Can AI save us from the arduous and time-consuming task of academic research collection? An international team of researchers investigated the credibility and efficiency of generative AI as an information-gathering tool in the medical field.

The research team, led by Professor Masaru Enomoto of the Graduate School of Medicine at Osaka Metropolitan University, fed identical clinical questions and literature selection criteria to two generative AIs; ChatGPT and Elicit. The results showed that while ChatGPT suggested fictitious articles, Elicit was efficient, suggesting multiple references within a few minutes with the same level of accuracy as the researchers.

“This research was conceived out of our experience with managing vast amounts of medical literature over long periods of time. Access to information using generative AI is still in its infancy, so we need to exercise caution as the current information is not accurate or up-to-date.” Said Dr. Enomoto. “However, ChatGPT and other generative AIs are constantly evolving and are expected to revolutionize the field of medical research in the future.”

Their findings were published in Hepatology Communications.

But integrating 2D materials into devices and systems like computer chips is notoriously difficult. These ultrathin structures can be damaged by conventional fabrication techniques, which often rely on the use of chemicals, high temperatures, or destructive processes like etching.

To overcome this challenge, researchers from MIT and elsewhere have developed a new technique to integrate 2D materials into devices in a single step while keeping the surfaces of the materials and the resulting interfaces pristine and free from defects.

Their method relies on engineering surface forces available at the nanoscale to allow the 2D material to be physically stacked onto other prebuilt device layers. Because the 2D material remains undamaged, the researchers can take full advantage of its unique optical and electrical properties.

They used this approach to fabricate arrays of 2D transistors that achieved new functionalities compared to devices produced using conventional fabrication techniques. Their method, which is versatile enough to be used with many materials, could have diverse applications in high-performance computing, sensing, and flexible electronics.

Core to unlocking these new functionalities is the ability to form clean interfaces, held together by special forces that exist between all matter, called van der Waals forces.

However, such van der Waals integration of materials into fully functional devices is not always easy, says Farnaz Niroui, assistant professor of electrical engineering and computer science (EECS), a member of the Research Laboratory of Electronics (RLE), and senior author of a new paper describing the work.

“Van der Waals integration has a fundamental limit,” she explains. “Since these forces depend on the intrinsic properties of the materials, they cannot be readily tuned. As a result, there are some materials that cannot be directly integrated with each other using their van der Waals interactions alone. We have come up with a platform to address this limit to help make van der Waals integration more versatile, to promote the development of 2D-materials-based devices with new and improved functionalities.”

Niroui wrote the paper with lead author Peter Satterthwaite, an electrical engineering and computer science graduate student; Jing Kong, professor of EECS and a member of RLE; and others at MIT, Boston University, National Tsing Hua University in Taiwan, the National Science and Technology Council of Taiwan, and National Cheng Kung University in Taiwan. The research will be published in Nature Electronics.

Advantageous attraction

Making complex systems such as a computer chip with conventional fabrication techniques can get messy. Typically, a rigid material like silicon is chiseled down to the nanoscale, then interfaced with other components like metal electrodes and insulating layers to form an active device. Such processing can cause damage to the materials.

Recently, researchers have focused on building devices and systems from the bottom up, using 2D materials and a process that requires sequential physical stacking. In this approach, rather than using chemical glues or high temperatures to bond a fragile 2D material to a conventional surface like silicon, researchers leverage van der Waals forces to physically integrate a layer of 2D material onto a device.

Van der Waals forces are natural forces of attraction that exist between all matter. For example, a gecko’s feet can stick to the wall temporarily due to van der Waals forces. Though all materials exhibit a van der Waals interaction, depending on the material, the forces are not always strong enough to hold them together. For instance, a popular semiconducting 2D material known as molybdenum disulfide will stick to gold, a metal, but won’t directly transfer to insulators like silicon dioxide by just coming into physical contact with that surface.

However, heterostructures made by integrating semiconductor and insulating layers are key building blocks of an electronic device. Previously, this integration has been enabled by bonding the 2D material to an intermediate layer like gold, then using this intermediate layer to transfer the 2D material onto the insulator, before removing the intermediate layer using chemicals or high temperatures.

Instead of using this sacrificial layer, the MIT researchers embed the low-adhesion insulator in a high-adhesion matrix. This adhesive matrix is what makes the 2D material stick to the embedded low-adhesion surface, providing the forces needed to create a van der Waals interface between the 2D material and the insulator.

Making the matrix

To make electronic devices, they form a hybrid surface of metals and insulators on a carrier substrate. This surface is then peeled off and flipped over to reveal a completely smooth top surface that contains the building blocks of the desired device.

This smoothness is important, since gaps between the surface and 2D material can hamper van der Waals interactions. Then, the researchers prepare the 2D material separately, in a completely clean environment, and bring it into direct contact with the prepared device stack.

“Once the hybrid surface is brought into contact with the 2D layer, without needing any high-temperatures, solvents, or sacrificial layers, it can pick up the 2D layer and integrate it with the surface. This way, we are allowing a van der Waals integration that would be traditionally forbidden, but now is possible and allows formation of fully functioning devices in a single step,” Satterthwaite explains.

This single-step process keeps the 2D material interface completely clean, which enables the material to reach its fundamental limits of performance without being held back by defects or contamination.

And because the surfaces also remain pristine, researchers can engineer the surface of the 2D material to form features or connections to other components. For example, they used this technique to create p-type transistors, which are generally challenging to make with 2D materials. Their transistors have improved on previous studies, and can provide a platform toward studying and achieving the performance needed for practical electronics.

Their approach can be done at scale to make larger arrays of devices. The adhesive matrix technique can also be used with a range of materials, and even with other forces to enhance the versatility of this platform. For instance, the researchers integrated graphene onto a device, forming the desired van der Waals interfaces using a matrix made with a polymer. In this case, adhesion relies on chemical interactions rather than van der Waals forces alone.

In the future, the researchers want to build on this platform to enable integration of a diverse library of 2D materials to study their intrinsic properties without the influence of processing damage, and develop new device platforms that leverage these superior functionalities.

This research is funded, in part, by the U.S. National Science Foundation, the U.S. Department of Energy, the BUnano Cross-Disciplinary Fellowship at Boston University, and the U.S. Army Research Office. The fabrication and characterization procedures were carried out, largely, in the MIT.nano shared facilities.

Hubs are critical for connectivity

Hubs are everywhere. The hub of a bicycle wheel, with spokes shooting out from the center, keeps the wheel from collapsing when the bicycle is ridden. Airport hubs connect cities across the world. And social hubs like coffee shops or online social networks are places people gather for interaction.

The human brain has hubs, too — the intersection of many neuronal pathways that help coordinate brain activity required for complex functions like understanding and responding to speech. However, whether highly interconnected brain hubs are irreplaceable for certain brain functions has been controversial. By some accounts the brain, as an already highly interconnected neural network, can in principle immediately compensate for the loss of a hub, in the same way that traffic can be redirected around a blocked off city center.

With a rare experimental opportunity, the UI neurosurgical and research teams led by Matthew Howard III, MD, professor and DEO of neurosurgery, and Christopher Petkov, PhD, professor and vice chair for research in neurosurgery, have achieved a breakthrough in understanding the necessity of a single hub. By obtaining evidence for what happens when a hub required for language meaning is lost, the researchers showed both the intrinsic importance of the hub as well as the remarkable and rapid ability of the brain to adapt and at least partially attempt to immediately compensate for its loss. The findings were reported recently in the journal Nature Communications.

Evaluating the impact of losing a brain hub

The study was conducted during surgical treatment of two patients with epilepsy. Both patients were undergoing procedures that required surgical removal of the anterior temporal lobe — a brain hub for language meaning — to allow the neurosurgeons access to a deeper brain area causing the patients’ debilitating epileptic seizures. Before this type of surgery, neurosurgery teams often ask the patients to conduct speech and language tasks in the operating room as the team uses implanted electrodes to record activity from parts of the brain close to and distant from the planned surgery area. These recordings help the clinical team effectively treat the seizures while limiting the impact of the surgery on the patient’s speech and language abilities.

Typically, the recording electrodes are not needed after the surgical resection procedure and are removed. The innovation in this study was that the neurosurgery team was able to safely complete the procedure with the recording electrodes left in place or replaced to the same location after the procedure. This made it possible to obtain rare pre- and post-operative recordings allowing the researchers to evaluate signals from brain areas far away from the hub, including speech and language areas distant from the surgery site. Analysis of the change in responses to speech sounds before and after the loss of the hub revealed a rapid disruption of signaling and subsequent partial compensation of the broader brain network.

“The rapid impact on the speech and language processing regions well removed from the surgical treatment site was surprising, but what was even more surprising was how the brain was working to compensate, albeit incompletely within this short timeframe,” says Petkov, who also holds an appointment at Newcastle University Medical School in the UK.

The findings disprove theories challenging the necessity of specific brain hubs by showing that the hub was important to maintain normal brain processing in language.

“Neurosurgical treatment and new technologies continue to improve the treatment options provided to patients,” says Howard, who also is a member of the Iowa Neuroscience Institute. “Research such as this underscores the importance of safely obtaining and comparing electrical recordings pre and post operatively, particularly when a brain hub might be affected.”

According to the researchers, the observation on the nature of the immediate impact on a neural network and its rapid attempt to compensate provides evidence in support of a brain theory proposed by Professor Karl Friston at University College London, which posits that any self-organizing system at equilibrium works towards orderliness by minimizing its free energy, a resistance of the universal tendency towards disorder. These neurobiological results following human brain hub disconnection were consistent with several predictions of this and related neurobiological theories, showing how the brain works to try to regain order after the loss of one of its hubs.

In addition to Petkov and Howard, the research team included researchers in the UI Departments of Neurosurgery, Radiology, and Psychological and Brain Sciences, as well as colleagues from Newcastle University, UCL, and University of Cambridge in the UK, and from Carnegie Mellon University, University of Wisconsin-Madison, and Gonzaga University in the United States.

The research was funded in part by grants from National Institutes of Health, the Wellcome Trust. and the European Research Council.

In the study, the research team showed that Zika’s enzyme — NS2B-NS3 — is a multipurpose tool with two essential functions: breaking up proteins (a protease) and dividing its own double-stranded RNA into single strands (a helicase).

“We found that Zika’s enzyme complex changes function based on how it’s shaped,” says Alexey Terskikh, Ph.D., associate professor at Sanford Burnham Prebys and senior author of the paper. “When in the closed conformation, it acts as a classic protease. But then it cycles between open and super-open conformations, which allows it to grab and then release a single strand of RNA — and these functions are essential for viral replication.”

Zika is an RNA virus that’s part of a family of deadly pathogens called flaviviruses, which include West Nile, dengue fever, yellow fever, Japanese encephalitis and others. The virus is transmitted by mosquitoes and infects uterine and placental cells (among other cell types), making it particularly dangerous for pregnant women. Once inside host cells, the virus re-engineers them to produce more Zika.

Understanding Zika on the molecular level could have an enormous payoff: a therapeutic target. It would be difficult to create safe drugs that target the domains of the enzyme needed for protease or helicase functions, as human cells have many similar molecules. However, a drug that blocks Zika’s conformational changes could be effective. If the complex can’t shape-shift, it can’t perform its critical functions, and no new Zika particles would be produced.

An efficient machine

Researchers have long known that Zika’s essential enzyme was composed of two units: NS2B-NS3pro and NS3hel. NS2B-NS3pro carries out protease functions, cutting long polypeptides into Zika proteins. However, NS2B-NS3pro’s abilities to bind single-stranded RNA and help separate the double-stranded RNA during viral replication were only recently discovered.

In this study, the researchers leaned on recent crystal structures and used protein biochemistry, fluorescence polarization and computer modeling to dissect NS2B-NS3pro’s life cycle. NS3pro is connected to NS3hel (the helicase) by a short amino acid linker and becomes active when the complex is in its closed conformation, like a closed accordion. The RNA binding happens when the complex is open, whereas the complex must transition through the super-open conformation to release RNA.

These conformational changes are driven by the dynamics of NS3hel activity, which extends the linker and eventually “yanks” the NS3pro to release RNA. NS3pro is anchored to the inside of the host cell’s endoplasmic reticulum (ER) — a key organelle that helps shepherd cellular proteins to their appropriate destinations — via NS2B and, while in the closed conformation, cuts up the Zika polypeptide, helping generate all viral proteins.

On the other side of the linker, NS3hel separates Zika’s double-stranded RNA and conveniently hands a strand over to NS3pro, which has positively charged “forks” to grab on to the negatively charged RNA.

“There’s a very nice groove of positive charges,” says Terskikh. “So, RNA just naturally follows that groove. Then the complex shifts to the closed conformation and releases the RNA.”

As NS3hel reaches forward to grab the double-stranded RNA, it pulls the complex with it; however, since the NS3pro is anchored in the ER membrane, and the linker can only extend so far, the complex snaps into the super-open conformation and releases RNA. The complex then relaxes back to the open conformation, ready for a new cycle.

Meanwhile, when NS3pro detects a viral polypeptide to cut, it forces the complex into the closed conformation, becoming a protease. The authors call this process “reverse inchworm,” because grabbing and releasing the single-stranded RNA resembles inchworm movements, but backward, with the jaws (the protease) trailing behind.

In addition to providing a possible therapeutic target for Zika, this detailed understanding could be applied to other flaviviruses, which share similar molecular machinery.

“Versions of the NS2B-NS3pro complex are found throughout the flaviviruses,” says Terskikh. “It could potentially constitute a whole new class of drug targets for multiple viruses.”

This device, fabricated on a silicon wafer, takes input laser light of one frequency and converts it into an evenly spaced set of many distinct frequencies that form a train of pulses whose length can be as short as 100 femtoseconds (quadrillionths of a second). (The comb in the name comes from the frequencies being spaced like the teeth of a hair comb.)

Now Vahala (BS ’80, MS ’81, PhD ’85), Caltech’s Ted and Ginger Jenkins Professor of Information Science and Technology and Applied Physics and executive officer for applied physics and materials science, along with members of his research group and the group of John Bowers at UC Santa Barbara, have made a breakthrough in the way the short pulses form in an important new material called ultra-low-loss silicon nitride (ULL nitride), a compound formed of silicon and nitrogen. The silicon nitride is prepared to be extremely pure and deposited in a thin film.

In principle, short-pulse microcomb devices made from this material would require very low power to operate. Unfortunately, short light pulses (called solitons) cannot be properly generated in this material because of a property called dispersion, which causes light or other electromagnetic waves to travel at different speeds, depending on their frequency. ULL has what is known as normal dispersion, and this prevents waveguides made of ULL nitride from supporting the short pulses necessary for microcomb operation.

In a paper appearing in Nature Photonics, the researchers discuss their development of the new microcomb, which overcomes the inherent optical limitations of ULL nitride by generating pulses in pairs. This is a significant development because ULL nitride is created with the same technology used for manufacturing computer chips. This kind of manufacturing technique means that these microcombs could one day be integrated into a wide variety of handheld devices similar in form to smartphones.

The most distinctive feature of an ordinary microcomb is a small optical loop that looks a bit like a tiny racetrack. During operation, the solitons automatically form and circulate around it.

“However, when this loop is made of ULL nitride, the dispersion destabilizes the soliton pulses,” says co-author Zhiquan Yuan (MS ’21), a graduate student in applied physics.

Imagine the loop as a racetrack with cars. If some cars travel faster and some travel slower, then they will spread out as they circle the track instead of staying as a tight pack. Similarly, the normal dispersion of ULL means light pulses spread out in the microcomb waveguides, and the microcomb ceases to work.

The solution devised by the team was to create multiple racetracks, pairing them up so they look a bit like a figure eight. In the middle of that ‘8,’ the two tracks run parallel to each other with only a tiny gap between.

If we continue with the racetrack analogy, this would be like two tracks sharing one straightaway. As the cars from each track converge on that shared section, they encounter something like a traffic jam. Just like two lanes of traffic merging into one on a freeway forces cars to slow down, the conjoined section of the two microcombs forces the paired laser pulses to bunch up. This bunching up counteracts the pulses’ tendency to spread out and allows the microcombs to work properly.

“In effect, this counteracts the normal dispersion and gives the overall composite system the equivalent of anomalous dispersion,” says graduate student and co-author Maodong Gao (MS ’22).

The idea extends when one adds even more racetracks, and the team has shown how three racetracks will also operate by creating two sets of pulse pairs. Vahala believes the phenomenon will continue to work even with many coupled racetracks (microcombs), thereby offering a way to create large photonic circuit arrays for the soliton pulses.

As noted above, these ULL microcombs are fabricated with the same equipment used to make computer chips based on complementary metal-oxide-semiconductor (CMOS) technology. Bowers, a professor of electrical and computer engineering, collaborated on the research and notes that “The manufacturing scalability of the CMOS process means that it will now be easier and more economical to manufacture the short-pulse microcombs and integrate them into existing technologies and applications.”

Concerning these applications, Vahala says “a comb is like a Swiss army knife for optics. It has many different functions, and that’s why it’s such a powerful tool.”

Funding for the research was provided by the Defense Advanced Research Projects Agency, the Defense Threat Reduction Agency Joint Science and Technology Office for Chemical and Biological Defense, and the Air Force Office of Scientific Research.

Expanding on Dr. Fang Liu’s earlier work that identified a novel drug target for the treatment of MS, she and her team have now created a small molecule compound that is effective in two different animal models of MS. This represents a key advancement that brings this MS research closer to the clinic to impact patient care.

MS is a progressive neurological disease that currently has no cure. It is associated with a wide-range of debilitating symptoms, including problems with coordination, cognition, muscle weakness and depression. For unknown reasons, it is more common in northern latitudes and more than twice as common in women.

It is known that MS damages myelin, a protective sheath that forms around nerves in the brain and spinal cord. As the myelin damage is triggered by inflammation in the immune system, up until now all current drug treatments for MS target the immune system.

In this study, CAMH Senior Scientist Dr. Fang Liu and her team treated MS in a completely different way — targeting the glutamate system. Study results showed that the newly synthesized lead compound not only reduced MS-like symptoms, it also may repair the damaged myelin in two different pre-clinical models of MS.

“Our compound had a stunning effect on rescuing myelin and motor function in the lab models, and I hope these effects will translate to the clinic to add to current treatments and bring new hope to patients with MS,” said Dr. Liu. “As with cancer chemotherapy drug cocktails, simultaneous targeting of the MS disease pathway at multiple points can have synergistic effects and result in better outcomes.”

Dr. Iain Greig, Reader in Medicinal Chemistry at the University of Aberdeen, alongside his team, are working to turn the molecules identified by Dr. Liu into advanced “drug-like” molecules suitable for continued development towards clinical use in patients. He added: “In all my years as a medicinal chemist, I have never seen a more promising starting point for a drug development project. It has been a huge pleasure to be involved in this program and I am looking forward to continuing to drive it towards to the clinic.”

Much of the funding for this novel treatment for MS, which Dr. Fang and her team have been investigating for over a decade, has come from the Multiple Sclerosis Society of Canada and the National Multiple Sclerosis Society USA’s Fast Forward commercial research program.

“We are pleased to have helped enable the early development of a novel neuroprotective strategy for MS, and look forward to seeing it progress through the critical next stages needed to determine its potential benefits for people living with MS,” said Walt Kostich, PhD, head of the National MS Society (USA)’s Fast Forward commercial research program.

Dr. Liu believes that the evidence of efficacy and tolerability generated in this study for the small molecule drug makes it a good candidate to be developed for human trials. The next steps in drug development will involve some further pre-clinical research, including investigating safety and stability of the compound. CAMH and the University of Aberdeen have already filed patent applications to protect this research and are actively seeking industry partners to further advance this work towards clinical trials over the next few years.

In 1957, evolutionary biologist George Williams proposed that genetic mutations that contribute to aging could be favored by natural selection if they are advantageous early in life in promoting earlier reproduction or the production of more offspring. Williams was an assistant professor at Michigan State University at the time.

Williams’ idea, now known as the antagonistic pleiotropy theory of aging, remains the prevailing evolutionary explanation of senescence, the process of becoming old or aging. While the theory is supported by individual case studies, it has lacked unambiguous genome-wide evidence.

In the new study, scheduled for publication Dec. 8 in Science Advances, U-M evolutionary biologist Jianzhi Zhang and a Chinese colleague tested the Williams hypothesis using genetic, reproductive and death-registry information from 276,406 participants in the United Kingdom’s Biobank database.

They found reproduction and lifespan to be genetically strongly negatively correlated, meaning that genetic mutations that promote reproduction tend to shorten lifespan.

In addition, individuals carrying mutations that predispose them to relatively high reproductive rates have lower probabilities of living to age 76 than those carrying mutations that predispose them to relatively low reproductive rates, according to the study.

However, the authors caution that reproduction and lifespan are affected by both genes and the environment. And compared with environmental factors — including the impacts of contraception and abortion on reproduction and medical advances on lifespan — the genetic factors discussed in the study play a relatively minor role, according to the authors.

“These results provide strong support for the Williams hypothesis that aging arises as a byproduct of natural selection for earlier and more reproduction. Natural selection cares little about how long we live after the completion of reproduction, because our fitness is largely set by the end of reproduction,” said Zhang, the Marshall W. Nirenberg Collegiate Professor in the U-M Department of Ecology and Evolutionary Biology.

Fitness is a concept biologists use to describe the degree to which an organism’s characteristics increase its number of offspring.

“Interestingly, we found that when you control for the genetically predicted amount and timing of reproduction, having two kids corresponds to the longest lifespan,” Zhang said. “Having fewer or more kids both lower the lifespan.” That result supports the findings of several previous studies.

Zhang’s co-author on the Science Advances paper is Erping Long of the Chinese Academy of Medical Sciences and Peking Union Medical College. Long was a visiting student at U-M when the study began.

In genetics, the concept of pleiotropy posits that a single mutation can influence multiple traits. The idea that the same mutation can be both beneficial and harmful, depending on the situation, is known as antagonistic pleiotropy and was proposed by Williams to underlie the origin of aging in a paper titled “Pleiotropy, natural selection, and the evolution of senescence.”

To a biologist, senescence refers specifically to a gradual decline of bodily functions that manifests as a decline in reproductive performance and an increase in the death rate with age.

The U.K.’s Biobank database enabled Zhang and Long to assess the genetic relationship between reproduction and lifespan at the genomic scale.

The researchers examined the frequency of 583 reproduction-associated genetic variants in the database and found that several of the variants associated with higher reproduction have become more common in recent decades, despite their simultaneous associations with shorter lifespan. The increased frequency of the variants is presumably a result of natural selection for higher reproduction.

“The antagonistic pleiotropy hypothesis predicts that most mutations that increase reproduction but reduce lifespan have larger fitness advantages than disadvantages so are selectively favored,” Zhang said.

Even so, human life expectancy, birth rate and reproductive behavior have all changed drastically in the last few decades. Specifically, more than half of humans live in areas of the world where birth rates have declined, along with increased incidences of contraception, abortion and reproductive disorder, according to the new study.

Global human life expectancy at birth, on the other hand, has steadily increased from 46.5 years in 1950 to 72.8 years in 2019.

“These trends are primarily driven by substantial environmental shifts, including changes in lifestyles and technologies, and are opposite to the changes caused by natural selection of the genetic variants identified in this study,” Zhang said. “This contrast indicates that, compared with environmental factors, genetic factors play a minor role in the human phenotypic changes studied here.”

Funding for the study was provided by the U.S. National Institutes of Health, the National Natural Science Foundation of China and the Chinese Academy of Medical Sciences Innovation Fund.