Once More Unto the Breach, Dear Friends
Building out a research practice at Dimension.
I’m incredibly honored to join Dimension. I’m thrilled to begin collaborating with a group of founders I’ve admired since beginning my investment career. There’s no greater professional privilege than tackling the hardest problems with friends who support, inspire, and challenge you.
Over the years, we’ve bonded over our shared belief that scientific progress is inexorable. Our society’s need for more sensitive diagnostics, more effective therapeutics, and more powerful bio-based solutions is independent from interest rate volatility or the market’s palpable melancholy.
Dimension’s founders and I agree that the boundary between biology and technology has dissolved. We believe that software, hardware, and biological wetware are coalescing into a single amalgam. Modern biotech platforms can interconvert between atoms and bits with astonishing magnitude and velocity—each revolution creating compounding economies of scale and enormous data advantages. Properly admixed, we believe these technologies will produce shocking, discontinuous leaps in scientific knowledge and enterprise value.
I subscribe to the Galisonian notion that new tools are the driving force behind scientific paradigm shifts. The transition from analog to digital devices has transformed myriad scientific domains including biotechnology. Emerging tools are slashing experimental costs, flipping the script on the scientific method. Researchers can inexpensively generate terabytes of bio-data in hours, allowing hypotheses to fall out of experiments rather than the other way around.
I’m a tinkerer at my core. As a kid, Lego sets lost their luster as soon as I was done building them. In seventh grade, my first “business” consisted of me soldering custom microcontrollers into video game consoles to help students boost their skills. By studying chemical engineering in college, I gained a deeper appreciation for the way disciplines harmonize. Not only did I learn how atoms click together to form cellular machinery, but also how nonlinear interactions between heat and pressure can transform proofs-of-concept into industrial-scale products.
Unsurprisingly, the first research focus of my investment career was DNA sequencing—a beautiful example of machines leveraging multiple disciplines to change an industry (and launch new ones). Using this blog page, I plan to write about many exciting technologies. For now, I want to briefly show my affection for sequencing by glancing under the hoods of these metal contraptions.
Sequencers underpin the ‘read’ in the canonical ‘read-write’ paradigm shared across software and, increasingly, biology. Sequencers are singular microcosms of convergent technologies. But they’re only one unit operation. High-throughput biotech platforms operated by companies like Recursion Pharmaceuticals (RXRX) incorporate hundreds of hardware, software, and wetware operations in a highly orchestrated dance—one aimed at industrializing the process of drug discovery. Without sequencing, though, the platform wouldn’t exist.
Sequencers are analog-to-digital converters. They transform physical base pairs of DNA (or RNA) into digital information to be passed on to a growing body of analytical tools. Unlike most electronics, sequencers require biology to function. By reading, comparing, and annotating strings of DNA code, we gain an understanding of how these letters come together to form words (genes) and how those words form sentences (gene pathways) and so on.
Today, market forces have selected for three sequencing archetypes—patterned flow-cells, semiconductors, and nanopores. Each sensing method tells a rhyming story about how hardware, software, and biological subsystems work in confluence.
Flow-cell-based sequencers (e.g., Illumina (ILMN) sequencers) arrange monoclonal clusters of ~300-letter DNA fragments on a solid support tightly arrayed with nanoscopic landing pads. Under a camera’s watchful eye, a coordinated series of liquid washes containing dye-labeled DNA bases and custom-engineered enzymes will bind, measure, cleave, and extend the myriad DNA clusters. Onboard software algorithms watch each cycle, converting light signals into DNA base calls while being careful to filter out impure, noisy clusters from the final readout.
Semiconductor-based sequencers (e.g., PacBio (PACB) sequencers) carefully load ~20,000-letter DNA circles into millions of zeptoliter (10-21 L) confinement chambers nano-etched into a metal substrate. Smaller than the wavelength of incident light, these so-called zero-mode waveguide (ZMW) chambers allow a camera to watch in real-time as highly-engineered polymerase enzymes incorporate dye-labeled DNA bases into the template DNA circles. Uniquely, these enzymes are armed with “photoshields”—chemical structures that shade fragile DNA molecules from the prolonged assault of laser light necessary to read off tens of thousands of bases. Finally, the camera passes a tensor object to a transformer algorithm, converting twinkling light pulses into sequence data.
Nanopore-based sequencers (e.g., Oxford Nanopore (ONT.L) sequencers) don’t use cameras. Instead, ~100,000-letter DNA fragments are unzipped by helicase enzymes before being ratcheted through an array of biological nanopores embedded in a membrane support. As each stretch of nucleotides passes through a pore, it creates a characteristic electronic trace (a squigglegram) that is translated by a deep-learning algorithm into sequence information.
Sequencing is just one example of the boundary between biology and technology fully dissolving. However, it’s one I’m passionate about given its foundational position within modern biotechnology. Reinvigorated sequencing vendor competition is unlocking new perspectives into previously obscured layers of biology—methylation, chromatin structure, and structural variation, just to name a few. I’m eager to see how the next wave of entrepreneurs harnesses this information for the betterment of humankind.
There are many more molecular tools to build and many more facets of biology to digitize. We’re excited to see new developments in spatial biology, biosensors and analytical tools, therapeutic platforms, software infrastructure solutions for the life sciences, and cell programming technologies—amongst other areas. I’m thrilled to be partnering with entrepreneurs at the earliest stages of technology development whether bench-side or breadboard. What’s more, I’m grateful to share this experience with Dimension’s founders who’ve entrusted me with forging our research practice. The future is bright and we’re just getting started.