 million single cells stained for 11,998 proteins across 11 cell lines in the Human Protein Atlas were analyzed for organelle, pathway, and single-protein levels in association with cellular shapespace. We found that cell and nuclear shapes across cell lines exist in a shared continuum. The subcellular organelle topology varies across cell lines but remains consistent within each cell line's shapespace. At the single-protein level, cells of different shapes in the same cell-cycle phase might be preparing for different fates, and many non-cell-cycle proteins expressed shape-based abundance variation. Using a shape-based coordinate framework, we analyzed the distribution shift of protein spatial localization under drug perturbation.

  PublisherDOI

• A high-resolution spatial map of cilia-associated proteins in the human fallopian tube

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-03-16

  articleOpen access

  ABSTRACT Molecular alterations in the fallopian tubes play a pivotal role in the development of cancer and reproductive disorders, yet their molecular landscape at the protein level remains poorly defined. Here, we map key fallopian tube proteins at single-cell resolution utilizing an integrated transcriptomics and proteomics approach. Based on RNA-seq analysis we identify 310 genes with elevated expression in fallopian tube, the majority of which are associated with motile cilia function. We spatially characterize 133 of the corresponding proteins in fallopian tube and other human tissues with motile cilia to subcellular structures of ciliated cells, validating the findings with single-cell RNA-seq and mass-spectrometry data. Eleven proteins previously only studied on the transcript level without information in cilia databases are further analyzed in a hydrosalpinx patient, showing a thinner epithelium, lower density of FOXJ1 expression, and reduced expression of FHAD1, RIIAD1, and C2orf81. Our high-resolution spatial map aids in dissecting the pathways underlaying infertility and diseases linked to cilia-specific functions.

  PublisherDOI

• Abstract 6902: Remodeling of cancer cell architecture by chemotherapy.

  Cancer Research · 2026-04-03

  article

  Abstract How chemotherapy reshapes tumor cells—and how these changes influence outcomes such as drug resistance—remains largely unclear. Here, we present a multimodal, global characterization of tumor subcellular organization and its reorganization by chemotherapy. We use self-supervised learning to encode protein coordinates across four orthogonal data modalities: proteome-wide size-exclusion chromatography fractionation (before and after treatment with cisplatin or vorinostat), native-state immunofluorescence imaging, affinity purification, and primary sequence information covering 7,579 proteins. This integrated map resolves 174 subcellular components, spanning molecular assemblies from protein complexes to organelles across a size range of ∼10-9 to10-5nm. 58 components undergo significant remodeling upon treatment, recapitulating known mechanisms of action and revealing previously unrecognized alterations in pathways such as cytoskeletal organization and metabolic rewiring. We systematically validate these “chemotherapy-response assemblies” using genome-wide CRISPR knockout drug-sensitivity profiling, identifying which assemblies confer drug sensitivity versus resistance. Chemotherapy-remodeled components serve as convergence points for cancer mutations that predict therapeutic response—including those involved in homologous recombination repair, chromatin remodeling, and double-strand break repair. Citation Format: Gege Qian, Xiaoyu Zhao, Leah V. Schaffer, Kyung-Mee Moon, Jiahao Gao, Emma Lundberg, Leonard Foster, Trey Ideker. Remodeling of cancer cell architecture by chemotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 6902.

  PublisherDOI

• Generative machine learning unlocks the first proteome-wide image of human cells

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-04-02

  articleOpen accessSenior authorCorresponding

  The spatial organization of proteins within cells governs virtually all cellular functions. Yet, current imaging technologies can simultaneously visualize only tens of proteins, orders of magnitude below the thousands that populate a single human cell. Here, we present ProtiCelli , a deep generative model that simulates microscopy images for 12,800 human proteins from just three cellular landmark stains. Trained on 1.23 million images from the Human Protein Atlas, ProtiCelli outperforms existing methods in reconstruction accuracy and textural fidelity, and generalizes to unseen cell types and drug perturbations absent from training. We demonstrate that ProtiCelli -generated images preserve hierarchical subcellular organization, recapitulate known protein–protein interaction landscapes, and resolve compartment-specific functions of moonlighting proteins at the single-cell level. Remarkably, the model infers drug-induced changes in protein expression and localization from cell morphology alone, predicts cell cycle stage without dedicated cell cycle markers, and enables unsupervised segmentation of subcellular compartments as well as spatial decomposition of gene sets into functional regions. Ultimately, we leverage ProtiCelli to generate Proteome2Cell , an unprecedented dataset of 30.7 million simulated images creating 2,400 “virtual cells” across 12 human cell lines. These proteome-scale images enable the construction of hierarchical single-cell models that distinguish conserved from dynamic protein architectures. Integration of Proteome2Cell into the Human Protein Atlas democratizes the exploration of these “virtual cells”. By computationally bridging the experimental scalability gap, ProtiCelli establishes a foundation for spatial virtual cell modeling and paves an avenue for transforming spatial proteomics from cataloging proteins to simulating complete cellular systems.

  PublisherDOI

• Intrinsic heterogeneity of primary cilia revealed through spatial proteomics

  Cell · 2025-09-26 · 20 citations

  articleOpen accessSenior author

  Primary cilia are critical organelles found on most human cells. Their dysfunction is linked to hereditary ciliopathies with a wide phenotypic spectrum. Despite their significance, the specific roles of cilia in different cell types remain poorly understood due to limitations in analyzing ciliary protein composition. We employed antibody-based spatial proteomics to expand the Human Protein Atlas to primary cilia. Our analysis identified the subciliary locations of 715 proteins across three cell lines, examining 128,156 individual cilia. We found that 69% of the ciliary proteome is cell-type specific, and 78% exhibited single-cilia heterogeneity. Our findings portray cilia as sensors tuning their proteome to effectively sense the environment and compute cellular responses. We reveal 91 cilia proteins and found a genetic candidate variant in CREB3 in one clinical case with features overlapping ciliopathy phenotypes. This open, spatial cilia atlas advances research on cilia and ciliopathies.

  PublisherDOI

• Dataset: Multiplexed Deep Visual Proteomics Unveils Spatial Heterogeneity and Rare Endocrine States in Human Adult Pancreatic Islets

  2025-04-21

  datasetSenior author
  PublisherDOI

Frequent coauthors

• Mathias Uhlén

  KTH Royal Institute of Technology

  345 shared

• Fredrik Pontén

  Uppsala University

  121 shared

• Wei Ouyang

  Jilin University

  121 shared

• Christian Gnann

  Science for Life Laboratory

  85 shared

• Charlotte Stadler

  Science for Life Laboratory

  71 shared

• Peter Thul

  Science for Life Laboratory

  71 shared

• Diana Mahdessian

  68 shared

• Trang Le

  Stanford University

  68 shared