Friday, January 05, 2018

What to look out for in science in 2018

I wrote a piece for the Guardian on what we might expect in science, and what some of the big issues will be, in 2018. It was originally somewhat longer than the paper could accommodate, explaining some issues in more detail. Here’s that longer version.


Quantum computers
This will be the year when we see a quantum computer solve some computational problem beyond the means of the conventional ‘classical’ computers we currently use. Quantum computers use the rules of quantum mechanics to manipulate binary data – streams of 1s and 0s – and this potentially makes them much more powerful than classical devices. At the start of 2017 the best quantum computers had only around 5 quantum bits (qubits), compared to the billions of transistor-based bits in a laptop. By the close of the year, companies like IBM and Google said that they are testing devices with ten times that number of qubits. It still doesn’t sound like much, but many researchers think that just 50 qubits could be enough to achieve “quantum supremacy” – the solution of a task that would take a classical computer so long as to be practically impossible. This doesn’t mean that quantum computers are about to take over the computer industry. For one thing, they can so far only carry out certain types of calculation, and dealing with random errors in the calculations is still extremely challenging. But 2018 will be the year that quantum computing changes from a specialized game for scientists to a genuine commercial proposition.

Quantum internet
Using quantum rules for processing information has more advantages than just a huge speed-up. These rules make possible some tricks that just aren’t imaginable using classical physics. Information encoded in qubits can be encrypted and transmitted from a sender to a receiver in a form that can’t be intercepted and read without that eavesdropping being detectable by the receiver, a method called quantum cryptography. And the information encoded in one particle can in effect be switched to another identical particle in a process dubbed quantum teleportation. In 2017 Chinese researchers demonstrated quantum teleportation in a light signal sent between a ground-based source and a space satellite. China has more “quantum-capable” satellites planned, as well as a network of ground-based fibre-optic cables, that will ultimately comprise an international “quantum internet”. This network could support cloud-based quantum computing, quantum cryptography and surely other functions not even thought of yet. Many experts put that at a decade or so off, but we can expect more trials – and inventions – of quantum network technologies this year.

RNA therapies
The announcement in December of a potential new treatment for Huntington’s disease, an inheritable neurodegenerative disease for which there is no known cure, has implications that go beyond this particularly nasty affliction. Like many dementia-associated neurodegenerative diseases such as Parkinson’s and Alzheimer’s, Huntington’s is caused by a protein molecule involved in regular brain function that can ‘misfold’ into a form that is toxic to brain cells. In Huntington’s, which currently affects around 8,500 people in the UK, the faulty protein is produced by a mutation of a single gene. The new treatment, developed by researchers at University College London, uses a short strand of DNA that, when injected into the spinal cord, attaches to an intermediary molecule involved in translating the mutated gene to the protein and stops that process from happening. The strategy was regarded by some researchers as unlikely to succeed. The fact that the current preliminary tests proved dramatically effective at lowering the levels of toxic protein in the brain suggests that the method might be a good option not just for arresting Huntington’s but other similar conditions, and we can expect to see many labs trying it out. The real potential of this new drug will become clearer when the Swiss pharmaceuticals company Roche begins large-scale clinical trials.

Gene-editing medicine
Diseases that have a well defined genetic cause, due perhaps to just one or a few genes, can potentially be cured by replacing the mutant genes with properly functioning, healthy ones. That’s the basis of gene therapies, which have been talked about for years but have so far failed to deliver on their promise. The discovery in 2012 of a set of molecular tools, called CRISPR-Cas9, for targeting and editing genes with great accuracy has revitalized interest in attacking such genetic diseases at their root. Some studies in the past year or two have shown that CRISPR-Cas9 can correct faulty genes in mice, responsible for example for liver disease or a mouse form of muscular dystrophy. But is the method safe enough for human use? Clinical trials kicked off in 2017, particularly in China but also the US; some are aiming to suppress the AIDS virus HIV, others to tackle cancer-inducing genetic mutations. It should start to become clearing 2018 just how effective and safe these procedures are – but if the results are good, the approach might be nothing short of revolutionary.

High-speed X-ray movies
Developing drugs and curing disease often relies on an intimate knowledge of the underlying molecular processes, and in particular on the shape, structure and movements of protein molecules, which orchestrate most of the molecular choreography of our cells. The most powerful method of studying those details of form and function is crystallography, which involves bouncing beams of X-rays (or sometimes of particles such as electrons or neutrons) off crystals of the proteins and mathematically analysing the patterns in the scattered beams. This approach is tricky, or even impossible, for proteins that don’t form crystals, and it only gives ‘frozen’ structures that might not reflect the behaviour of floppy proteins inside real cells. A new generation of instruments called X-ray free-electron lasers, which use particle-accelerator technologies developed for physics to produce extremely bright X-ray beams, can give a sharper view. In principle they can produce snapshots from single protein molecules rather than crystals containing billions of them, as well as offering movies of proteins in motion at trillions of frames per second. A new European X-ray free-electron laser in Hamburg inaugurated in September is the fastest and brightest to date, while two others in Switzerland and South Korea are starting up too, and another at Stanford in California is getting an ambitious upgrade. As these instruments host their first experiments in 2018, researchers will acquire a new window into the molecular world.

100,000 genomes
By the end of 2018 the private company Genomics England, set up by the UK Department of Health, should have completed its goal of reading the genetic information in 100,000 genomes of around 75,000 voluntary participants. About a third of these people will be cancer patients, who will have a separate genome read from cancer cells and healthy cells; the others will be people with rare genetic diseases and their close relatives. With such a huge volume of data, it should be possible to identify gene mutations linked to cancer and to some of the many thousands of known rare diseases. This information could help diagnoses of cancer and disease, and perhaps also to improve treatments. For example, a gene mutation that causes a rare disease (one of which is likely to affect around one person in 17 at some point in their lives) supplies a possible target for new drugs. Genetic information for cancer patients can also help to tailor specific treatments, for example by identifying those not at risk of side effects from what can otherwise be effective anti-cancer drugs.

Gravitational-wave astronomy
The 2017 Nobel prize in physics was awarded to the chief movers behind LIGO, the US project to detect gravitational waves. These are ripples in spacetime caused by extreme astrophysical events such as the merging of two neutron stars or black holes, which have ultra-strong gravitational fields. The ripples produce tiny changes in the dimensions of space itself as they pass, which LIGO – comprising two instruments in Washington State and Louisiana – detects from changes in the distances travelled by laser beams sent along channels to mirrors a few kilometres away. The first gravitational wave was detected in late 2015 and announced in 2016. Last year saw the announcement of a few more detections, including one in August from the first known collision of two neutron stars. Gravitational-wave detectors now also exist or are being built in Europe, Korea and Japan, while others are planned that will use space satellites. The field is already maturing into a new form of astronomy that can ‘see’ some of the most cataclysmic events in the universe – and which so far fully confirm Einstein’s theory of general relativity, which explains gravitation. We can expect to see more cataclysmic events detected in 2018 as gravitational-wave astronomy becomes a regular tool in the astronomer’s toolkit.

Beyond the standard model
It’s a glorious time for fundamental physics – but not necessarily for the reasons physicists might hope. The so-called standard model of particle physics, which accounts for all the known particles and forces in nature, was completed in 2013 with the discovery of the Higgs boson using the Large Hadron Collider (LHC), the world’s most powerful particle accelerator, at CERN in Switzerland. The trouble is, it can’t be the whole story. The two most profound theories of physics – general relativity (which describes gravity) and quantum mechanics – are incompatible; they can’t both be right as they stand. That problem has loomed for decades, but it’s starting to feel embarrassing. Physicists have so far failed to find ways of breaking out beyond the standard model and finding ‘new physics’ that could show the way forward. String theory offers one possible route to a theory of quantum gravity, but there’s no experimental evidence for it. What’s needed is some clue from particle-smashing experiments for how to extend the standard model: some glimpse of particles, forces or effects outside the current paradigm. Researchers were hoping that the LHC might have supplied that already – in particular, many anticipated finding support for the theory called supersymmetry which some see as the best candidate for the requisite new physics. But so far there’s been zilch. If another year goes by without any chink in the armour appearing, the head-scratching may turn into hair-pulling.

Crunch time for dark matter
That’s not the only embarrassment for physics. It’s been agreed for decades that the universe must contain large amounts of so-called dark matter – about five times as much, in terms of mass, than all the matter visible as stars, galaxies, and dust. This dark matter appears to exert a gravitational tug while not interacting significantly with ordinary matter or light (whence the ‘dark’) in other ways. But no one has any idea what this dark matter consists of. Experiments have been trying to detect it for years, primarily by looking for very rare collisions of putative dark-matter particles with ordinary particles in detectors buried deep underground (to avoid spurious detections caused by other particles such as cosmic rays) or in space. All have drawn a blank, including results from separate experiments in China, Italy and Canada reported in the late summer and early autumn. The situation is becoming grave enough for some researchers to start taking more seriously suggestions that what looks like dark matter is in fact a consequence of something else – such as a new force that modifies the apparent effects of gravity. This year could prove to be crunch time for dark matter: how long do we persist in believing in something when there’s no direct evidence for it?

Return to the moon
In 2018, the moon is the spacefarer’s destination of choice. Among several planned missions, China’s ongoing unmanned lunar exploration programme called Chang’e (after a goddess who took up residence there) will enter its fourth phase in June with the launch of a satellite to orbit the moon’s ‘dark side’ (the face permanently facing away from the Earth, although it is not actually in perpetual darkness). That craft will then provide a communications link to guide the Long March 5 rocket that should head out to this hidden face of the moon in 2019. The rocket will carry a robotic lander and rover vehicle to gather information about the mineral composition of the moon, including the amount of water ice in the south polar basin. It’s all the prelude to a planned mission in the 2030s that will take Chinese astronauts to the lunar surface. Meanwhile, tech entrepreneur Elon Musk has claimed that his spaceflight business SpaceX will be ready to fly two paying tourists around the moon this year in the Falcon Heavy rocket and the Dragon capsule the company has developed. Since neither craft has yet had a test flight, you’d best not hold your breath (let alone try to buy a ticket) – but the rocket will at least get its trial launch this year.

Highway to hell
Exploration of the solar system won’t all be about the moon, however. The European Space Agency and the Japanese Aerospace Exploration Agency are collaborating on the BepiColombo mission, which will set off in October on a seven-year journey to Mercury, the smallest planet in the solar system and the closest to the Sun. Like the distant dwarf planet Pluto until the arrival of NASA’s New Horizons mission in 2015, Mercury has been a neglected little guy in our cosmic neighbourhood. That’s partly because of the extreme conditions it experiences: the sunny side of the planet reaches a hellish 430 oC or so, and the orbiting spacecraft will feel heat of up to 350 oC – although the permanently shadowed craters of Mercury’s polar regions stay cold enough to hold ice. BepiColombo (named after renowed Italian astronomer Giuseppe Colombo) should provide information not just about the planet itself but about the formation of the entire solar system.

Planets everywhere
While there is still plenty to be learnt about our close planetary neighbours, their quirks and attractions have been put in cosmic perspective by the ever-growing catalogue of “exoplanets” orbiting other stars. Over the past two decades the list has grown to nearly 4,000, with many other candidates still being considered. The majority of these were detected by the Kepler space telescope, launched in 2009, which identifies planets from the very slight dimming of their parent star as the planet passes in front (a ‘transit’). But the search for other worlds will hot up in 2018 with the launch of NASA’s Transiting Exoplanet Survey Satellite, which will monitor the brightness of around 200,000 stars during its two-year mission. Astronomers are particularly interested in finding ‘Earth-like planets’, with a size, density and orbit comparable to that of Earth and which might therefore host liquid water - and life. Such candidates should then be studied in more detail by the James Webb Space Telescope, a US-European-Canadian collaboration widely regarded as the successor to the Hubble Space Telescope, due for launch in spring 2019. The Webb might be able to detect possible signatures of life within the chemical composition of exoplanet atmospheres, such as the presence of oxygen. With luck, within just a couple of years or so we may have good reason to suspect we are not alone in the universe.

Mapping the brain
It’s sometimes said, with good reason, that understanding outer space is easier than understanding inner space. The human brain is arguably the most complex object in the known universe, and while no one seems to be expecting any major breakthrough in 2018 in our view of how it works, we can expect to reach next Christmas with a lot more information. Over the summer of 2017 the €10bn European Human Brain Project got a reboot to steer it away from what many saw as an over-ambitious plan to simulate a human brain on a computer and towards a more realistic goal of mapping out its structure down to the level of connections between the billions of individual neurons. This shift in emphasis was triggered by an independent review of the project after 800 neuroscientists threatened to boycott it in 2014 because of concerns about the way it was being managed. One vision now is to create a kind of Google Brain, comparable to Google Earth, in which the brain structures underpinning such cognitive functions as memory and emotion can be ‘zoomed’ from the large scale revealed by MRI scanning down to the level of individual neurons. Such information might guide efforts to simulate more specific ‘subroutines’ of the brain. But one of the big challenges is simply how to collect, record and organize the immense volume of data these studies will produce.

Making clean energy
Amidst the excitement and allure of brains, genes, planets and the cosmos, it’s easy for the humbler sciences, such as chemistry, to get overlooked. That should change in 2019, which UNESCO has just designated as the International Year of the Periodic Table, chemistry’s organizing scheme of elements. But there are good reasons to keep an eye on the chemical sciences this year too, not least because they may hold the key to some of our most pressing global challenges. Since nature has no reason to heed the ignorance of the current US president, we can expect the global warming trend to continue – and some climate researchers believe that the only way to limit future warming to within 2 oC (and thus to avoid some extremely alarming consequences) is to develop chemical technologies for capturing and storing the greenhouse gas carbon dioxide from the atmosphere. At the start of 2017 a group of researchers warned that lack of investment in research on such “carbon capture and storage” technologies was one of the biggest obstacles to achieving this target. By the end of this year we may have a clearer view of whether industry and governments will rise to the challenge. In the meantime, development of carbon-free energy-generating technologies needs boosting too. The invention last year at the Massachusetts Institute of Technology of a device that uses an ultra-absorbent black “carbon nanomaterial” to convert solar heat to light suggests one way to make solar power more efficient, capturing more of the energy in the sun’s rays than current solar cells can manage even in principle. We can hope for more such innovation, as well as efforts to turn the smart science into commercially viable technologies. Don’t expect any single big breakthrough in these areas, though; success is likely to come, if at all, from a portfolio of options for making and using energy in greener ways.