The Microscope That Sees the Invisible

TL;DR

• Researchers at Biohub and UC Berkeley have built a laser phase plate — a device housing a laser 100 million times brighter than the surface of the Sun — and integrated it into a cryo-electron microscope.
• The result: for the first time, scientists can clearly image more than 90% of human proteins that were previously too small for cryo-EM to capture.
• Two papers published: one in Science (June 11, 2026), one preprint on bioRxiv describing an even more advanced dual-laser system. A third paper in Nature Communications provides the theoretical foundation for the dual-laser design.
• The next frontier is cryo-electron tomography — imaging proteins inside living cells, in their natural environment, where they interact, assemble, and malfunction.
• This is not an incremental improvement. It is a step-function change for structural biology and drug discovery.

What Happened

For more than a decade, cryo-electron microscopy has been the reigning champion of structural biology. It earned its inventors the 2017 Nobel Prize in Chemistry. It has revealed the atomic architecture of ribosomes, ion channels, and viral spike proteins — the molecular machines that run every living cell. It has become, in the words of the Nobel Committee, the method that "moved biochemistry into a new era."

But cryo-EM has a brutal limitation: it cannot see small things clearly.

More than 90% of the proteins inside a human cell are too small to generate enough contrast in a cryo-EM image. They are, for all practical purposes, invisible. The technique works beautifully for large molecular complexes — ribosomes, viruses, massive enzyme assemblies — but the average human protein falls below the detection threshold. The very molecules most likely to be drug targets, disease mechanisms, and answers to decades-old biological questions have been sitting in a blind spot.

That changed last week.

On June 11, 2026, researchers at Biohub and UC Berkeley announced they had successfully built and demonstrated a laser phase plate — a device that shifts the phase of the electron beam inside a cryo-electron microscope using one of the most intense steady-state lasers ever constructed. The laser, concentrated into a spot roughly 1/1000th the width of a human hair, reaches an intensity of 350–400 gigawatts per square centimetre. That is approximately 100 million times brighter than the surface of the Sun.

The result is dramatically improved contrast. Small proteins that were previously at or beyond the detection limit — like haemoglobin, the oxygen-carrying protein in blood — now resolve clearly. The improvement is not subtle. It is the difference between a smudge and a structure.

The UC Berkeley paper, published in Science, demonstrated improved resolution across six different biological samples of varying sizes and preparation quality. Critically, the smaller the sample, the greater the improvement — exactly what the field has needed for two decades. For the most challenging specimens — small proteins with imperfect preparation — the laser produced what Müller called "a very considerable advantage."

Biohub simultaneously published a preprint describing a next-generation dual-laser system with twice the complexity, using two perpendicular laser beams at lower individual power. This design reduces component burn risk, suppresses optical aberrations, and eliminates a known artefact of single-beam systems called ghost images — faint copies of high-contrast objects that can obscure the even fainter biological signal researchers are actually trying to see.

A third paper, published the previous week in Nature Communications, provided the theoretical foundation for the dual-laser concept, authored by Müller and his UC Berkeley colleagues.

Three papers. Two working instruments. One field transformed.

The Physics: Why This Works

To understand why the laser phase plate matters, you need to understand the problem it solves — and the problem is almost a century old.

In 1930, the Dutch physicist Frits Zernike realised something that had escaped every microscopist before him. When light passes through a biological sample — a cell, say — two things happen. The light is scattered, which changes its brightness, or amplitude. But it is also slowed down, which shifts its phase — the timing of the peak of the waveform — by a tiny amount.

That phase shift is invisible to the human eye. But Zernike discovered that if you also shift the unscattered light by 90 degrees, the phase differences become visible as brightness differences. Suddenly, transparent specimens became vivid. Zernike won the 1953 Nobel Prize in Physics for this insight, and the phase-contrast light microscope became a standard tool in every biology lab on Earth.

By the early 1940s, scientists were already asking: could you do the same thing with an electron microscope? An electron beam has roughly 10,000 times the resolving power of visible light. If you could apply Zernike's trick to electrons, you could see individual proteins — the molecular machines that run every living cell.

The problem was the phase plate itself. In a light microscope, you can insert a small glass element to shift the phase of the unscattered light. In an electron microscope, any physical material placed in the electron beam gets bombarded, charged, and damaged — ruining the image. For 80 years, this was the unsolved problem.

One partial solution emerged in the 2010s: the Volta phase plate, which used a thin carbon film that exploited charging as a feature rather than a bug. It worked — sometimes. But it was unstable, it blurred high-resolution detail, and it never became a routine tool. The contrast problem remained.

In 2010, Holger Müller and Robert Glaeser proposed a radically different approach. Instead of putting a physical object in the electron beam, they would use a laser. The laser's electromagnetic field would interact with the electrons, shifting their phase without any material in the way. The catch: because light barely interacts with electrons, the laser would have to be extraordinarily intense — focused to a spot measured in microns, sustained for hours, and stable to within nanometres.

Most of the field considered it nearly impossible. Müller spent 15 years proving them wrong.

The Engineering Feat

The laser phase plate is, by any measure, one of the most extraordinary pieces of precision engineering ever built for biology.

Inside the device, a laser beam bounces between two concave mirrors almost 10,000 times, building up to the extraordinary intensity required. The mirrors themselves are polished to "atomic-level smoothness" — a surface roughness of less than one angstrom, roughly the diameter of a single atom.

"The mirrors must be extremely lossless to prevent them from melting," Müller said, "and in fact are so lossless that they barely warm up, despite being bombarded by a laser that could easily cut inches of steel."

The alignment precision is equally staggering. The mirrors must be angled to within 1/1000th of a degree. The laser beam and electron beam must be aligned to within 50 nanometres — on a standing wave 500 nanometres across — and held there for half an hour at a stretch during data acquisition.

"It's like a surfer trying to hold perfectly to the peak of a wave, not for seconds, but for half an hour at a stretch," said Bridget Carragher, founding technical director of imaging at Biohub.

The entire optical cavity housing this system is less than four inches wide, tucked inside a Thermo Scientific Krios microscope that stands 14 feet tall. Müller named the complete system Theia, after the ancient Greek Titaness of light and radiance.

"It's 75 kilowatts focused to a few microns," Müller said. "That's more powerful than what you use for welding. It's more power than a military laser. It builds up the brightest continuous laser focus ever."

Theia is not just a cryo-EM with a laser bolted on. It is a custom instrument, built in collaboration with Thermo Fisher Scientific, with extra electron optics that give it better resolution than a standard Krios even without the laser engaged. "Theia is the Formula 1 microscope," Müller said. "With the addition of the laser phase plate, we hope that it really becomes the world's best instrument overall."

Biohub's dual-laser system — the crossed laser phase plate, or xLPP — takes the concept further. Two perpendicular laser beams, each in its own cavity, operate at roughly half the power of the single-beam design. This reduces stress on the mirrors, suppresses ghost images, and makes the system easier to operate. "I like to say ours is not street legal," Müller said of Theia. "It's optimised for peak performance, so it requires a bit more user training. The other configurations of the microscope that are in the works are more user-friendly."

What It Actually Means

The laser phase plate solves the fundamental contrast problem that has limited cryo-EM since its inception. But the implications cascade outward in ways that matter far beyond the microscopy community.

The protein universe expands

Here is the core fact: the average human protein is too small to be imaged by today's cryo-EM. Every one of those proteins is a potential drug target, a potential disease mechanism, a potential answer to a question biologists have been asking for decades. The laser phase plate opens the door to seeing all of them.

"If you look at all the proteins in a human, they all have various sizes," Müller said. "And all of these proteins are potential disease mechanisms and drug targets. The problem is, the average human protein is too small to be imaged by cryo-EM. The laser phase plate could fill an enormous gap in our knowledge of protein structures that can't be processed with today's cryo-EM."

Scott Fraser, president of dynamic imaging at Biohub, put the numbers in stark terms: "Rough estimates suggest that scientists can image 10% of the human proteome in purified form using existing cryo-EM — and fewer than 1% of proteins in their native cellular environment." The laser phase plate could push that figure past 50%.

The real prize: cryo-electron tomography

The most profound impact, however, will not be in imaging individual proteins in isolation. It will be in cryo-electron tomography (cryo-ET) — a variant of cryo-EM that captures proteins inside actual cells, in their natural environment, revealing how molecular machines assemble, interact, and malfunction in disease.

Cryo-ET today is painfully slow and frustratingly low-contrast. "Doing a cell biology experiment with cryo-ET today can take up a postdoc's entire career," said David Agard, founding scientific director of imaging at Biohub. "We need to speed that up, and the laser phase plate, along with better processing, all working seamlessly with AI algorithms, will get us there."

Carragher described the challenge with a metaphor that has become the field's rallying cry: "With cryo-ET, we're looking at small, very complicated cellular material that's incredibly crowded inside the cell. It's like a forest of trees, and you're trying to find one leaf on one tree in there. Cryo-ET needs a dramatic step forward in contrast, so we can start to see what's going on inside the cell. That's what the laser phase plate promises to give us."

Biohub scientists have already used standard cryo-ET to image structures called lysosomes in disease states. Once dismissed as mere cellular garbage disposal units, lysosomes are now understood to be complex signalling hubs. Defects in their function have been linked to dozens of rare diseases, as well as common neurodegenerative conditions including Alzheimer's disease. In experiments with mutated lysosomes, Biohub scientists can see that something is wrong — but they lack the high-contrast images needed to reveal how specific defects affect protein–protein interactions.

The laser phase plate changes that equation.

AI meets imaging

The timing is not accidental. The laser phase plate arrives at a moment when AI-driven protein structure prediction — AlphaFold and its successors — has already transformed computational biology. But AI prediction and experimental imaging are complements, not substitutes. AI can predict a structure; cryo-EM can verify it. AI can propose a drug target; cryo-EM can show you the binding site. And for proteins that AI cannot yet predict reliably — disordered regions, dynamic complexes, transient interactions — cryo-EM remains the only game in town.

"With advances in AI, this breakthrough in contrast will start to open up a new frontier in structural biology," said Alex Rives, Biohub's head of science, "that will allow us to see the molecular machines of the cell, and how they assemble into far more complex and dynamic systems, and understand how they work."

The Arc of the Story

It is worth stepping back to appreciate the full arc of this achievement, because it is rare in modern science.

The phase-contrast problem was first posed for electron microscopy in 1942. It resisted solution for 68 years. In 2010, Müller and Glaeser published the theoretical proposal. In 2019, they demonstrated a working prototype in an ageing microscope in Glaeser's lab — a proof of principle published in Nature Methods. In 2021, Biohub made what Stephani Otte, its vice president of imaging science, called "a big bet" — funding Müller to purchase a custom Krios and build Theia. In 2025, installation and final development were completed with support from a Berkeley Lab LDRD award. Within days, the system was producing striking high-contrast images.

On June 11, 2026, the Science paper landed. The preprint followed. The Nature Communications theoretical paper had appeared the week before.

Sixteen years from theory to demonstration. Eighty-four years from problem to solution.

"This technology is a step function change for biology," Otte said. "We are going to be able to see how molecules interact in the cell, and that is going to change everything about how we understand disease."

Stakeholder Landscape

Who benefits directly: Structural biologists, drug discovery teams, and anyone working on diseases caused by proteins currently too small to image. Pharmaceutical companies with cryo-EM pipelines should be paying very close attention. The technology is especially relevant for teams working on membrane proteins, intrinsically disordered proteins, and small signalling molecules — all categories that have been largely invisible to cryo-EM.

Who benefits second-order: Patients. The majority of drug targets are proteins. If you can see the target, you can design against it. This technology will accelerate drug discovery across oncology, neurodegeneration, infectious disease, and rare genetic disorders. The lysosome example is instructive: dozens of rare lysosomal storage diseases have known genetic causes but poorly understood molecular mechanisms. The laser phase plate could change that.

Who should be paying attention now: Research institutions with cryo-EM facilities. Müller expects commercial microscopes fitted with laser phase plates to become available "in the coming years." Early adopters will have a substantial advantage. Biohub's CryoET Data Portal — which freely shares all tomography data, including tens of thousands of annotated tomograms — is the place to watch.

Who benefits from the noise: Cryo-EM manufacturers, particularly Thermo Fisher Scientific, which collaborated on both instruments. AI-based protein structure prediction companies and cryo-ET data processing startups also stand to gain, as the laser phase plate will generate vast new datasets that require analysis.

Who is not affected (yet): Clinical practice. This is a research-tool advance, not a therapy. The timeline from better imaging to approved drugs is measured in years, not months. Patients should not expect new treatments next year because of this — but they should expect new treatments that would not have been possible without it.

Cross-Layer Implications

The laser phase plate sits at the intersection of several trends that amplify each other:

• AI + imaging: The combination of AI-driven structure prediction and high-contrast cryo-EM/ET creates a feedback loop. Better images train better AI; better AI guides better imaging experiments. Biohub is explicitly building this pipeline.
• Open science: Biohub's CryoET Data Portal makes all tomography data freely available. This is not typical for a technology this expensive and this competitive. It accelerates the entire field.
• Manufacturing precision: The mirrors in the laser phase plate are polished to atomic-level smoothness. The alignment tolerances are measured in thousandths of a degree and tens of nanometres. This is precision engineering at a level that pushes the boundaries of what is manufacturable — and it will have spillover effects in optics, semiconductor fabrication, and quantum sensing.
• The talent pipeline: Cryo-ET experiments currently take "a postdoc's entire career." Compressing that timeline changes who can do the science — and how many questions get asked.

What This Means for You

If you work in biomedical research or drug discovery: Track Biohub's CryoET Data Portal. The dual-laser preprint on bioRxiv is essential reading. Begin evaluating whether your targets of interest fall into the "currently too small for cryo-EM" category — if so, the laser phase plate changes your experimental roadmap. The technology is particularly relevant for targets below ~70 kilodaltons, which includes roughly 90% of the human proteome. Müller's team is pushing toward imaging proteins as small as 17 kilodaltons.

If you invest in biotech or life sciences tools: The cryo-EM market is about to expand its addressable protein universe by roughly an order of magnitude. Companies building cryo-EM workflows, AI-based protein structure prediction, and cryo-ET data processing pipelines stand to benefit disproportionately. Thermo Fisher's collaboration with both the Berkeley and Biohub teams positions it as the likely commercialisation partner.

If you are a general reader: This is one of those moments where a tool changes what it is possible to know. The last time this happened in structural biology — when cryo-EM itself became practical — it won a Nobel Prize. The laser phase plate is that kind of advance. It will not produce a new drug next year. But a decade from now, when a drug exists for a disease that currently has no treatment, there is a reasonable chance the target was first seen clearly through a laser phase plate.

Uncertainty Ledger

• The dual-laser system is still in preprint — not yet peer-reviewed at the same level as the single-laser Science paper. The Nature Communications theoretical paper is peer-reviewed, but the Biohub experimental demonstration is not.
• Cryo-ET with the laser phase plate has not yet been demonstrated. Both teams believe it is the next frontier, and Carragher says they are "optimistic we'll be doing data collection by the end of the year." But it remains prospective.
• Commercial availability timeline is uncertain. Müller says "coming years," but integration into production microscopes involves Thermo Fisher and other industrial partners — timelines can slip. Müller himself notes that Theia is "not street legal" and requires expert operation.
• The technology requires extraordinary precision to operate. Whether it can be made robust enough for routine lab use — as opposed to being a Formula 1 instrument in a handful of elite labs — remains an open question.
• The ultimate resolution limit for the smallest proteins (below ~17 kilodaltons) is unknown. Müller's team is pushing toward that threshold, but it has not been demonstrated yet.

Bottom Line

The laser phase plate is the most significant advance in biological imaging since cryo-EM itself. It solves the contrast problem that has kept more than 90% of human proteins invisible — a problem that resisted solution for 84 years. When combined with cryo-electron tomography, it will let scientists watch molecular machines operate inside living cells for the first time. That changes how we understand disease. It changes how we find drugs to treat it. And it arrives at precisely the moment when AI-driven structure prediction needs experimental validation to fulfil its own promise. This is not hype. This is a tool that makes the invisible visible. The science it enables has only just begun.

Sources: Biohub press release (June 11, 2026) [Tier 1]; UC Berkeley News — Robert Sanders (June 11, 2026) [Tier 1]; Berkeley Lab News Center (June 11, 2026) [Tier 1]; Petrov, Zhang et al., "Laser phase plate improves structure determination of small proteins by cryo-EM," Science (June 11, 2026) [Tier 1]; Yu, Olshin, Carragher, Agard et al., "Crossed laser phase plate for cryo-electron microscopy," bioRxiv preprint (June 2026) [Tier 2]; Müller et al., dual-laser concept, Nature Communications (June 2026) [Tier 1]; phys.org / UC Berkeley (June 11, 2026) [Tier 2]; GEN News (June 12, 2026) [Tier 2]; Biohub blog — "Making the invisible visible" (June 11, 2026) [Tier 2]