Inside 10x Genomics’ Mission to Measure Biology at Scale

We still don’t understand most of biology because it’s hard to measure. The molecular machinery of life operates across staggering complexity, yet researchers have often relied on technologies that blur the details. Serge Saxonov, co-founder and CEO of 10x Genomics, believes that limitation is finally changing.

“Fundamentally, we still don’t understand most of biology,” Saxonov says. “And that’s something people consistently underestimate, even within the field.”

His company’s strategy is straightforward but ambitious. The goal is to build technologies that measure biology at the scale and resolution needed to capture its complexity and accelerate scientific discoveries that improve human health.

Founded in 2012 by Saxonov and co-founders Ben Hindson and Kevin Ness, 10x Genomics has grown into one of the most influential toolmakers for biology. The publicly traded company now commands a market capitalization of roughly $2.5 billion and generates around $600 million in annual revenue. Its platforms are used widely across academia and industry to profile individual cells and map molecular activity inside tissues.

But Saxonov insists the company is not fundamentally about any single technology.

“We try hard not to be attached to a particular technology,” he explains. “We’re driven by questions and opportunities. Where are the big gaps in biology that, if solved, would unlock major progress?”

One of those gaps was single-cell analysis.

When 10x Genomics launched its Chromium platform in 2016, the technology allowed researchers to monitor gene expression in tens of thousands of individual cells simultaneously. Today, they can routinely analyze millions. The result has been what many researchers describe as a single-cell revolution.

“Back then it was clear that the scale simply wasn’t sufficient,” Saxonov says. “So we built tools designed from the start to measure biology at scale and the right resolution.”

More recently, the company has pushed further into spatial biology, a rapidly growing field that maps molecular activity within intact tissues. Platforms such as Visium and Xenium allow scientists to see not only which genes are active in a cell, but exactly where those cells sit in a tissue and how they interact.

That spatial context is critical, particularly in diseases like cancer where cellular neighborhoods shape disease progression.

“Spatial gives you the ultimate combination of cell biology, pathology, and genomics,” Saxonov says. “Instead of measuring one biomarker at a time, you can look at hundreds or thousands while preserving the tissue context.”

These capabilities are increasingly important at the intersection of biology and artificial intelligence.

“Everyone talks about AI,” he says. “But in parallel a more quiet revolution has been happening in technology to measure biology. The convergence of these two will absolutely transform the world.”

Large biological datasets are now being generated at scales that were unimaginable even five years ago. The Biohub's (formerly Chan Zuckerberg Initiative) Initiative’s Billion Cells Project, for example, aims to generate a dataset of one billion single cells using 10x technology to fuel new AI models in biology.

Similarly ambitious efforts are emerging globally. In Singapore, 10x is collaborating with the A*STAR Genome Institute of Singapore on the TISHUMAP initiative, which will analyze thousands of tumor samples used to uncover new biomarkers and therapeutic targets.

One of the newest efforts pushing this vision forward is PharosAI, a UK consortium bringing together King’s College London, Queen Mary University of London, and major NHS research hospitals. Using 10x Genomics’ Xenium spatial platform, the initiative aims to convert decades of archived cancer samples into one of the world’s largest multimodal cancer datasets. By combining spatial biology, genomics, imaging data, and AI models, the project hopes to uncover hidden patterns in tumor biology, enabling earlier diagnoses, more precise therapies, and faster drug discovery. 

Meanwhile, collaborations with groups such as the Arc Institute aim to generate massive perturbation datasets that could enable “virtual cell” models capable of predicting how cells respond to genetic or chemical changes.

“If you want to build a virtual model of cells or tissues, you need massive amounts of high-quality data,” he says. “That’s exactly the type of data our platforms were designed to generate.”

But producing data is only part of the equation. Extracting meaning requires combining well-annotated biological samples, technologies that measure those samples at scale, and AI capable of interpreting the resulting data.

“The answers to disease are locked inside clinical samples,” Saxonov says. “To unlock that biology you need the samples, the technology to measure them, and the AI to make sense of the measurements.”

The long-term vision is predictive models of biology that guide drug discovery, diagnostics, and personalized medicine. Saxonov believes these advances could eventually enable therapies tailored to individual patients.

For now, however, the mission remains focused on building the tools that enable discovery.

“Our goal is to accelerate fundamental science,” Saxonov says. “If we do that well, the impact on human health will follow.”