The initiative, called MOSAIC (Multi-Omics Spatial Atlas in Cancer), is a multi-center clinical omics study that has profiled more than 2,700 cancer samples spanning multiple tumor types. The consortium brings together researchers from Owkin and clinical and academic partners including Gustave Roussy in France, Charité Universitätsmedizin Berlin and University Hospital Erlangen in Germany, Lausanne University Hospital in Switzerland, and the University of Pittsburgh in the United States. The study was funded by Owkin Inc. and conducted as a non-interventional, multicenter clinical protocol, registered as NCT06625203, with ethics approval at each participating site. What distinguishes MOSAIC from many previous atlas projects is not just its size but its deliberate integration of data modalities that are usually collected and analyzed in isolation.

Technically, each tumor sample in MOSAIC is characterized through several complementary lenses. Spatial transcriptomics captures gene expression across intact tissue sections, preserving the two-dimensional geography of the tumor and allowing researchers to measure which genes are active at which coordinates on a slide. Single-nuclei transcriptomics resolves expression at the level of individual cells, distinguishing malignant subclones from the diverse non-malignant cells of the tumor microenvironment. These are layered alongside digitized histology scans, bulk whole-transcriptome sequencing, and whole-exome sequencing, which together provide the genomic mutations, the aggregate expression landscape, and the classical pathological view of the tissue. Extensive curated clinical information, including treatment regimens and outcomes, is attached to each sample, so that molecular patterns can eventually be linked to how patients actually responded.

The rationale for this multi-modal design is that no single technology can capture the full complexity of a tumor. Histology reveals architecture but not molecular identity; bulk sequencing averages away the very heterogeneity researchers want to measure; single-cell methods lose spatial context; and spatial methods, until recently, lacked single-cell resolution. By generating all of these representations from the same samples under standardized protocols, MOSAIC aims to create a resource in which computational methods, including artificial intelligence, can learn to connect what a pathologist sees under the microscope with what the genome and transcriptome reveal, and ultimately to identify clinically relevant biomarkers and cancer subtypes that no single modality alone could expose.

Alongside the full study design, the consortium has released an initial public dataset called the MOSAIC Window, now available through the European Genome-phenome Archive under study ID EGAS50000000689. This first release covers 60 patients across five tumor types and includes quality-controlled data from spatial transcriptomics, single-nuclei RNA sequencing, bulk RNA sequencing, and whole-exome sequencing, together with curated clinical descriptors such as age, smoking status, disease stage, and survival. The consortium describes this release as a glimpse into the project’s potential, with further data releases planned as the study matures. The cohort design spans eleven major cancer types overall, including bladder, breast, colorectal, gastric, glioblastoma, head and neck squamous cell carcinoma, mesothelioma, non-small cell lung cancer, ovarian, and pancreatic cancers, sampled at baseline diagnosis, after neoadjuvant therapy, and at recurrence or progression.

The early analyses reported in the paper demonstrate the analytical power that comes from combining these modalities. Using the MOSAIC Window data, the researchers quantified both intra-patient and inter-patient heterogeneity, uncovering distinct malignant cell subsets within individual tumors. They then examined how these malignant subpopulations relate to their surroundings and found a correlation between the intrinsic oncogenic signaling activity of malignant cells and the colocalization of specific tumor microenvironment cell types. In other words, the signaling programs running inside cancer cells appear to be associated with which immune and stromal cells gather around them, a link that could help explain why some tumor regions are immunologically cold and others inflamed, and why immunotherapy can succeed in one part of a tumor while failing elsewhere.

To make these concepts concrete, the paper presents four diverse case studies, each illustrating a different way that integrated multi-omics data can decipher heterogeneity. In these examples, the team used computational tools such as differential expression analysis and gene set enrichment analysis to characterize malignant cell clusters identified from single-nuclei data, and quantified pathway activities with methods like PROGENy and GSVA to compare signaling states across subpopulations. Statistical measures, including variance analyses and Pearson correlations between spot-level cellular fractions, were used to test whether the associations between malignant signaling and microenvironment composition were robust across samples and indications. The case studies span different tumor types and different biological questions, collectively highlighting how the same standardized data framework can support a wide range of investigations.

The technical rigor behind the resource is considerable. Supplementary documentation accompanying the paper details quality metrics for every modality, from median gene counts per spatial spot and read-mapping percentages in transcriptomics to sequencing depth, contamination estimates, and GC content in whole-exome data. A detailed data management plan describes how samples are pseudonymized, how data flows across modalities are linked to a common study identifier, and how the resource adheres to FAIR principles, making the data findable, accessible, interoperable, and reusable. The clinical study protocol, covering objectives, endpoints, eligibility criteria, and the statistical analysis plan, is also published, giving the community full visibility into how the resource was constructed.

For the field of computational oncology, the significance of MOSAIC lies partly in addressing a persistent bottleneck: the scarcity of large, consistent, well-annotated datasets. Machine learning models for histology and multi-omics integration have advanced rapidly, but they are often trained on small or inconsistently processed cohorts, limiting their generalizability and clinical translation. A standardized atlas of thousands of tumors, each with matched spatial, single-cell, genomic, histological, and clinical data, provides exactly the kind of training substrate that multimodal artificial intelligence approaches require. The consortium explicitly states its aim to use artificial intelligence and other computational approaches to integrate all data modalities and extract clinically actionable insights, positioning MOSAIC as infrastructure for the next generation of biomarker discovery.

There are, of course, caveats worth noting. The study is funded entirely by a single company, Owkin, and several authors are current or former employees holding shares or stock options in the company, alongside a range of disclosed consulting relationships with pharmaceutical and diagnostics firms at the academic sites. The paper is published as a shared early version, citable with a permanent DOI and subject to further editorial updates. And while the MOSAIC Window release demonstrates feasibility and analytical promise, the clinical value of the resource will ultimately depend on whether the biomarkers and subtypes it reveals can be validated prospectively and shown to improve patient outcomes. Still, with more than 2,700 samples profiled, an open first dataset already in researchers’ hands, and a standardized framework linking molecular, spatial, and clinical dimensions of cancer, MOSAIC represents one of the most ambitious attempts yet to map the internal diversity of tumors at scale, and a foundation on which the precision oncology community can build for years to come.

Subject of Research: Multi-omics characterization of intra-tumoral heterogeneity using spatial and single-cell profiling in cancer

Article Title: MOSAIC: Intra-tumoral heterogeneity characterization through large-scale spatial and cell-resolved multi-omics profiling

Article References: MOSAIC Consortium, Cornish, A. J., Bayard, Q., Karabajakian, A., Madissoon, E., Ferrarini, G., Youssef, A., Badoual, C., de Leval, L., Dressman, D., Durand, E. Y., Erber, R., Florian, S., Garberis, I., Haignere, C., Homicsko, K., Keilholz, U., Lee, A. V., Lehar, J., … Hoffmann, C. (2026). MOSAIC: Intra-tumoral heterogeneity characterization through large-scale spatial and cell-resolved multi-omics profiling. Genome Medicine. https://doi.org/10.1186/s13073-026-01791-y

Image Credits: AI Generated

DOI: 10.1186/s13073-026-01791-y

Keywords: MOSAIC, intra-tumoral heterogeneity, spatial transcriptomics, single-nuclei RNA-seq, multi-omics, tumor microenvironment, precision oncology, whole-exome sequencing, cancer biomarkers, artificial intelligence, Genome Medicine, open data

The research team, led by scientists at the Olivia Newton-John Cancer Research Institute, the Walter and Eliza Hall Institute, and collaborators in Australia and France, tackled a problem that has long frustrated the field. When a pathologist examines a needle biopsy, how much of the tumor’s clonal complexity is actually being seen? When circulating tumor DNA is detected in a patient’s plasma, which clones are shedding that DNA and which are hiding? Previous comparisons of the two approaches often focused on a handful of well-known driver mutations such as TP53 or KRAS, rather than the full landscape of clonal diversity. To go deeper, the team turned to a powerful lineage-tracing technology known as genetic barcoding.

The technique works like this: cancer cells are infected with a lentiviral library containing thousands of unique DNA sequences, each of which integrates permanently into the genome of a single cell and is inherited by all of its descendants. Every clone in the resulting tumor therefore carries a molecular name tag that can be recovered by PCR amplification and sequencing. In the study, the researchers applied this strategy to six preclinical breast cancer models, comprising two established cell lines, MDA-MB-231 and MDA-MB-468, and four patient-derived xenografts, or PDXs, established from drug-naive triple-negative breast cancer patients. Because the barcodes provide an unbiased record of clonal identity independent of any particular mutation, the researchers could track exactly which clones ended up in which sample.

The first major finding concerned geography. Using a stochastic agent-based simulation of three-dimensional tumor growth, adapted from a previously published model, the team predicted that clonal density would be higher in the center of a tumor than at its periphery. When they then dissected real barcoded tumors grown in mice into central and peripheral pieces and sequenced the barcode repertoire of each, the prediction held. In the MDA-MB-231 model, the tumor center contained on average roughly three times more distinct barcodes than peripheral pieces, and across all six models the center harbored a substantially larger share of the total clonal diversity. In one cell line model the center accounted for 66 percent of all detected barcodes, and in the other, 77 percent. The pattern was reproduced in all four PDX models, with the intriguing exception that a highly necrotic PDX showed a smaller center-periphery difference, presumably because cell death had erased many barcodes from the core.

This spatial bias has direct clinical implications, because radiologists typically aim needles toward the center of a lesion. The barcoding experiments showed that a needle biopsy reaching the tumor core captures a large fraction of the tumor’s biomass, with clones detected in needle samples representing roughly 80 to 90 percent of the total tumor reads. However, the content of any single needle varied considerably depending on the direction from which it was inserted. Several dominant clones present in the primary tumor were missed entirely by some needle tracks, and biopsies taken from opposite sides of the same tumor contained non-overlapping sets of barcodes. Minor clones, in particular, were frequently absent from needle samples, a sobering result given that low-frequency clones can drive metastasis and drug resistance.

The team then turned to the liquid side of the comparison. Blood was collected from tumor-bearing mice at three time points, corresponding to tumor volumes of 100, 300, and 800 cubic millimeters, and cell-free DNA was extracted from the plasma before the barcodes were amplified and sequenced. Barcode recovery in cfDNA proved strikingly variable. Some models, such as PDX-4295 and CRCM412, shed detectable DNA early and reliably, while others yielded little or nothing even when tumors were large. Recovery also increased with tumor burden, and within a given model the same clones could appear and disappear in the plasma of the same mouse over time. When the researchers implanted identical clonal pools into multiple recipient mice, the clonal composition of the resulting sister tumors was well conserved, yet barcode recovery still fluctuated between animals, suggesting that both intrinsic clone properties and stochastic factors influence how much DNA a tumor releases into the bloodstream.

When cfDNA barcodes were detected, they told a meaningful story. The plasma repertoire at the final time point broadly reflected the clonal makeup of the primary tumor, and the clones represented in cfDNA accounted for up to 80 percent of tumor biomass in the best-shedding model. Notably, cfDNA also served as a reasonable surrogate for the clonal composition of lung metastases, hinting that DNA in the blood may report on disease beyond the primary site. But the method had clear blind spots: only a small percentage of the tumor’s total barcodes appeared in the plasma in most models, and some dominant clones in the primary tumor were never detected in cfDNA at all.

The head-to-head comparison revealed that the two biopsy types are complementary rather than interchangeable. Deep needle biopsies generally captured a larger share of primary tumor barcodes and higher clonal diversity than cfDNA, except in the CRCM412 model, where cfDNA outperformed both deep and shallow needle sampling. Three distinct patterns emerged across individual mice: sometimes cfDNA better recapitulated tumor heterogeneity, sometimes the needle did, and sometimes both performed comparably. Crucially, combining the two methods significantly increased the percentage of tumor biomass captured compared with either alone in several models. In the MDA-MB-231 model, the researchers went one step further and retrieved circulating tumor cells from the blood; adding their barcodes to the combined needle-plus-cfDNA dataset pushed the captured biomass above 80 percent.

The authors are careful to note the limitations of their system. Xenograft experiments unfold over weeks rather than the years over which human tumors evolve, and barcode detection, which relies on amplifying tags shorter than 100 base pairs, may underestimate the cfDNA fragments detectable in clinical assays that query multiple genomic loci. Blood volumes in mice are also far smaller than those drawn from patients. Nevertheless, the barcoded models offer something patient samples cannot: a complete, ground-truth map of clonal composition against which every sampling method can be scored, tracked longitudinally, and manipulated experimentally without the confounding effects of treatment.

The broader takeaway for oncology is pragmatic. Solid biopsies deliver intact cells that support histology, receptor-status testing, single-cell omics, and assessment of tumor-infiltrating lymphocytes, while liquid biopsies offer a minimally invasive, repeatable readout that can integrate heterogeneity across multiple lesions that no needle could reach. This study provides quantitative backing for the growing consensus, supported by earlier analyses of hundreds of patient samples, that the two approaches should coexist in clinical practice. As barcoded preclinical models are combined with genomic and spatial transcriptomic profiling, researchers hope to identify the properties that make some clones prolific DNA shedders and others invisible, knowledge that could sharpen the interpretation of liquid biopsy results, flag false negatives, and ultimately guide more rational personalized therapies for breast cancer patients.

Subject of Research: Using genetic barcoding to quantify how well solid and liquid biopsies capture intra-tumoral heterogeneity in breast cancer models

Article Title: Genetic barcoding uncovers the clonal makeup of solid and liquid biopsies and their ability to capture intra-tumoral heterogeneity

Article References: Serrano, A., Weber, T. S., Berthelet, J., Ftouni, S., El-Saafin, F., Lee, S., Lim, E., Charafe-Jauffret, E., Ginestier, C., Williams, D., Hollande, F., Yeo, B., Dawson, S.-J., Naik, S. H., & Merino, D. (2026). Genetic barcoding uncovers the clonal makeup of solid and liquid biopsies and their ability to capture intra-tumoral heterogeneity. Molecular Systems Biology, 22(5), 659-684. https://doi.org/10.1038/s44320-026-00194-w

Image Credits: AI Generated

DOI: 10.1038/s44320-026-00194-w

Keywords: breast cancer, genetic barcoding, intratumoral heterogeneity, liquid biopsy, cell-free DNA, needle biopsy, patient-derived xenografts, circulating tumor DNA, clonal evolution, tumor sampling, precision oncology, Molecular Systems Biology