About
Origin builds a tissue-first cancer biology data and model engine that generates spatial, molecular, and functional maps from real patient tumor samples. It serves pharmaceutical and AI companies through data licensing, differentiating itself by preserving native tissue context and combining multimodal spatial profiling with live-tissue perturbation and response data.
Market
Origin competes in AI-enabled biotechnology, spatial biology, and precision-oncology data markets, positioning itself as a tissue-first data and model platform for studying cancer in real patient tumors. Its differentiation is the combination of spatial transcriptomics and morphology with functional perturbation-response measurements from living or native tumor tissue, followed by multimodal AI modeling. The competitive set includes spatial-biology platform providers such as 10x Genomics, Vizgen, Akoya, Bruker, and Bio-Techne, as well as tissue-AI companies such as Nucleai and PathAI; Origin emphasizes proprietary patient-derived datasets and therapeutic-response models rather than only assay instrumentation or image analysis.
Origin’s primary customers are pharmaceutical and AI companies developing cancer therapeutics, especially oncology R&D teams working on target discovery, drug-response modeling, biomarker discovery, or patient stratification. Its related Axis offering also targets cell- and gene-therapy developers seeking regulatory-DNA designs and proprietary sequence datasets.
At a Glance
Problem
Origin addresses a central data and translation problem in cancer therapeutics: researchers need to understand how drugs and genetic medicines affect tumors in their native architecture, not just in simplified models. The company’s economic value proposition is to give pharmaceutical and AI customers more actionable evidence for target discovery, drug-response prediction, and patient stratification, reducing uncertainty in expensive oncology R&D. The public evidence does not disclose a dollar estimate of the pain or customer ROI, so the economics are best characterized as high-value research uncertainty rather than a quantified cost.
The clearest use case is testing treatments directly on patient-derived tumor tissue and identifying the cell populations and tumor contexts that are sensitive or resistant. That can help drug developers decide which targets, therapies, combinations, or patient subgroups merit further investment.
Product / Service
Origin describes a tissue-first data and model engine. It sources FFPE tumor blocks and fresh tumor resections through hospital, biobank, and clinical partners, with anonymized clinical annotations. The company then combines spatial transcriptomics with H&E and immunofluorescence imaging, and subjects precision-cut tumor slices or dissociated cells to matched untreated and perturbed conditions, including drugs, cytokines, immune modulators, and genetic or molecular interventions.
The resulting product is a multimodal dataset and modeling layer: spatial maps of cell states and tumor niches, molecular and functional response profiles, and AI models trained on proprietary sequencing and spatial-transcriptomics data. Origin says these models support target discovery, perturbation-response prediction, cross-modal inference, and patient stratification. Its YC profile indicates a delivery model centered on licensing patient-derived cancer datasets to pharmaceutical and AI companies, supplemented by partnerships around data and therapeutic-design work.
Market
Origin competes at the intersection of AI-enabled cancer therapeutics, spatial and single-cell biology, patient-derived cancer models, and drug-discovery data licensing. Its differentiation is the combination of native patient tissue, spatial context, functional perturbation, and multimodal AI rather than a purely computational model or a conventional cell-line assay. The closest evidence-backed adjacent competitor is CrownBio, which also offers an ex vivo patient-tissue platform for evaluating monotherapy and combination drug responses in patient tumors.
Origin appears to be an early, pre-revenue company rather than a scaled commercial vendor. It was founded in 2025, joined Y Combinator’s Winter 2026 batch, is listed as active with four employees in San Francisco, and identifies itself as YC-backed. The public evidence shows that it is building its proprietary dataset and soliciting partnerships, but names no customers, revenue, or completed commercial deployments; therefore its demonstrated traction is funding-program validation and platform development, not yet disclosed sales traction.
Founders & Leadership
Funding History
Possible Ventures
Recent News
Origin announced the release of 10,000 AI-designed proximal enhancer-like sequences spanning three cell types, with predicted activity, transcription-factor binding annotations, and 3D structures. The release makes the dataset available for research use.
Origin published a benchmark comparing Muon and Adam variants for regulatory DNA modeling. The company reported that MuonW reached its target validation perplexity with approximately 37% fewer FLOPs than AdamH.
Y Combinator profiled Origin as a Winter 2026 company creating patient-derived cancer single-cell datasets for licensing to pharmaceutical and AI companies. The profile describes datasets that measure drug and genetic-medicine effects under controlled DNA-switch conditions.
Origin announced Axis, a multifunctional DNA model designed to both generate regulatory DNA elements and predict their function. The company stated that Axis beat DeepMind’s AlphaGenome on multiple tasks.
Active Roles
0No active roles right now.
Get notified when they postBusiness Model
Origin’s disclosed business model is licensing patient-derived cancer datasets to pharmaceutical and AI companies. It also describes partnerships around regulatory-sequence designs and access to proprietary biological datasets, but public sources do not disclose specific pricing or subscription terms.