The field scientists once jokingly called “blobography” have come a long way.
Cryogenic electron microscopy used to be known for providing amorphous images of biomolecules. Now it has become one of the most reliable methods available for viewing the tiniest building blocks of the body. It’s also a major pillar of Gandeeva, a new biotech company that emerged from stealth Monday with a $40 million Series A round. The company plans to combine these high-resolution images with a suite of machine learning tools to accelerate the drug discovery process.
“It took about 15 years of dedicated effort to realize this dream we had when we started: to literally visualize proteins in an electron microscope with atomic resolution. Once we and others showed that this was possible, it was clear to me that this was the critical tool needed to transform and revolutionize drug discovery,” co-founder and CEO Sriram Subramaniam told TechCrunch.
Gandeeva’s thesis, he continues, is about creating a platform that can actually… to learn of current advances in cryo-Em. The idea is that these high-resolution images could help reveal previously invisible pockets for drugs to bind to, if we can find drugs that fit the bill.
“Having the tools to dig for gold is one thing, but you have to know what to do with it — what product to turn it into. And in our case, that’s medicine for patients,” he said.
There are currently many companies tackling the mammoth problem of drug discovery. Gandeeva’s approach, quite bluntly, is that seeing is believing when it comes to finding drug targets in the body.
Numerous scientific breakthroughs have been achieved simply by observing the world around us. But when it comes to the building blocks of the body, this is impossible without specific microscope techniques. For decades, the leading technique in this field has been X-ray crystallography, in which a scientist literally packs proteins or molecules together into a crystal and fires X-rays at it, approximating its shape, size and orientation.
The problem with X-ray crystallography was the… crystallization bit — this process is arduous and time-consuming. The advantage of cryo-Em is that it does not require crystallization. Instead, the molecules are flash-frozen, creating a 2D sheet, which is then fired at with an electron gun. This sheet protects the biomolecule from the electrons and makes it possible to capture detailed images. That sheet also allows scientists to record the movement of biomolecules, a process not possible with a crystal-encased structure.
For example, it is possible to create images of structures as small as two ngströms in diameter – that is one-tenth of a nanometer. (For reference, the width of one human hair is about 1 million ngströms).
There are indications that cryo-EM is in the middle of a tree. As Nature reported in February 2020, some scientists have predicted that by 2024, more protein structures will be determined by cryo-EM than by X-ray crystallography. It’s becoming a bigger part of the scientific toolkit—despite the sometimes prohibitive cost of microscopes and equipment needed to perform the technique—as resolution has vastly improved.
Meanwhile, there are other developments in structural biology that work in Gandeeva’s favor. First, advances in machine learning have made it possible to predict exactly how proteins fold.
In particular, we’ve seen the development of two AI engines capable of predicting how proteins fold: AlphaFold, developed by the AI company DeepMind, which is owned by Alphabet, and RoseTTAFold, developed at the University of Washington. While it used to take hours of lab work to determine protein structures, RoseTTAFold claims to be able to predict that structure in 10 minutes on a regular game console.
Subramaniam has argued that these tools provide an unprecedented level of insight into the structure and function of proteins, but that there will still be gaps that need to be filled (for example, there are some elements of the AI predictions that are less reliable than others). . Cryo-EM, he notes, allows scientists to zoom in on specific parts of a protein, or to capture images of proteins in many different conformational states (think wobbling), perhaps filling those gaps.
“There’s a whole new revolution going on in AI, and I think everyone is wondering, what does all this mean? This combination of AI and cryo-EM, which has always been Gandeeva’s thesis, is really the ticket because it’s not just experimental or just prediction,” he said.
“You can [use] combine these AI-based insights into structural biology and interactions with high-speed, high-throughput precision imaging.”
So far, Gandeeva wants to prove that cryo-EM is really possible fast and easy outside a government or university sponsored context. That’s important, because much of Subramaniam’s work in this area has ended up in those environments.
Subramaniam spent most of his career at the NIH, where he headed the biophysics section of the National Cancer Institute. From there he went on to find the the National Cryo-EM Facility, a government-run laboratory. At the NIH, he hoped to get started on Gandeeva’s cryo-EM-based drug discovery platform, and realized that developing the lab alone would cost billions.
At the time, “VCs weren’t interested in that kind of approach,” he said. But the University of British Columbia does. He left the NIH to become president of Cancer Drug Design at the university.
“Over the years I’ve been here, I’ve set this up to show that I could replicate what we were doing at the NIH. So at UBC I was able to do a prototype and that’s really what convinced investors that this could be done quickly,” he said.
His proof-of-concept came in the form of a rapidly produced cryo-EM image of the spike protein of the Omicron variant, published in Science.
Ultimately, though, Gandeeva doesn’t just pack cryo-EM in the hopes of making beautiful biological pictures — it’s a research platform aimed at reducing the time it takes to create new drugs.
“We really think we can save a lot of time simply because of the power to see exactly where the drug binds and what protein it targets. That kind of information is extremely powerful because it keeps you from going dead ends,” he said. Subramaniam.
The company will have to prove that it can execute this technique at industrial speed and scale, and collect information that cannot be found anywhere else. Gandeeva has a six-year lease for a facility outside of Vancouver, where Subramaniam plans to build out the functionality of its platform.
Internally, the goal is to advance a number of programs to prove they can identify potential drug targets. Subramaniam said that if he had to guess, he would probably apply Gandeeva’s platform to oncology – but that’s not certain.
This round was led by Lux Capital and LEAPS by Bayer. The round includes participation from Obvious Ventures, Amgen Ventures, Amplitude Ventures and Air Street Capital. The company has raised $40 million so far.