Stately Bio founder explains how combining machine learning and imaging is redefining what it’s possible to measure in biology.
The recent launch of Stately Bio has highlighted a key bottleneck in regenerative medicine: the inability to monitor living cells in real time without destroying them. After three years in stealth mode, the Palo Alto-based startup emerged this month with $12 million in seed funding and a platform that combines machine learning and advanced imaging to address this longstanding challenge.
Stately’s ML-powered platform enables high-resolution, label-free imaging of live cells – tracking cell identity, quality, and behavior continuously. This allows researchers to observe how cells grow, mature, and respond to interventions without genetic or chemical labeling, and is believed to have the potential to accelerate the development cycle for cell therapies.
And, with its founder a former machine learning lead at Google’s aging focused R&D company Calico, the company has longevity in its DNA…
Longevity.Technology: Traditionally, scientists working on cell therapies have faced a tough trade-off – either kill cells to analyze their internal workings or settle for less informative, non-destructive methods. Stately Bio is aiming to eliminate the need for compromise, and the company’s early results are promising, successfully producing enhanced cells that are already being explored for applications ranging from drug toxicity screening and disease modeling to potential therapeutic use. To learn more about the company and its potential in longevity, we sat down with founder and CEO, Frank Li.
Li’s time at Calico opened his eyes to the world of aging biology and also the transformative potential of ML in biomedical research, particularly when it came to imaging.
“One of the things that really impressed me was how impactful the applications of technologies, especially on the computer vision side, could be,” he says. “There were just so many applications where it became clear that the machines were, in fact, able to do things that even humans could not even fathom doing.”
Unravelling temporal trajectories
This realization became foundational to Li’s vision for what would eventually become Stately Bio. At the core of the company is a technology platform built to solve a very specific challenge in the life sciences: how to study living cells in a non-destructive way.
“Biology is the study of life, and cells are the atoms of biology, if you will,” he says. “But the technologies that we have used to study cells, such as transcriptomics or immunofluorescent staining, oftentimes force us to kill them. So you’re studying dead matter at the end of the day.”
According to Li the latest developments in computer vision mean that there is now an alternative option.
“Using imaging, it became very apparent that, for the first time, you could keep the cells alive and start to unravel these highly dynamic temporal trajectories of how cells change over time,” he explains.
The real challenge facing Stately wasn’t the imaging hardware – microscopes and sensors have long been capable of capturing high-resolution biological images – but rather the software needed to analyze them.
“What we needed was to be able to tease out the relevant signals in an automatic, quantifiable way,” he says. “And that’s really where the technology basis for the company started to take root. We’re hoping to build a future where the predominant paradigm of medicine is going to be about healing the body using regenerative medicine. A future where we’re not just trying to treat symptoms – we’re trying to improve and repair damage that’s accumulated in our bodies and organs.”
The era of phenomics
After spending a couple of years building out its core technology, the team at Stately Bio turned its attention to a pressing bottleneck in the field of cell therapy: understanding how to control and guide the transformation of stem cells into specific, “therapeutically useful” cell types.
“Stem cells have the potential to become all the cell types of the human body,” says Li. “But for various technical reasons, they are not good direct therapeutic candidates. The real value lies in the differentiated cells – the cells that stem cells can become – but current methods for producing those downstream cells are inefficient and hard to scale. So that is really the key bottleneck that we’re hoping to be able to unlock using our technology.”
Li believes Stately’s technology has the potential to enable a paradigm shift he likens to the rise of genomics in the 2000s.
“I would love for the 2020s to become the era of phenomics, where we start to interrogate phenotypes directly through non-destructive imaging,” he says. “To be able to assess biological systems at any given point in time, but also how they evolve over time, without having to invest the considerable sums of time and money and labor typically needed to run these studies.”
Partial reprogramming potential?
Ultimately Li sees Stately as enabling a new kind of biology – one that captures the temporal nature of life. Understanding how cells respond to signals over time is crucial, particularly in stem cell differentiation.
“Biology is life – it’s dynamic,” he says. “The same signal applied 24 hours later can lead a cell down opposite fates, so when and for how long you apply signals are incredibly important factors to consider. We think of ourselves as unlocking that temporal dimension.”
While Stately is currently focused on forward differentiation – transforming stem cells into mature, functional cell types – Li suggests that the same technology could be used in the reverse direction: partial cellular reprogramming.
“I would bet that there are good signals that we would be able to tease out from partial reprogramming campaigns,” he says. “These applications could serve as useful benchmarks for longevity-related studies, particularly those aimed at reversing cellular aging. That obviously rests on certain hypotheses about the impact of those interventions on longevity and aging. But from a technology perspective, it’s really two sides of the same coin.”
Overcoming the dogma
Of course, pioneering a new methodology means challenging entrenched ways of thinking, especially in biology where imaging has traditionally played a secondary role to more invasive, molecular techniques.
“The big idea behind Stately is this almost heretical idea that you can actually study biology just using imaging-based readouts – it’s not accepted dogma,” says Li.
To overcome any skepticism, the company has focused on validating its models through rigorous internal studies and external collaborations.
“We’ve done a bunch of in-house work to compare and validate our technology against different ways of establishing ground truth,” says Li. “We’ve seen extremely high concordance between our quantifications and specific markers you can assess using established imaging-based methods such as immunofluorescent staining.”
While Stately’s technology offers clear benefits in terms of efficiency – saving time, labor, and cost – it also unlocks the potential to make scientific discoveries that were previously out of reach.
“The efficiency piece is key, and it enables everything else,” says Li, giving an example of how the platform has been used internally to improve stem cell differentiation protocols for hepatocytes, or liver cells. “We’ve been able to take an existing state-of-the-art differentiation protocol, and by using our machine learning models to predict hepatocyte maturation directly from an image, we can actually run these kinds of screenings much more efficiently, much larger scale, much more rapidly.”
The results were compelling. In functional assays used to validate the maturity and performance of liver cells, the company’s updated protocols showed improvements ranging from 3x to 10x over the current standard.
“It’s a validation of this strategy of building the tools that will allow us to unlock new science,” says Li. “We’re not just improving how biology is measured – we’re redefining what’s possible.”
The post Unlocking biology’s temporal dimension appeared first on Longevity.Technology – Latest News, Opinions, Analysis and Research.
