Skip to Content

A Step Closer to Perfect 3-D Data Storage

Using new classes of engineered fluorescent proteins, scientists get closer to storing data at unprecedented densities.

In the introduction to a paper in press in the Journal of Biotechnology, Virgile Adam of the Katholieke Universiteit Leuven in Belgium, along with a long list of collaborators from other institutions, describe the ultimate in holographic (three-dimensional) data storage: a chemically pure crystal composed solely of proteins that can be read and reversibly switched between at least two different states using nothing but light.

Embedded within the proper array of lasers (it would take at least two), such a crystal would represent something approaching the theoretical limit of data density in a storage medium: each bit would be represented by a single molecule.

With their latest work, Adam et al. take us a step closer to this dream, at least in the laboratory. Not only did they manage to encode and read data on chemically pure crystals of mutant descendants of fluorescent proteins originally derived from corals, but they also demonstrated that at least one of these proteins, known as IrisFP, actually has the ability to store data in four different states, versus the two different states (on and off) encoded by a traditional bit. In other words, this protein could store data in base 4 instead of base 2.

In order to understand what’s going on here, it helps to understand the substrate on which these researchers propose to store data: in a traditional optical disc like a DVD or CD, bits are stored as microscopic, physical pits or bumps, each representing either a 0 or a 1.

In a system using fluorescent proteins, researchers exploit the fact that a whole host of proteins discovered (and in some cases engineered) by biologists will either fluoresce or not when hit with a laser pulse of light, depending on their physical configuration. What’s more, some of these proteins are reversibly switchable–that is, given the proper stimulation with light, they can be turned “on” (a state in which they fluoresce when “read” using light of an appropriate wavelength) or “off” (a state in which they fluoresce very little or not at all when “read”).

In their research, Adam et al. experimented with a number of fluorescent proteins on both two dimensional surfaces (as an analog to traditional disc-style storage media) and as three-dimensional crystals. Growing proteins into chemically pure crystals is an ancient and dark art still practiced by scientists wishing to uncover the three-dimensional structure of a protein by shining X-rays through a crystal comprised of that protein. (Without this technique, we never would have uncovered the structure of DNA, for instance.)

When great numbers of proteins are in crystalline form, there is the additional challenge of how to address a single bit within the crystal. A laser must shine all the way through the crystal without lighting up or altering any of the other proteins that are between the origin of the laser and the protein of interest, which may be smack in the middle.

This problem is solved by another innovation from the biological sciences, known as two-photon microscopy. Without getting too heavy into the physics, the way this works is that two pulsing laser beams at right angles to each other both emit photos that by themselves don’t have enough energy to excite a protein. When the two beams are aimed so that they intersect within the crystal in the exact location of the protein of interest - i.e., the bit to be addressed - and two photons, each from a different laser, happen to hit that protein at the same time, it lights up. Meanwhile, both photons traveled all the way through the crystal to the point at which they simultaneously smacked into the same protein without having any effect on any of the proteins they passed on their journey, allowing for selective addressing in three dimensions.

The x-ray structure of IrisFP, surrounded by pictures of crystals of IrisFP in its different forms (switched on/off, green/red), recorded while fluorescing, courtesy ESRF

IrisFP is one of the fluorescent proteins that can be addressed with two-photon microscopy. In addition, it has not two states (on and off) but four: green on/off, and red on/off. The switch from the green to the green to the red state is not reversible, which means that data stored with this protein in base 4 would have to be of the “write once, read many” type. But within its green and red states, it is fully reversible.

This work is preliminary at best–the crystals grown in the course of this research were 100 micrometres in their smallest dimension, or barely the width of a human hair, and it’s not yet possible to focus laser beams precisely enough to address individual proteins. (Indeed, it might not even be physically possible, if the protein is smaller than the wavelength of the laser.)

However, in the future, as we move from disc-based storage media to polymer-based holographic (3D) storage media, the use of fluorescent proteins in a three dimensional matrix is a logical next step. Intriguingly, these storage media might not be designed so much as evolved - in order to create stable fluorescent proteins that can last as long as possible and don’t require an aqueous environment, researchers propose mutating them one amino acid at a time and testing the resulting proteins.

Follow Christopher Mims on Twitter. Or contact Mims if you would like to share something you’d like to see covered in this blog.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.