Researchers recently showed that a computer could “learn” from many examples of protein folding to predict the 3D structure of proteins with great speed and precision. Now a recent study in the journal Science shows that a computer also can predict the 3D shapes of RNA molecules . This includes the mRNA that codes for proteins and the non-coding RNA that performs a range of cellular functions.
This work marks an important basic science advance. RNA therapeutics—from COVID-19 vaccines to cancer drugs—have already benefited millions of people and will help many more in the future. Now, the ability to predict RNA shapes quickly and accurately on a computer will help to accelerate understanding these critical molecules and expand their healthcare uses.
Like proteins, the shapes of single-stranded RNA molecules are important for their ability to function properly inside cells. Yet far less is known about these RNA structures and the rules that determine their precise shapes. The RNA elements (bases) can form internal hydrogen-bonded pairs, but the number of possible combinations of pairings is almost astronomical for any RNA molecule with more than a few dozen bases.
In hopes of moving the field forward, a team led by Stephan Eismann and Raphael Townshend in the lab of Ron Dror, Stanford University, Palo Alto, CA, looked to a machine learning approach known as deep learning. It is inspired by how our own brain’s neural networks process information, learning to focus on some details but not others.
In deep learning, computers look for patterns in data. As they begin to “see” complex relationships, some connections in the network are strengthened while others are weakened.
One of the things that makes deep learning so powerful is it doesn’t rely on any preconceived notions. It also can pick up on important features and patterns that humans can’t possibly detect. But, as successful as this approach has been in solving many different kinds of problems, it has primarily been applied to areas of biology, such as protein folding, in which lots of data were available for researchers to train the computers.
That’s not the case with RNA molecules. To work around this problem, Dror’s team designed a neural network they call ARES. (No, it’s not the Greek god of war. It’s short for Atomic Rotationally Equivariant Scorer.)
To start, the researchers trained ARES on just 18 small RNA molecules for which structures had been experimentally determined. They gave ARES these structural models specified only by their atomic structure and chemical elements.
The next test was to see if ARES could determine from this small training set the best structural model for RNA sequences it had never seen before. The researchers put it to the test with RNA molecules whose structures had been determined more recently.
ARES, however, doesn’t come up with the structures itself. Instead, the researchers give ARES a sequence and at least 1,500 possible 3D structures it might take, all generated using another computer program. Based on patterns in the training set, ARES scores each of the possible structures to find the one it predicts is closest to the actual structure. Remarkably, it does this without being provided any prior information about features important for determining RNA shapes, such as nucleotides, steric constraints, and hydrogen bonds.
It turns out that ARES consistently outperforms humans and all other previous methods to produce the best results. In fact, it outperformed at least nine other methods to come out on top in a community-wide RNA-puzzles contest. It also can make predictions about RNA molecules that are significantly larger and more complex than those upon which it was trained.
The success of ARES and this deep learning approach will help to elucidate RNA molecules with potentially important implications for health and disease. It’s another compelling example of how deep learning promises to solve many other problems in structural biology, chemistry, and the material sciences when—at the outset—very little is known.
CRISPR gene-editing technology has tremendous potential for making non-heritable DNA changes that can treat or even cure a wide range of devastating disorders, from HIV to muscular dystrophy Now, a recent animal study shows that another CRISPR system—targeting viral RNA instead of human DNA—could work as an inhaled anti-viral therapeutic that can be preprogrammed to seek out and foil potentially almost any flu strain and many other respiratory viruses, including SARS-CoV-2, the coronavirus that causes COVID-19.
How can that be? Other CRISPR gene-editing systems rely on a sequence-specific guide RNA to direct a scissor-like, bacterial enzyme (Cas9) to just the right spot in the genome to cut out, replace, or repair disease-causing mutations. This new anti-viral CRISPR system also relies on guide RNA. But the guide instead directs a different bacterial enzyme, called Cas13a, to the right spot in the viral genome to bind and cleave viral RNA and stop viruses from replicating in lung cells.
The findings, recently published in the journal Nature Biotechnology , come from the lab of Philip Santangelo, Georgia Institute of Technology and Emory University, Atlanta. Earlier studies by other groups had shown the potential of Cas13 for degrading the RNA of influenza viruses in a lab dish [2,3]. In this latest work, Santangelo and colleagues turned to mice and hamsters to see whether this enzyme could actually work in the lung tissue of a living animal.
What’s interesting is how Santangelo’s team did it. Rather than delivering the Cas13a protein itself to the lungs, the CRISPR system works by supplying a messenger RNA (mRNA) with the instructions to make the anti-viral Cas13a protein. This is the same idea as the Pfizer and Moderna mRNA-based COVID-19 vaccines, which temporarily direct your muscle cells to produce viral spike proteins that launch an immune response. In this case, the lung cells translate the Cas13a mRNA to produce the protein. Directed by the guide RNA that was also delivered to the same cells, Cas13a degrades the viral RNA and stops the infection. Because mRNA doesn’t enter the cell’s nucleus, it doesn’t interact with DNA and raise potential concerns about causing unwanted genetic changes.
The researchers designed guide RNAs that were specific to a shared, highly conserved portion of influenza viruses involved in replicating their genome and infecting other cells. They also designed another set directed to key portions of SARS-CoV-2.
Next, they delivered the Cas13a mRNA and guides straight to the lungs of animals using an adapted nebulizer, just like those used to deliver medicines to the lungs of people. In mice with influenza, Cas13a degraded influenza RNA in the lungs and the animals recovered without any apparent side effects. In SARS-CoV-2-infected hamsters, the same approach limited the virus’s ability to replicate in cells as the animals COVID-19-like symptoms improved.
The findings are the first to show that mRNA can be used to express the Cas13a protein in living lung tissue, not just in cells in a dish. It’s also the first to show that the bacterial Cas13a protein is effective at slowing or stopping replication of SARS-CoV-2. The latter raises hope that this CRISPR system could be quickly adapted to fight any future novel coronaviruses that develop the ability to infect humans.
The researchers report that this approach has potential to work against the vast majority—99 percent—of the flu strains that have circulated around the world over the last century. It also should be equally effective against the new and more contagious variants of SARS-CoV-2 now circulating around the globe. While more study is needed to understand the safety of such an anti-viral approach before trying it in humans, what’s clear is basic research advances like this one hold great potential for helping us to fight life-threatening respiratory viruses of the past, present, and future.