A huge planet about the size of Jupiter has been spotted orbiting a tiny white dwarf star 80 light-years away. The discovery by an international team of astronomers is puzzling because the planet should have been swallowed up long ago, when the star expanded to become a red giant before contracting to a white dwarf.
This is the first planet known to orbit a white dwarf and its existence suggests that at least some of the Sun’s planets could survive when our star becomes a red giant in five billion years.
The white dwarf (called WD 1856+534) and the giant planet (WD 1856b) were studied using NASA’s Transiting Exoplanet Survey Satellite – which looks for fluctuations in starlight that occur when a planet passes in front of its star. TESS spotted the planet whizzing around the star once every 34 h in an extremely tight orbit that is about 20 times closer to its star than Mercury is to the Sun. Another extraordinary thing about this system is that the white dwarf is about the size of Earth, so the planet is much larger in size than the star it orbits.
No signs of destruction
“We were using the TESS satellite to search for transiting debris around white dwarfs, and to try to understand how the process of planetary destruction happens,” says Andrew Vanderburg at the University of Wisconsin-Madison, who is part of the team that made the discovery. “We were not necessarily expecting to find a planet that appeared to be intact.”
The team then studied the system in more detail using the Gemini Near-Infrared Spectrograph (GNIRS) on the Gemini North telescope in Hawaii. While the star could be seen by GNIRS, no light from an orbiting debris field, nor from the giant planet was detected.
Siyi Xu of the Gemini Observatory explains, “Because no debris from the planet was detected floating on the star’s surface or surrounding it in a disc we could infer that the planet is intact”. She adds, “because we didn’t detect any light from the planet itself, even in the infrared, it tells us that the planet is extremely cool, among the coolest we’ve ever found”. Indeed, NASA’s Spitzer Space Telescope was used to put an upper limit of 17 °C on the temperature of WD 1856b – which is similar to the average temperature of Earth. As a result, it is possible that life exists on the planet.
Hospitable planet
“I think the most exciting part of this work is what it means for both habitability in general – can there be hospitable regions in these dead solar systems – and also our ability to find evidence of that habitability,” says Vanderburg.
WD 1856+534 was once a star like the Sun before it ballooned out to become a red giant – which would have consumed WD 1856b had it been in its current orbit. Instead, astronomers believe that the planet was in a much larger orbit when the star was in its red-giant phase, so instead of being consumed, it was knocked into an eccentric orbit.
Eventually, the red giant burnt out, leaving the cool white dwarf behind. At that point, the planet could have wandered into its current tight orbit – possibly though gravitational interactions with other surviving planets. This journey would have taken billions of years, but astronomers believe that WD 1856+534 is almost 6 billion years old. Because the system is relatively close to Earth, it’s possible that astronomers could spot the other planets in the future.
An innovative fin-shaped design for light-emitting diodes (LEDs) could not only overcome the devices’ limited brightness, it could also help turn them into lasers. The new scheme, from researchers in the US, could prove valuable in applications including chemical sensing, hand-held communication technologies, high-definition displays and disinfection.
Wide bandgap semiconductor LED technology has developed substantially over the last few decades and is now in common use – for general lighting and displays, but also for a range of applications in areas such as photodetection and optoelectronics. The technology does, however, suffer from a major drawback: as an LED’s current density increases, its internal quantum efficiency (IQE) declines.
This decline in efficiency means that although an LED shines more brightly when supplied with stronger electrical currents, it does so only up to a certain point. After that, its brightness begins to drop off – a phenomenon known as efficiency droop. The extra heat generated in the LED as the applied current increases exacerbates the problem, with IQEs dropping by about 30% as the temperature increases from 23° to 177°C. As a result of these effects, the power of LEDs with an area less than a square micron tops out in the nanowatt (nW) range.
Two mechanisms are believed to contribute to efficiency droop. In the first, known as non-radiative recombination, excited charge carriers (electrons and holes) recombine without emitting light. This unwanted process lowers the efficiency of light generation and increases heat losses, since the electrons and holes recombine by producing phonons – thermal vibrations of the crystal lattice – instead of photons. In the second mechanism, known as Auger recombination, the energy of the electron-hole pair is transferred to another electron or hole, again without a photon being emitted. This charge carrier then normally loses its excess energy to thermal vibrations.
NIST team member Babak Nikoobakht, who conceived the new design, explains that he and his colleagues used the same materials as in conventional LEDs, but formed into a different shape. Unlike the flat, planar design commonly used, Nikoobakht and colleagues built their light source out of long, thin zinc oxide (ZnO) strands, or fins, measuring around 5 microns in length and approximately 160 nanometres in width. The resulting fin LED array, grown on a gallium nitride (GaN) substrate, looks like a tiny comb that extends over areas as large as a centimetre or more.
“We saw an opportunity in fins, as I thought their elongated shape and large side facets might be able to receive more electrical current,” Nikoobakht says. “At first we just wanted to measure how much the new design could take. We started increasing the current and figured we’d drive it until it burned out, but it just kept getting brighter.”
The group’s new ZnO-GaN fin LED emits light at violet to ultraviolet wavelengths and generates about 100 to 1000 times as much power as a typical micron-sized LED – up to 20 mW. The new design does not exhibit efficiency droop, even at record-high current densities of 1000 kA/cm2. Measurements of the device’s total spectral radiant flux also show that its output power increases linearly with drive current.
“One of the most efficient solutions”
Grigory Simin, a professor of electrical engineering at the University of South Carolina, US, who was not involved in the project, says that the new design is one of the most efficient solutions he has seen. “The community has been working for years to improve LED efficiency, and other approaches often have technical issues when applied to submicrometre wavelength LEDs,” he comments in a NIST press release. “This approach does the job well.”
Nikoobakht and colleagues also discovered that as they increased the current supplied to the LED to 1000 kA/cm2, the LED’s comparatively broad-band emission light (with a wavelength in the ultraviolet range, around 385 nm) narrowed to just two wavelengths (403 and 417 nm) of an intense violet colour. At this point, the device started to lase with a brightness of over 20 mW.
According to the researchers, their nanoLED’s enhanced light-emitting performance comes from the fin shape mitigating nonradiative pathways. The large side facets of the fins also allow for effective electrical injection and form a laser cavity.
Nikoobakht notes that converting an LED into a laser usually requires a lot of effort, and typically involves coupling the device to a resonance cavity that allows the light to bounce around and increase in intensity. In this case, however, “it appears that the fin design can do the whole job on its own, without needing to add another cavity,” he says.
While the nanoLEDs and nanolasers described in this work operate in the near UV range, the researchers say their concept could be applied to different materials systems, such as aluminium gallium nitride (AlGaN), boron nitride (BN) or their heterostructures, to develop far brighter deep-UV devices. They report their work in Science Advances.
Tirthraj Adhikari (far right) and colleagues commission a cobalt-60 radiotherapy system during the pandemic. (Courtesy: Tirthraj Adhikari)
I started work at the BP Koirala Memorial Cancer Hospital in Nepal at the beginning of this year as a fresh radiation oncology physicist. I had left Italy after graduating from the ICTP, just a month before the start of the COVID-19 pandemic. I was the first clinical physicist to be given an appointment from the Ministry of Health & Population to work as a clinical medical physicist in Nepal.
17 January was my first day in the radiation oncology department, which has seven physicians and four physicists. The hospital has three linacs, a simulator and a brachytherapy system for intracavitary treatment. It is the one of the busiest cancer centres in Nepal, with cancer patients visiting from all over the country. I interacted with patients from remote parts of Nepal, some of whom have to travel 16–20 hours by bus, after having trekked for one or two days. I was eager to chat with them, as I myself come from a remote village in far-west Nepal. For patients from my region, it takes on average 18 hours by bus to reach the cancer hospital.
During treatments, I noticed that – due to a lack of screening and diagnosis – patients would visit the hospital in the late stages of their disease. The major factor is poverty. When someone in a remote village feels sick, they usually visit the pharmacy and buy medicine without prescriptions from physicians. This makes them feel well for some time, but after a while the problems return. After several such attempts, the patient becomes serious and decides to visit the hospital. Unfortunately, by this time the cancer has reached a late stage and treatment becomes difficult.
The patient may have to undergo major surgery, but they cannot afford the cost. If they manage to afford surgery, they cannot afford radiotherapy, which is expensive. It is sad that for many patients who need treatment with a complex radiotherapy technique, we have to shift to a simple technique just because of the cost. For example, one patient with head-and-neck cancer was a candidate for volumetric-modulated arc therapy (VMAT), but we treated him using cobalt-60 radiotherapy because it is relatively inexpensive.
During this period, I created treatment plans for about 230 patients. Most of these were cobalt-60, 2D and 3D conformal radiotherapy (CRT) plans for head-and-neck, cervix and breast cancers. In addition, I created a few intensity-modulated radiotherapy (IMRT) and VMAT treatment plans for prostate and oesophagus cancers, and brachytherapy plans for cervical cancer. I remember that I created more plans for palliative than curative cases.
In addition to the treatment planning, I was involved in quality assurance and quality control in the department. Every day, I’d commute to the hospital early in the morning, do the daily quality assurance of the linacs, and make the CT simulator and brachytherapy machine ready for treatment and simulation. At our department, frequent problems with the machines arise during treatment delivery. I was engaged in troubleshooting of the linacs and CT simulator with senior physicists in the department.
The impact of COVID-19
The first COVID-19 patient in Nepal was identified in early February. He had flown from Wuhan in China to Nepal, and he soon recovered. Then the government declared us a corona-free state, while cases were growing in China, Italy and then the US. However, in March a girl who fled from Qatar to Nepal was diagnosed with COVID-19, and cases grew over time. The government then imposed a lockdown to prevent transmission to the community. This harshly affected all patients, including cancer patients.
As we did not have personal protective equipment (PPE) in our department, the hospital stopped treatments for three days the first time. Treatments soon resumed, but some patients had already left the hospital, as there was uncertainty when treatments would restart and difficulties with accommodation. Those patients who were living in the hospital wards and in the hospital periphery received the remaining fractions of their radiation treatments. But some patients had already travelled more than 600 km to their home.
Under normal conditions, the department used to treat 180–200 patients per day. But with lockdown, the number of patients declined rapidly, to 30–40 patients per day. This happened first because it was difficult for patients to travel the hospital – the lockdown meant there was no public transport and travel by ambulances was not affordable. Second, hospital personnel were afraid to touch or take care of patients without having PPE. During the first period of prolonged lockdown, around 30 patients who were having radiotherapy died at their homes.
After four months of lockdown, the hospital resumed normal activity and our department started to treat normally again. Patients who survived the lockdown came back to the hospital for their remaining treatments. For us, it was difficult to decide whether to treat patients with plans created four months earlier. Some patients had re-simulation after changes in their diagnosis and plans were recreated for them. For those who had already received some radiation fractions, the gap was calculated and dose was managed accordingly.
Following a spike in COVID-19-infected patients, the government imposed a second lockdown from the middle of August. At the time, some hospital personnel were also infected, though fortunately none of our department. Treatment was stopped, again for some days, with the same problems – it was a high risk for us to treat patients without proper PPE and testing. We demanded mandatory PCR tests for patients and their visitors to keep ourselves safe. Then the treatments resumed, but with fewer patients than under normal conditions.
A patient from Bardiya, a district of West Nepal, had started treatment in February. His head-and-neck cancer was being treated by 3D CRT. With three fractions remaining, his treatment was impeded because of the pandemic, and he was not able to resume treatment due to the ongoing pandemic. He contacted me recently saying that he had difficulty in breathing. I suggested that he visit the department soon for follow-up. But in the meantime, the government had imposed the second lockdown. I am wondering how he will manage to travel 600 km. The cancer patients in Nepal have been affected miserably by this COVID-19 pandemic.
Time management
This pandemic has adversely affected everyone’s lives. To keep myself physically active, I do regular yoga in the morning, wearing a mask, and I walk in the evening, maintaining physical distance.
In addition to the usual departmental clinical tasks, I used to talk with seniors, physicians and technicians about treatments, innovations, politics and science-economics. During lockdown, I followed more than 80 national and international webinars. In addition, I composed articles in my native language about nuclear laws in Nepal, radiation protection, cancer radiotherapy, managing errors in radiation therapy, the importance of radiation dosimetry in cancer care, and the increasing needs of the public cancer hospital in Nepal. I also prepared a proposal to establish a cancer hospital in my region, where there is no cancer centre and people have to travel at least 16 hours to the nearest cancer hospital.
To engage myself, I have taken 25 online/remote training courses during this pandemic, covering themes including radiation protection, nuclear safety, radiotherapy safety, image-guided radiation therapy and stereotactic radiosurgery. More importantly, to stay updated, keep myself safe from COVID-19 infection and answer people’s questions, I took 10 online training courses from the World Health Organization. With this training, I gained confidence in how to deal with COVID-19.
Finally, I attended the 2020 AAPM|COMP Virtual Meeting. It was possible to attend this conference as a friend of mine from the US paid the registration fee for me. Because of the time difference between Nepal and the US, I stayed up the whole night to follow the presentations.
Being a fresher, I have never felt nervous in dealing with radiation treatment during the pandemic, despite the risks. We continue to treat cancer patients with limited protective resources from COVID-19. But because of pandemic, I have had plenty of time to learn and implement ideas and techniques that I learnt while abroad. I personally took this pandemic as an opportunity – keeping myself busy by learning and teaching others remotely. Nobody knows what will happen tomorrow, so keeping yourself ready to face any unpredicted situation is wise.
“Marangoni surfers” can manipulate surface tension gradients at fluid interfaces to propel themselves to ultrafast speeds, researchers in Switzerland, Germany and the UK have found. The team, led by Kilian Dietrich at ETH Zurich, demonstrated the effect by illuminating Janus spheres with a special laser, enabling them to reach speeds of up to 10,000 times their length per second.
Active particles are unique in their ability to convert energy from their surrounding environments into straight-line motions – a behaviour made possible through the in-built asymmetry in their shapes or compositions. Owing to their simplicity, particles called Janus spheres are a popular choice for these experiments.
These are silica microbeads with one side left bare, and the other coated in gold. One mechanism that can propel the particles is asymmetric heating of the two surfaces by a light source, with the gold side absorbing more heat than the silica. The resulting temperature gradient surrounding the particle propels the particle in a straight path.
On its own, this mechanism is highly inefficient – only propelling particles to speeds of a few times their body length per second. In their study, Dietrich’s team sought to boost this efficiency by coupling the temperature gradient to surface-tension gradients in the surrounding fluid. They achieved this by positioning Janus spheres at a flat oil-water interface, then illuminating them with a powerful, specially-shaped laser beam.
Surface tension gradients
The resulting temperature asymmetry surrounding the particle lowered the surface tension at the oil-water interface on the warmer gold side, inducing the Marangoni effect – whereby flows arise along surface tension gradients at fluid interfaces. The Janus spheres could then “surf” along the resulting flows, in the direction of the surface tension gradients they created themselves.
In their wake, Dietrich and colleagues found that the Marangoni-surfing spheres left areas in which concentrations of surface tension-reducing surfactant molecules were depleted. This created a second surface tension gradient which acted in the opposite direction to the first – imposing a restriction on their motion. Therefore, the overall velocities of the spheres arose from a balance between these two components.
Dietrich’s team could easily control the velocities of their Marangoni surfers by varying either the power of the laser, or the concentration of surfactants at the interface. Their demonstrated velocities spanned four orders of magnitude: from microns, to several centimetres – around 10,000 times a sphere’s body length – per second. This represented a vast increase on the speeds of particles which propel themselves along temperature gradients alone.
The team’s discoveries offer new insights into systems whose dynamics are far from thermal equilibrium. Through further research, they could lead to new ways to manipulate active matter – in which large ensembles of suspended particles borrow energy from their surrounding environments to propel themselves and exert mechanical forces.
I’m not a machine (as far as I’m aware) but one day AI bots will probably replace at least some aspects of my job as a science journalist. Still, could these computer systems ever really be capable of doing “proper” science journalism? And what is science journalism anyway?
These were some of the questions at the heart of a session on AI journalism at the European Conference of Science Journalism (ECSJ), which took place on 1 September in Trieste, Italy, with most participants joining remotely.
“If we deploy this technology in a way that’s helpful for us, I think it can help science journalism, enrich it and make our jobs easier,” said Mićo Tatalović, news editor of Research Fortnight. Tatalović pointed to several examples where AI has already entered the newsroom and other forms of science communication.
They include (e) Science News, a website that employs an AI editor to curate its news across a range of disciplines. Elsewhere, there’s Science Surveyor, which can provide background information to a field, and SciNote, which can help you write your next scientific paper.
Tatalović spent a year working with computer scientists at Massachusetts Institute of Technology developing an AI bot to automate science news-writing. The group trained a neural network to write jargon-free text by feeding it with news articles from Science Daily along with the corresponding research papers behind the stories. Tatalović believes that the system would become increasingly skilled with additional data.
So could such systems eventually remove the need for science reporters like me?
Not likely, said Harry Collins, a sociologist from Cardiff University, another speaker on the ECSF panel. “There is a carapace of public descriptions of what AI can do, which doesn’t necessarily reflect what AI really can actually do.”
A case in point was the Guardian’s op-ed last week with the somewhat misleading headline “A robot wrote this entire article. Are you scared yet, human?“, which was produced by the GPT-3 language generator. A note following the reasonably well-written piece revealed that editors had cherrypicked and edited the best bits from eight different essays – giving readers an exaggerated view of GPT-3’s capabilities.
Interested in the sociological dimensions of big-science projects, Collins spent over four decades following the quest to directly detect gravitational waves. He believes the real key to that historic first detection in 2015 was human interaction, underlining that the LIGO-Virgo collaboration had previously rejected several apparent detections. “The real art of science is working out what’s good data and what’s rubbish,” he said.
Collins argued that if AI is of limited use in breakthrough science, then by extension the same is true for science journalism. “Serious journalism is where people are trying to get themselves to the frontier of science and develop a real serious understanding of science,” he said.
You can see the full discussion in this video, also featuring Fabiana Zollo from Ca’ Foscari University of Venice and Charlie Beckett of the London School of Economics.
A comparison of three commercially available artificial intelligence (AI) systems for breast cancer detection has found that the best of them performs as well as a human radiologist. Researchers applied the algorithms to a database of mammograms captured during routine cancer screening of nearly 9000 women in Sweden. The results suggest that AI systems could relieve some of the burden that screening programmes impose on radiologists. They might also reduce the number of cancers that slip through such programmes undetected.
Population-wide screening campaigns can cut breast-cancer mortality drastically by catching tumours before they grow and spread. Many of these programmes employ a “double-reader” approach, in which each mammogram is assessed independently by two radiologists. This increases the procedure’s sensitivity – meaning that more breast abnormalities are caught – but it can strain clinical resources. AI-based systems might alleviate some of this strain – if their effectiveness can be proved.
Fredrik Strand. (Courtesy: Martin Stenmark)
“The motivation behind our study was curiosity about how good AI algorithms had become in relation to screening mammography,” says Fredrik Strand at Karolinska Institutet in Stockholm. “I work in the breast radiology department, and have heard many companies market their systems but it was not possible to understand exactly how good they were.”
The companies behind the algorithms that the team tested chose to keep their identities hidden. Each system is a variation on an artificial neural network, differing in details such as their architecture, the image pre-processing they apply and how they were trained.
The researchers fed the algorithms with unprocessed mammographic images from the Swedish Cohort of Screen-Age Women dataset. The sample included 739 women who had been diagnosed with breast cancer less than 12 months after screening, and 8066 women who had received no diagnosis of breast cancer within 24 months. Also included in the dataset, but not accessible to the algorithms, were the binary “normal/abnormal” decisions made by the first and second human readers for each image.
The three AI algorithms rate each mammogram on a scale of 0 to 1, where 1 corresponds to maximum confidence that an abnormality is present. To translate this approach into the binary system used by radiologists, Strand and colleagues chose a threshold for each AI algorithm so that the binary decisions assumed a specificity (the proportion of negative cases classified correctly) of 96.6%, equivalent to the average specificity of the first readers. This meant that only mammograms that scored above the threshold value for each algorithm were classed as abnormal cases. The ground truth to which they were compared comprised all cancers detected at screening or within 12 months thereafter.
Under this system, the researchers found that the three algorithms, AI-1, AI-2 and AI-3, achieved sensitivities of 81.9%, 67.0% and 67.4%, respectively. In comparison, the first and second readers averaged 77.4% and 80.1%. Some of the abnormal cases identified by the algorithms were in patients whose images the human readers had classified as normal, but who then received a cancer diagnosis clinically (outside of the screening programme) less than a year after the examination.
This suggests that AI algorithms could help correct false negatives, particularly when used within schemes based on single-reader screening. Strand and colleagues showed that this was the case by measuring the performance of combinations of human and AI readers: pairing AI-1 with an average human first reader, for example, increased the number of cancers detected during screening by 8%. However, this came with a 77% rise in the overall number of abnormal assessments (including both true and false positives). The researchers say that the decision to use a single human reader or high-performing AI algorithm, or a human–AI hybrid system, would therefore need to be made after a careful cost–benefit analysis.
As the field advances, we can expect the performance of AI algorithms to improve. “I have no idea how effective they might become, but I do know that there are several avenues for improvement,” says Strand. “One option is to analyse all four images from an examination as one entity, which would allow better correlation between the two views of each breast. Another is to compare to prior images in order to better identify what has changed, as cancer is something that should grow over the years.”
Full details of the research are published in JAMA Oncology.
A gamma ray flare originating from a distant blazar was likely generated by magnetic reconnection within a black hole’s relativistic jet, a pair of researchers in Germany have proposed. Amit Shukla at the Indian Institute of Technology Indore and Karl Mannheim at the University of Würzburg used observations from NASA’s Fermi-LAT space telescope to reveal how “mini-jets” form within the blazar’s larger plasma jets, producing high-energy gamma rays. Their conclusions provide new insights into how the magnetic fields surrounding supermassive black holes dissipate their vast amounts of energy.
Powerful magnetized jets are common features of the spinning supermassive black holes that occupy the centres of large galaxies. Within these features, plumes of accelerated matter can extend to hundreds of thousands of light-years along the black hole’s rotational axis; dissipating their energy by emitting radiation from across the entire electromagnetic spectrum. These emissions are thought to be boosted by shock waves travelling along the jets, accelerating particles to highly relativistic speeds. However, Shukla and Mannheim propose that these boosts would be too inefficient within a black hole’s magnetically dominated plasma to fully explain how the jets dissipate their energy.
The duo explores this idea in their study, through observations gathered by Fermi-LAT, which is a space-based gamma-ray detector. In 2018, Fermi-LAT observed a giant gamma-ray flare in the distant blazar 3C 279, which endured for almost six months. Yet within this time, the flare displayed a distinct flickering; sometimes doubling in brightness on timescales of just a few minutes. The observations provided Shukla and Mannheim with an ideal opportunity to examine how energy is dissipated within the innermost parts of black hole jets.
Magnetic topologies
Based on the timescales of the flickering they observed, the researchers concluded that the regions of gamma-ray emission within the burst were limited in size. This suggested that the accelerations responsible are driven by structures far smaller than jet-spanning shock waves. Instead, Shukla and Mannheim argue that they can be better explained by the process of magnetic reconnection – which describes how the topologies of magnetic fields within highly conductive plasmas can be rearranged. This process converts the magnetic energy of the plasma into kinetic and heat energy, driving particle accelerations.
In addition, Shukla and Mannheim found that gamma rays in the burst were not being attenuated by pair production – in which electron-positron pairs are created during collisions between gamma and ultraviolet photons. This would suggest that the responsible accelerations were taking place at light-year distances from the central black hole. This far away, kinks emerge within the jet’s thin, relativistic plasma columns, introducing turbulence. In these conditions, magnetic reconnection can readily occur.
The duo tested these ideas by incorporating them into a model black hole jet. They found that through turbulence-driven reconnection, the jet’s magnetic field fragments to form smaller clumps of plasma. These interact with each other and grow within the reconnection region; eventually forming mini-jets within the larger jet, which dissipate their energy through smaller-scale gamma bursts. If correct, this conclusion could suitably explain the characteristic flickering observed by Fermi-LAT, and may ultimately improve astronomers’ understanding of the complex, often mysterious physics of black hole jets.
On 23 March 2020 UK prime minister Boris Johnson announced a lockdown to tackle the spread of coronavirus, following the example of other countries around the world who chose this strategy to halt the virus’ progression. This decision came days after Johnson’s government toyed with the idea of letting the virus spread and infect up to 70% of the population, in order to develop so-called “herd immunity”. The stark policy shift left people wondering what had changed.
To many, the models produced by the physicist-turned-epidemiologist Neil Ferguson and his group at Imperial College London were critical. They predicted that should no action be taken, the death toll in the UK could reach 500,000, and may exceed 2 million in the US. As well as providing a shocking reality check about the pandemic, the work highlighted an increasingly popular new tool that is profoundly changing medical research.
While in vitro and in vivo experiments have long been a staple of medical-based research, the rapid increase of computational power in recent decades has enabled the emergence of a new experimental field: computational (in silico) modelling. From surgery to drug design, these numerical models are not only used to describe physiological phenomena, but also to derive useful information and even drive clinical decisions.
“Modelling is about encapsulating our knowledge into a set of rules or equations. It is thus at the core of any science,” says Pablo Lamata from King’s College London, UK. “We are currently experiencing a ‘computational boost’ in our modelling capabilities.”
Twin hearts Pablo Lamata and his group at King’s College London create digital twins of a patient’s heart to infer properties that are not readily available to doctors. They use a modelling blend to teach computers how to learn patterns from data (statistical modelling) and how to make predictions based on knowledge of the workings of the cardiovascular system (mechanistic modelling). (Courtesy: Pablo Lamata)
The ever-increasing amount of available data – from wearable sensors to digital medical images – has also sped up the applications of modelling. Lamata’s group, for example, combines in silico heart models with medical images of the heart to create patient-specific numerical heart models – so-called digital twins. Such models could in future provide doctors with vital information regarding cardiac properties that are currently unavailable, such as heart stiffness. This is important because when the heart fills up with blood (during diastole), stiffness can prevent the ventricle from filling up properly, a phenomenon associated with heart failure in about 50% of patients. These models could also provide new understandings on the mechanisms leading to this outcome.
“We obviously cannot touch a beating heart to know the stiffness, but we can use these models governed by the rules and laws of the material properties to infer that important piece of diagnostic and prognostic information,” Lamata explains. “The stiffness of the heart becomes another key biomarker that will tell us how the health of the heart is coping with disease.”
Reducing uncertainty
Similar approaches are used in other medical fields. Paul Sweeney, a mathematician from the University of Cambridge in the UK, for example, uses in silico models to predict perfusion – the passage of fluids through the circulatory or lymphatic systems to an organ or tissue – and drug delivery across whole tumours at the scale of the smallest blood vessels (approximately 3 μm). “Our models allow us to understand how a tumour’s micro-architecture influences the distribution of fluid and mass through cancerous tissue, which is important for engineering new anti-cancer drugs, or optimizing strategies of current therapeutics,” Sweeney says. As with heart stiffness, such data are otherwise unavailable through conventional experiments in isolation, which means that introducing modelling can inform the development of cancer therapeutics.
But getting to the point where information can be derived from these models is only the last stage in a long and thorough development process. From defining the problem to selecting the modelling strategy to address it, each step requires crucial choices, which are reassessed later to ensure they reflect reality. “It is only through many iterations of searching for the agreement between model and data that the weak links are revealed and addressed,” notes Lamata. This is why, just as with weather forecasting or death-toll projections in the COVID-19 crisis, models are constantly updated as more data become available.
Sometimes this means revisiting the assumptions that models are based on: they are necessary at the beginning to facilitate the modeller’s task, but their impact on predictions should be carefully handled. “The best solution to deal with this is the use of sensitivity analysis [assessing a parameter’s influence on the model prediction by varying it while keeping all other parameters constant] – but the limit is always going to be in those aspects that your model can’t account for,” Lamata says. It is important to keep this in mind when considering what can be inferred from the model output and what cannot, while also leaving room for improvement.
Modelling blood The tumour models that Paul Sweeney builds at the University of Cambridge use well-established mathematical models, such as those for microcirculatory blood flow (image at top of article), in combination with new approaches for predicting fluid distribution through the interstitial tissue (above). Both models are parameterized and validated against biomedical imaging data. (Courtesy: Paul Sweeney)
To Sweeney, there is always scope for increasing the accuracy of models, whether by supplying additional experimental data or incorporating more models of other complex biological mechanisms. “A potential trade-off is always between the accuracy of the model and the amount of experimental data available,” he says. “In other words, will additional data make the model more accurate or cause overfitting to the empirical data?”
Indeed, this is the caveat that all models face: the need to balance accuracy against simplicity. If a model relies too heavily on the data that it was developed from, its prediction for other datasets might not be accurate – the problem of overfitting Sweeney refers to. Conversely, developing as basic as possible a model, with little reliance on data or patient characteristics, can also yield unreliable predictions, as the “one-size-fits-all” approach fails to account for a patient’s individualities.
The purpose of the model dictates where to put the emphasis, although simplicity is usually favoured. The focus hence shifts back to the importance of accurately defining the problem that the model is addressing in the first place. “You want the minimum model to correctly address the problem,” concludes Lamata.
Designing new technologies
Confronting a model’s predictions with reality remains the quickest method of validation, although this can be achieved faster in some fields than others. Take biomaterials, where in silico models help us understand how molecules behave and interact with their environment, which can quickly be replicated in vitro.
Over at the University of Nottingham in the UK, for example, bioengineer Alvaro Mata and his group use modelling as a stepping stone for developing innovative materials and therapies for tissue engineering and regenerative medicine. “Molecular dynamic simulations are key to elucidate mechanisms by which molecules assemble together,” explains Mata. “This allows us to translate those assemblies at the molecular level into fabrication platforms capable of engineering functional structures.”
His group recently used this approach to understand how graphene oxide can exploit the flexible regions of a protein to create a new bioink material for 3D-printing tissue-like vasculature structures. Through the models, the researchers learnt how to guide their assembly at various size scales, from the cellular level to the final complex structure. “Simulations can dramatically facilitate testing and optimization of materials, structures and processes, saving both time and money,” Mata says.
Functional boost By using molecular dynamics simulations of the interaction between elastin proteins (gold/green flexible filaments) and graphene oxide (rigid gold with red and white balls), Alvaro Mata and colleagues at the University of Nottingham can understand and optimize the underlying mechanism of the assembly of tubular structures. These studies considerably boost the chances that the functional structure engineered from this interaction is viable. (Courtesy: Alvaro Mata)
Compared with in vitro and in vivo experiments, in silico simulations have the advantage of being fast, cheap, safe, easy to implement and free of experimental errors. Consequently, they are becoming increasingly helpful in designing new technologies and strategies.
A prime example of this is cardiac resynchronization therapy (CRT) in patients with heart failure. This treatment involves placing two pacing leads controlled by a pacemaker into the patient’s heart to augment the electrical activation and synchronize the beating of the two ventricles. Traditionally, the leads’ location and timings for stimulation are derived from electrocardiograms (ECGs) and medical images, but 30% of patients do not see a clear benefit from this strategy. By producing computational heart models from the patient’s scans and simulating various pacing strategies on them, Steven Niederer’s group at King’s College London can identify the best area to electrically stimulate the heart and investigate the effects of changing the pacing.
Such models are particularly complex, requiring multi-scale modelling to link cellular dynamics, blood flow, electrophysiology and tissue deformations within a common anatomical heart geometry. There is still a long way to go before the technology can be adopted as a support tool for clinical decision-making, but studies in a small number of patients have shown that such models can perform patient-specific predictions about the acute haemodynamic CRT response. This further demonstrates that clinical intervention guided by in silico models is no longer just a pipe dream.
In fact, in 2015 Alberto Figueroa and his team at the University of Michigan in the US helped perform the first surgical intervention that used computerized blood flow simulation. The team’s open-source software, CRIMSON, uses MRI scans and haemodynamic variables (blood pressure and flow) to produce 3D computer models of a patient’s circulatory system. This can be used to simulate different surgical alternatives and determine which yields the best prognosis before heading to the operating theatre.
The CRIMSON software has already been used to plan several complex cardiac operations, such as “Fontan procedures”, which involves rewiring pulmonary circulation in patients born with just one functioning ventricle. The venous return – the flow of blood back into the heart – is rerouted to bypass the heart and connected directly to the pulmonary arteries for transport to the lungs. Simulations help surgeons decide where to make the surgical connections so that blood flow is ideally balanced between the lungs.
In silico trials in reality
With the surge in applications of computational modelling and the demonstration of its clinical relevance, regulatory bodies are taking note and have already acknowledged the benefit of these models.
For example, in 2011 the US Food and Drug Administration (FDA) approved the first in silico diabetes type 1 model as a possible substitute for pre-clinical animal testing for new control strategies for type 1 diabetes. A few years later, the FDA went further by approving FFRCT software that had been developed by the US medical firm Heartflow to measure coronary blockages non-invasively from CT scans. This was therefore the first clinical technology based on subject-specific modelling to get the green light. The software has also received CE marking and regulatory approval in Japan.
In specific cases, such as the assessment of drug toxicity on the heart, the FDA is now even sparking new collaborations between academia and industry to rise to the challenge.
One success story comes from the University of Oxford in the UK, where Blanca Rodriguez’s group developed Virtual Assay – software that can run in silico drug trials in populations of human cardiac cell models. Designed to predict drug safety and efficacy, the software simulates the effects of drugs on the electrophysiology and calcium dynamics of human cardiomyocytes – cells that control the heart’s ability to contract. Specific heart rhythm patterns can therefore be inferred for each drug compound and dose simulated, with the objective of spotting any drug-induced arrythmias. An in silico trial of 62 drugs, led by Rodriguez’s team in collaboration with Janssen Pharmaceutica, showed that the software was more accurate at predicting abnormal heart rhythms than animal trials (89% accuracy versus 75% in rabbits).
These results captured the attention of other pharmaceutical companies – a sector in which developing new compounds costs several billions of dollars, and 20–50% of drug candidates are abandoned due to cardiovascular safety issues. Eight major pharmaceutical companies are now evaluating Virtual Assay in the early drug development process to assess arrhythmic risk. This strategy is also backed by animal protection groups, as in silico models for pharmaceutical R&D could reduce the need for animal use by a third.
This conjunction of interest from regulatory bodies, industry, clinics, academia and even animal-welfare groups has led to the establishment of networks and initiatives around the world to promote the development, validation and use of in silico medicine technologies.
The Virtual Physiological Human Institute led the way when it was opened to members in 2011. As part of a two year EU-funded project starting in 2013, it produced a roadmap for the introduction of in silico clinical trials. Lamata, Rodriguez and Niederer are part of a network of universities, industries and regulatory bodies working to bring personalized in silico cardiology to the clinics. Just as research is becoming more and more interdisciplinary, this combination of expertise is vital to steer the use of models in the right direction and facilitate their clinical translation.
“Oddly, one challenge with our models is knowing where to begin analysing their predictions as they produce a vast wealth of data. The interdisciplinary expertise on our team makes this process far easier by providing unique perspectives on how to tackle this challenge,” says Sweeney.
As Lamata puts it, it’s easy to be tempted to “think big” and include a lot of variables and complexity to have a huge model prediction power. “But it’s important to keep things as simple as possible for validation, which is an incremental process. Working with clinicians helps to keep the end-goal in sight.”
If incorporating in silico models to complement traditional in vitro and in vivo experiments is a new paradigm, all stakeholders seem to adapt well. Sweeney did not perceive much resistance from biologists and clinicians, who are more used to working with statistics and correlations drawn from large cohorts than equation-based, patient-specific simulations. For Lamata, the challenge with demonstrating the accuracy of in silico-based predictions remains the same as for any scientific advance: the need for evidence. And this takes time to generate. “The main cultural shift we require is one towards open science, where we make our data and tools available for the fast generation of the required evidence,” he notes.
The momentum is there: industrialists, policy-makers and clinicians are on board; personalized medicine is gaining traction as computational power keeps increasing and initiatives flourish; evidence of computational models’ capacity to enhance diagnosis, prognosis and treatment is mounting. The COVID-19 pandemic has shown that models can even influence government decisions. It might just be a matter of time before in silico models in medicine become as ubiquitous as the computers they are run on.
Ultimately, if all models are limited by their hypotheses, the possibilities that they offer are limitless. You just need to know exactly what you are looking for.
When I was in high school I loved physics and hoped to study theoretical physics in college. However, I didn’t do very well in the national college entrance examination and ended up choosing cell biology. I became very interested in the physical aspects of living cells and did a Master’s in biophysics at what is now Peking University Health Science Center. I learned to use various biophysics instruments and techniques to study cytoskeletons and cell membranes. I also started to realize that biophysical signals could be as important as biochemical signals in understanding cells, tissues and organs.
How did physics help your PhD in cell biology?
I did my PhD with Michael Sheetz at Duke University Medical Center in the US. This involved using laser optical tweezers to precisely measure the tension of cell membranes. We did this by putting a latex bead on the cell’s surface and then removed the bead to pull out a membrane tube – which is very elastic – from the cell. We then measured the force needed to pull the tube and calculated the cell membrane tension. It was the first time that laser tweezers were used to study living cells and the work involved a lot of physics.
On the one hand, the potential market for academic publishing in China is huge. On the other, there still needs to be an improvement in the quantity and quality of our publications.
What do you currently work on?
My current research focuses on tissue and organ regeneration, especially through the design of collagen scaffolds. Our lab recently completed a clinical trial with about 400 patients for a “smart bone formation” product and now we’re waiting for official clearance from the National Medical Products Administration. We also did a clinical study on intrauterine tissue regeneration with about 100 patients who had Asherman’s syndrome – a rare condition in which scar tissue forms in the uterus. More than 60 healthy babies have been born as a result of that project.
What are some of the key research topics carried out at the Institute of Genetics and Developmental Biology?
A significant part of the institute’s research efforts are dedicated to agriculture-based molecular breeding. My colleagues use genome sequencing and other modern biotechnologies to cultivate products such as rice, wheat and corn with high-quality yields and better resistance to disease. Our institute has also partnered with local governments to set up several breeding centers across the country.
What are some of the main issues facing researchers at your institute?
One issue is that about 10% of our research groups face funding shortage, in particular those that are not directly engaged in the institute’s major research projects. Another is that my colleagues and I don’t have enough graduate students or postdocs to work with. The quota cap on student admission in China has favoured universities over research institutes. For instance, I can only take on one doctoral student each year. This has become a bottleneck problem for the institute. The mechanism for industrial translation also needs to be improved.
How has COVID-19 affected your lab and are you now starting to reopen?
There certainly have been interruptions. My lab was closed for a couple of months during the worst of the pandemic. Some students are now back in the lab, especially those who are about to graduate. We’ve recovered about 60–70% of our research productivity. But we still can’t get all the lab supplies we need because the suppliers are manufacturing at a reduced capacity. I believe the impacts of COVID-19 in our field are temporary rather than long term. Students might face delays, but they shouldn’t have problems finding a job or further training opportunities as they move ahead.
You recently became the editor-in-chief of Biomedical Materials (BMM). Why did you decide to take on this role?
I’ve worked on biomedical materials for two decades and I’m aware of the importance of high-quality publications for our field. Early on, I was thinking of starting my own journal, but that could be a time-consuming process. So, when IOP Publishing, which publishes Physics World, contacted me about the BMM appointment, I immediately realized it’s a great opportunity and accepted.
What are the strengths and challenges of the journal?
This year is the 15th anniversary of BMM and the editorial board has worked hard to process manuscripts in an efficient manner. However, the journal faces increasing challenges from both the rapidly evolving discipline itself and from rival journals. We plan to distinguish BMM from similar publications by highlighting special topics such as the study of physical and mechanical mechanisms of biomedical materials in tissue regeneration. We will also expand the journal’s scope to include the latest advances in industry.
How are you going to make it more appealing to researchers in the field, especially those from China?
China has an expanding biomedical materials research community. To attract more authors and readers from China, we’re working on assembling a team to promote BMM through Chinese social-media performs and via online talks.
How do you see academic publishing in China?
On the one hand, the potential market for academic publishing in China is huge. On the other, there still needs to be an improvement in the quantity and quality of our publications. While research papers are essential, review articles, outreach and educational content are equally important in the world of academic publishing. For example, reviews play a major role in interpreting and synthesizing new research findings for a broader audience and they are particularly helpful for graduate students and postdocs who are interested in the overall development of their field.
Phosphine, which is a gas produced exclusively by microbes on Earth and considered to be a strong signature of life on other worlds, has been detected in the clouds of Venus. The discovery is perhaps the strongest evidence yet of life beyond Earth.
The idea to search for phosphine as a biosignature on other worlds is a recent one, developed in 2019 by astronomers led by Clara Sousa-Silva at the Massachusetts Institute of Technology, and independently by Greaves. Phosphine is a molecule derived from phosphorus and is an essential building block of RNA and DNA. On Earth it is produced by anaerobic bacteria, which are microbes that do not require oxygen. They absorb phosphate minerals and combine them with hydrogen, releasing phosphine in the process. Importantly, phosphine is not produced by any known geological process, at least not on Earth.
“In terms of the most distinctive biomarkers [in the solar system] where we can’t find a geological explanation, this is very strong,” Greaves tells Physics World.
A very different planet
Sousa-Silva’s work, which addressed how astronomers might detect phosphine in the atmospheres of Earth-like exoplanets, concluded that the presence of phosphine would act as an iron-clad biomarker since no other process on Earth-like planets is known to produce it. Venus, however, is a different type of planet to Earth. Its surface swelters at an average temperature of 460 °C, and is crushed under an atmospheric pressure of 93 bar – compared with 1 bar on Earth. The planet’s dense atmosphere is almost entirely made of carbon dioxide, laced with clouds of sulphuric acid. It is possible that some unknown chemical reaction in these extreme conditions could be producing the phosphine, but one problem is the lack of hydrogen.
Phosphine is formed from a phosphorus atom bonded to three hydrogen atoms. In the outer solar system, Jupiter and Saturn are able to produce phosphine via a non-biological process. These planets are hydrogen rich, however, and with so much hydrogen available in the high temperatures and pressures deep within their interiors, it is a relatively straightforward process for them to produce phosphine that is then dredged into their upper atmosphere by convection currents.
Venus, on the other hand, has very little hydrogen, having lost it to space long ago, along with most of the planet’s water. Instead, Venus is carbon rich. Without free hydrogen, it is difficult to conceive of a non-biological process to create phosphine. Furthermore, even if some geological reaction were taking place, adding all the possible sources such as volcanoes and the existence of favourable minerals, it would still come in at 10,000 times short of the observed phosphine abundance, which is 20 parts-per-billion.
No known geological process
“That doesn’t mean that the biological origin is the correct idea,” says Greaves. “It just means that we can’t find a really viable geological process.”
There is little previous evidence for phosphorous, too, on Venus, the only other detection having been made by the Soviet Union’s Vega 2 lander in 1985.
“This new detection of phosphine is significant because it suggests the widespread presence of phosphorous in Venus’ clouds,” says Sanjay Limaye, who is an atmospheric physicist from the University of Wisconsin, Madison, but who was not involved in the phosphine discovery. Limaye is a former chair of NASA’s Venus Exploration Analysis Group.
Scientists have speculated about the existence of microbial life in Venus’ clouds, attributing said life to unidentified ultraviolet-absorbing particles present within the planet’s atmosphere. These particles are currently being mapped by the Japanese Aerospace Exploration Agency’s Akatsuki orbiter.
Venusian habitable zone
Despite Venus’ generally hellish conditions, some regions of the planet are more clement than others. “The altitude that we probed is the top end of what is sometimes called the Venusian habitable zone,” says Greaves. This extends about 47–60 km above the surface, where temperatures range between 0 and 100 °C and atmospheric pressures average about 1 bar. However, the clouds also pose a hazard to life: it is not clear how microbes could survive in conditions that are 95% sulphuric acid.
More observations are required, says Limaye: “Confirmation of the presence of phosphine by other means is very much needed.”
It has been suggested that future missions to Venus could incorporate balloons or winged craft that could explore this potentially habitable region of the atmosphere. In the meantime, NASA is considering two future missions to Venus: VERITAS, which will study the planet from orbit, observing primarily with a synthetic aperture radar; and DAVINCI+, which will be a probe that will dive through Venus’ atmosphere.
“The radar should be able to provide some clues about the presence of liquid water on the surface in the past, while more crucially, the probe may be able to sample the cloud composition and search for the presence of phosphine,” says Limaye.
If the phosphine does prove to be biological in origin, it would mean that, surprisingly, Venus would be the first planet beyond Earth to be found to harbour life. Given its considerably harsh conditions, it would throw the idea of the habitable zone wide open.
