AI Recognises the Biological Activity of Natural Products

Nature has a vast store of medicinal substances. “Over 50 percent of all drugs today are inspired by nature,” says Gisbert Schneider, Professor of Computer-​Assisted Drug Design at ETH Zurich. Nevertheless, he is convinced that we have tapped only a fraction of the potential of natural products. Together with his team, he has successfully demonstrated how artificial intelligence (AI) methods can be used in a targeted manner to find new pharmaceutical applications for natural products. Furthermore, AI methods are capable of helping to find alternatives to these compounds that have the same effect but are much easier and therefore cheaper to manufacture.

And so the ETH researchers are paving the way for an important medical advance: we currently have only about 4,000 basically different medicines in total. In contrast, estimates of the number of human proteins reach up to 400,000, each of which could be a target for a drug. There are good reasons for Schneider’s focus on nature in the search for new pharmaceutical agents.

Most natural products are by definition potential active ingredients that have been selected via evolutionary mechanisms,” he says.
Whereas scientists used to trawl collections of natural products on the search for new drugs, Schneider and his team have flipped the script: first, they look for possible target molecules, typically proteins, of natural products so as to identify the pharmacologically relevant compounds. “The chances of finding medically meaningful pairs of active ingredient and target protein are much greater using this method than with conventional screening,” Schneider says.


New Biosensor Measures The Concentration Of Covid-19 In The Air

A team of researchers from Empa, ETH Zurich and Zurich University Hospital has succeeded in developing a novel sensor for detecting the new coronavirus. In future it could be used to measure the concentration of the virus in the environment – for example in places where there are many people or in hospital ventilation systems.

Jing Wang and his team at Empa and ETH Zurich usually work on measuring, analyzing and reducing airborne pollutants such as aerosols and artificially produced nanoparticles. However, the challenge the whole world is currently facing is also changing the goals and strategies in the research laboratories. The new focus: a sensor that can quickly and reliably detect SARS-CoV-2 – the new coronavirus.

But the idea is not quite so far removed from the group’s previous research work: even before the COVID-19 began to spread, first in China and then around the world, Wang and his colleagues were researching sensors that could detect bacteria and viruses in the air. The sensor will not necessarily replace the established laboratory tests, but could be used as an alternative method for clinical diagnosis, and more prominently to measure the virus concentration in the air in real time: For example, in busy places like train stations or hospitals.

Fast and reliable tests for the new coronavirus are urgently needed to bring the pandemic under control as soon as possible. Most laboratories use a molecular method called reverse transcription polymerase chain reaction, or RT-PCR for short, to detect viruses in respiratory infections. This is well established and can detect even tiny amount of viruses – but at the same time it can be time consuming and prone to error.

Jing Wang and his team have developed an alternative test method in the form of an optical biosensor. The sensor combines two different effects to detect the virus safely and reliably: an optical and a thermal one.

The sensor uses an optical and a thermal effect to detect the COVID-19-Virus safely and reliably

The sensor is based on tiny structures of gold, so-called gold nanoislands, on a glass substrate. Artificially produced DNA receptors that match specific RNA sequences of the SARS-CoV-2 are grafted onto the nanoislands. The coronavirus is a so-called RNA virus: Its genome does not consist of a DNA double strand as in living organisms, but of a single RNA strand. The receptors on the sensor are therefore the complementary sequences to the virus’ unique RNA sequences, which can reliably identify the virus.

The technology the researchers use for detection is called LSPR, short for localized surface plasmon resonance. This is an optical phenomenon that occurs in metallic nanostructures: When excited, they modulate the incident light in a specific wavelength range and create a plasmonic near-field around the nanostructure. When molecules bind to the surface, the local refractive index within the excited plasmonic near-field changes. An optical sensor located on the back of the sensor can be used to measure this change and thus determine whether the sample contains the RNA strands in question.


Ultrasound Can Selectively Kill Cancer Cells

A new technique could offer a targeted approach to fighting cancer: low-intensity pulses of ultrasound have been shown to selectively kill cancer cells while leaving normal cells unharmed.

Ultrasound wavessound waves with frequencies higher than humans can hear—have been used as a cancer treatment before, albeit in a broad-brush approach: high-intensity bursts of ultrasound can heat up tissue, killing cancer and normal cells in a target area. Now, scientists and engineers are exploring the use of low-intensity pulsed ultrasound (LIPUS) in an effort to create a more selective treatment.

A study describing the effectiveness of the new approach in cell models was published in Applied Physics Letters. The researchers behind the work caution that it is still preliminary—it still has not been tested in a live animal let alone in a human, and there remain several key challenges to address—but the results so far are promising.

The research began five years ago when Caltech‘s Michael Ortiz, Frank and Ora Lee Marble Professor of Aeronautics and Mechanical Engineering, found himself pondering whether the physical differences between cancer cells and healthy cells—things like size, cell-wall thickness, and size of the organelles within them—might affect how they vibrate when bombarded with sound waves and how the vibrations might trigger cancer cell death.

I have my moments of inspiration,” Ortiz says wryly.

And so Ortiz built a mathematical model to see how cells would react to different frequencies and pulses of sound waves. Together with then-graduate student Stefanie Heyden (PhD ’14), who is now at ETH Zurich, Ortiz published a paper in 2016 in the Journal of the Mechanics and Physics of Solids showing that there was a gap in the so-called resonant growth rates of cancerous and healthy cells. That gap meant that a carefully tuned sound wave could, in theory, cause the cell membranes of cancerous cells to vibrate to the point that they ruptured while leaving healthy cells unharmed. Ortiz dubbed the process “oncotripsy” from the Greek oncos (for tumor) and tripsy (for breaking).


Real Gold, Almost As Light As Air

Researchers at ETH Zurich have created a new type of foam made of real gold. It is the lightest form ever produced of the precious metal: a thousand times lighter than its conventional form and yet it is nearly impossible to tell the difference with the naked eye. There are many possible applications.

A nugget of real 20 carats gold, so light that it does not sink in a cappuccino, floating instead on the milk foam – what sounds unbelievable has actually been accomplished by researchers from ETH Zurich. Scientists led by Raffaele Mezzenga, Professor of Food and Soft Materials, have produced a new kind of foam out of gold, a three-​dimensional mesh of gold that consists mostly of pores. It is the lightest gold nugget ever created.

Even when it seems unbelievable: this is a genuine photograph, in which nothing has been faked. The 20 carats gold foam is lighter than milk foam

The so-​called aerogel is a thousand times lighter than conventional gold alloys. It is lighter than water and almost as light as air,” says Mezzenga.

The new gold form can hardly be differentiated from conventional gold with the naked eye – the aerogel even has a metallic shine. But in contrast to its conventional form, it is soft and malleable by hand. It consists of 98 parts air and only two parts of solid material. Of this solid material, more than four-​fifths are gold and less than one-​fifth is milk protein fibrils. This corresponds to around 20 carat gold.

The scientists created the porous material by first heating milk proteins to produce nanometre-​fine protein fibres, so-​called amyloid fibrils, which they then placed in a solution of gold salt. The protein fibres interlaced themselves into a basic structure along which the gold simultaneously crystallised into small particles. This resulted in a gel-​like gold fibre network.

One of the big challenges was how to dry this fine network without destroying it,” explains Gustav Nyström, postdoc in Mezzenga’s group and first author of the corresponding study in the journal Advanced Materials. As air drying could damage the fine gold structure, the scientists opted for a gentle and laborious drying process using carbon dioxide. They did so in an interdisciplinary effort assisted by researchers in the group of Marco Mazzotti, Professor of Process Engineering.


CRISPR Can Now Edit Multiple Genes At Once

Researchers at ETH Zurich have refined the famous CRISPR-Cas method. Now, for the very first time, it is possible to modify dozens, if not hundreds, of genes in a cell simultaneously.

Everyone’s talking about CRISPR-Cas. This biotechnological method offers a relatively quick and easy way to manipulate single genes in cells, meaning they can be precisely deleted, replaced or modified. Furthermore, in recent years, researchers have also been using technologies based on CRISPR-Cas to systematically increase or decrease the activity of individual genes. The corresponding methods have become the worldwide standard within a very short time, both in basic biological research and in applied fields such as plant breeding.

To date, for the most part, researchers could modify only one gene at a time using the method. On occasion, they managed two or three in one go; in one particular case, they were able to edit seven genes simultaneously. Now, Professor Randall Platt and his team at the Department of Biosystems Science and Engineering at ETH Zurich in Basel have developed a process that – as they demonstrated in experiments – can modify 25 target sites within genes in a cell at once. As if that were not enough, this number can be increased still further, to dozens or even hundreds of genes, as Platt points out. At any rate, the method offers enormous potential for biomedical research and biotechnology. “Thanks to this new tool, we and other scientists can now achieve what we could only dream of doing in the past.

Genes and proteins in cells interact in many different ways. The resulting networks comprising dozens of genes ensure an organism’s cellular diversity. For example, they are responsible for differentiating progenitor cells to neuronal cells and immune cells. “Our method enables us, for the first time, to systematically modify entire gene networks in a single step,” Platt says.

Moreover, it paves the way for complex, large-scale cell programming. It can be used to increase the activity of certain genes, while reducing that of others. The timing of this change in activity can also be precisely controlled.

This is of interest for basic research, for example in investigating why various types of cells behave differently or for the study of complex genetic disorders. It will also prove useful for cell replacement therapy, which involves replacing damaged with healthy cells. In this case, researchers can use the method to convert stem cells into differentiated cells, such as neuronal cells or insulin-producing beta cells, or vice versa, to produce stem cells from differentiated skin cells.


The Mind Controls Remotely Videogames

Scientists in Switzerland have developed a system which allows people with severely-impaired motor functions, such as quadriplegia, to use video games using only the power of their brain.

Samuel Kunz, who was paralysed after an accident, uses the brain-computer interface to control an avatar through a race course in a specially-designed computer game called ‘Brain Driver’. The ultimate aim of the research is to develop technology to control devices such as wheelchairs for those with a limited ability to move. Kunz, who is taking part in the trial, is able to ‘pilot’ the digital race-car using only his brain signals transmitted to a computer via electrodes placed on his head.


These electrodes are connected to an amplifier and then to the computer and to our algorithms in the end. The algorithms are then calculating the brain signal and sending commands to the game that our pilot can actually control,” Dr. Rea Lehner, a neuroscientist at ETH Zurich explained. Lehner added Kunz is training his mind by imagining certain actions which are then translated into signals to control the race car. Thinking about moving his left hand makes the car turn left, thinking about moving his right hand turns the car right, and moving both together makes the car go straight. A fourth command – fully relaxing and clearing his mind – slows the car down. Kunz said it has taken a lot of practise to train his mind to control the game; which will be made even more difficult in a stadium full of people. He will be among those taking part in a special championship next year called Cybathlon in which people with physical disabilities compete against each other using state-of-the-art technology.

I have to be very concentrated. The connection between my fingers and my brain is not there anymore. I still try to move my fingers just in my head and so that needs a lot of concentration to do it exactly the same way every time,” Kunz told Reuters during a training session in Zurich.


Trees Will Save The Climate

Around 0.9 billion hectares of land worldwide would be suitable for reforestation, which could ultimately capture two thirds of human-made carbon emissions. The Crowther Lab of ETH Zurich has published a study in the journal Science that shows this would be the most effective method to combat climate changeThe Crowther Lab at ETH Zurich investigates nature-based solutions to climate change. In their latest study the researchers showed for the first time where in the world new trees could grow and how much carbon they would store.

Reforestation would be the most effective method to combat climate change

One aspect was of particular importance to us as we did the calculations: we excluded cities or agricultural areas from the total restoration potential as these areas are needed for human life,” explains study lead author and postdoc at the Crowther Lab Jean-François Bastin.

The researchers calculated that under the current climate conditions, Earth’s land could support 4.4 billion hectares of continuous tree cover. That is 1.6 billion more than the currently existing 2.8 billion hectares. Of these 1.6 billion hectares, 0.9 billion hectares fulfill the criterion of not being used by humans. This means that there is currently an area of the size of the US available for tree restoration. Once mature, these new forests could store 205 billion tonnes of carbon: about two thirds of the 300 billion tonnes of carbon that has been released into the atmosphere as a result of human activity since the Industrial Revolution.

According to Prof. Thomas Crowther, co-author of the study and founder of the Crowther Lab at ETH Zurich: “We all knew that restoring forests could play a part in tackling climate change, but we didn’t really know how big the impact would be. Our study shows clearly that forest restoration is the best climate change solution available today. But we must act quickly, as new forests will take decades to mature and achieve their full potential as a source of natural carbon storage.”

The study also shows which parts of the world are most suited to forest restoration. The greatest potential can be found in just six countries: Russia (151 million hectares); the US (103 million hectares); Canada (78.4 million hectares); Australia (58 million hectares); Brazil (49.7 million hectares); and China (40.2 million hectares).

Many current climate models are wrong in expecting climate change to increase global tree cover, the study warns. It finds that there is likely to be an increase in the area of northern boreal forests in regions such as Siberia, but tree cover there averages only 30 to 40 percent. These gains would be outweighed by the losses suffered in dense tropical forests, which typically have 90 to 100 percent tree cover.

A tool on the Crowther Lab website< enables users to look at any point on the globe, and find out how many trees could grow there and how much carbon they would store. It also offers lists of forest restoration organisations. The Crowther Lab will also be present at this year’s Scientifica to show the new tool to visitors.

The Crowther Lab uses nature as a solution to:

1) better allocate resources – identifying those regions which, if restored appropriately, could have the biggest climate impact;

2) set realistic goals – with measurable targets to maximise the impact of restoration projects;

3) monitor progress – to evaluate whether targets are being achieved over time, and take corrective action if necessary.


Nanorobots Deliver Drugs Directly To Diseased Tissue

Scientists at EPFL and ETH Zurich in Switzerland have developed tiny elastic robots that can change shape depending on their surroundings. Modeled after bacteria and fully biocompatible, these robots optimize their movements so as to get to hard-to-reach areas of the human body. They stand to revolutionize targeted drug delivery.

One day we may be able to ingest tiny robots that deliver drugs directly to diseased tissue, thanks to research being carried out at EPFL and ETH Zurich.


The robots are modeled after bacteria and fully biocompatible© 2019 EPFL/ ETHZ

The group of scientists – led by Selman Sakar at EPFL and Bradley Nelson at ETH Zurich – drew inspiration from bacteria to design smart, biocompatible microrobots that are highly flexible. Because these devices are able to swim through fluids and modify their shape when needed, they can pass through narrow blood vessels and intricate systems without compromising on speed or maneuverability. They are made of hydrogel nanocomposites that contain magnetic nanoparticles allowing them to be controlled via an electromagnetic field.

In an article appearing in Science Advances, the scientists describe the method they have developed for “programming the robot’s shape so that it can easily travel through fluids that are dense, viscous or moving at rapid speeds. When we think of robots, we generally think of bulky machines equipped with complex systems of electronics, sensors, batteries and actuators. But on a microscopic scale, robots are entirely different.

Fabricating miniaturized robots presents a host of challenges, which the scientists addressed using an origami-based folding method. Their novel locomotion strategy employs embodied intelligence, which is an alternative to the classical computation paradigm that is performed by embedded electronic systems.Our robots have a special composition and structure that allow them to adapt to the characteristics of the fluid they are moving through. For instance, if they encounter a change in viscosity or osmotic concentration, they modify their shape to maintain their speed and maneuverability without losing control of the direction of motion,” says Sakar.

These deformations can be “programmed” in advance so as to maximize performance without the use of cumbersome sensors or actuators. The robots can be either controlled using an electromagnetic field or left to navigate on their own through cavities by utilizing fluid flow. Either way, they will automatically morph into the most efficient shape.