Connect with us

Tech

Powerful algorithms used by Netflix, Amazon and Facebook could “hack the language of cancer and Alzheimer’s”

Published

on

Powerful algorithms used by Netflix, Amazon and Facebook could "hack the language of cancer and Alzheimer's"


Fluorescence microscopy image of protein condensates formed inside living cells. Credit: Weitz Laboratory, Harvard University.

Artificial intelligence can predict the biological language of cancer and neurodegenerative diseases such as Alzheimer’s disease, scientists have found out.

Scientists have found that powerful algorithms used by Netflix, Amazon and Facebook can “predict” the biological language of cancer and neurodegenerative diseases such as Alzheimer’s.

Big data from decades of research has been fed into a computer language model to see if artificial intelligence can make more advanced discoveries than humans.

Scientists at St John’s College, Cambridge University have discovered that machine learning technology can decipher the “biological language” of cancer, Alzheimer’s and other neurodegenerative diseases.

Their groundbreaking research has been published in a scientific journal. PNAS April 8, 2021 and may be used in the future to “correct grammatical errors within cells that cause disease.”

Protein condensates formed inside living cells by fluorescence microscopy

Fluorescence microscopy image of protein condensates formed inside living cells. Credit: Weitz Laboratory, Harvard University.

Professor Tuomas Knowles, lead author of the article and researcher at St John’s College, said: “The use of machine learning technology in neurodegenerative disease and cancer research is game-changing. Ultimately, the goal will be to use artificial intelligence to develop targeted drugs to dramatically alleviate symptoms or prevent dementia altogether. ”

Whenever Netflix recommends a show to watch or Facebook invites someone to befriend, the platforms use powerful machine learning algorithms to make educated guesses about what people will do next. Voice assistants like Alexa and Siri can even recognize individuals and instantly “talk” to you.

Dr.Kadi Liis Saar, the article’s first author and research assistant at St John’s College, used similar machine learning technology to train a large-scale language model to see what happens when something goes wrong with proteins inside the body to make disease.

She said: “The human body is home to thousands and thousands of proteins, and scientists do not yet know the functions of many of them. We asked a neural network-based language model to learn the language of proteins.

Formation of Protein Condensates by Fluorescence Microscopy

Fluorescence microscopy image of protein condensates formed inside living cells. Credit: Weitz Laboratory, Harvard University.

“We specifically asked the program to learn the language of shape-altering biomolecular condensates – droplets of proteins found in cells – that scientists really need to understand in order to crack the language of biological function and malfunction that causes cancer and neurodegenerative diseases like Alzheimer’s. We found that he can learn, without explicit explanation, what scientists have already discovered about the language of proteins over decades of research. “

Proteins are large, complex molecules that play an important role in the body. They do most of the work in cells and are required for the structure, function, and regulation of tissues and organs in the body – for example, antibodies are a protein that protects the body.

Alzheimer’s, Parkinson’s and Huntington’s are the three most common neurodegenerative diseases, but scientists believe there are several hundred.

In Alzheimer’s disease, which affects 50 million people worldwide, proteins don’t work, clot and kill healthy nerve cells. A healthy brain has a quality control system that effectively gets rid of potentially dangerous masses of proteins known as aggregates.

Scientists now believe that some disordered proteins also form liquid droplets of proteins called condensates, which have no membrane and merge freely with each other. Unlike irreversible protein aggregates, protein condensates can form and be transformed, and they are often compared to the shape-shifting wax droplets in lava lamps.

Professor Knowles said: “Protein condensates have recently received a lot of attention in the scientific world because they control key events in the cell such as gene expression – like ours DNA turns into proteins – and protein synthesis – as cells produce proteins.

“Any defects associated with these protein droplets can lead to diseases such as cancer. This is why the use of natural language processing technology in research on the molecular origins of protein malfunctions is vital if we want to be able to correct the grammatical errors within cells that cause disease. ”

Dr. Saar said: “We fed the algorithm all the data about known proteins so that it could learn and predict the language of proteins in the same way that models learn about human language and how WhatsApp knows how to suggest words for you.

“Then we were able to ask him about a specific grammar that causes only a few proteins to condense inside cells. This is a very difficult problem, and solving it will help us learn the rules of the language of disease. “

Machine learning technology is evolving at a rapid pace due to the increasing availability of data, increased computing power, and technological advances that have led to the creation of more powerful algorithms.

The continued use of machine learning could change the future of cancer and neurodegenerative disease research. Discoveries can be made that go beyond what scientists currently know and think about disease, and perhaps even beyond what the human brain can understand without the aid of machine learning.

Dr. Saar explained: “Machine learning can be free of the constraints that researchers believe are the goals of scientific research, and this will mean that new connections will be discovered that we have not even thought about. It’s really very interesting. “

The network created is now done freely available researchers around the world so that other scientists can work on advances.

Citation: “Sequence Determinants Based Molecular Grammar of Protein Condensates and Embedding” Kadi L. Saar, Alexey S. Morgunov, Runjang Qi, William E. Arter, Georg Krainer, Alpha A. Lee and Tuomas PJ Knowles, April 7 2021 Proceedings of the National Academy of Sciences
DOI: 10.1073 / pnas.2019053118





Originally reported by Source link

The featured images are, as they appear on the original report.

Tech

Harvesting energy from radio waves to power wearable electronic devices

Published

on

Flexible Self Powered Electronics


An international team of researchers led by Huanyu “Larry” Cheng, Dorothy Quiggle’s career development professor at the Pennsylvania Department of Engineering and Mechanics, has developed a stretchable antenna and rectenna system that harvests energy from radio waves in the environment to generate energy for wearable devices. Credit: Larry Cheng, PA.

Radio waves entering the environment, from microwaves to Wi-Fi connections, are not only signals of energy consumption, but are also energy sources themselves. An international team of researchers led by Huanyu “Larry” Cheng, Dorothy Quiggle’s career development professor at the Pennsylvania Department of Engineering and Mechanics, has developed a way to harvest radio wave energy to power wearable devices.

Researchers recently published their method in Materials Today Physics.

Current sources of energy for wearable health monitoring devices have their place in powering sensory devices, Cheng said, but each has its own drawbacks. For example, solar energy can only harvest energy when exposed to the sun. A self-powered triboelectric device can only collect energy when the body is in motion.

“We do not want to replace any of these current energy sources,” Cheng said. “We are trying to provide additional permanent energy.”

Researchers have developed a stretchable broadband dipole antenna system capable of wirelessly transmitting data from health monitoring sensors. The system consists of two expandable metal antennas embedded in the conductive graphene metal coated material. The system’s broadband design allows it to maintain its frequency functions even when stretched, bent and twisted. This system is then connected to a stretchable rectifier circuit, creating a rectified antenna, or “rectenna,” capable of converting electromagnetic wave energy into electricity. This electricity can be used to power wireless devices or to charge energy storage devices such as batteries and supercapacitors.

This rectenna can convert radio or electromagnetic waves from the environment into energy to power sensor modules on the device that pulse temperature, hydration, and oxygen levels. Compared to other sources, less power is generated, but the system can generate power continuously, which Cheng says is a significant advantage.

“We use the energy that already surrounds us – radio waves are everywhere and always,” Cheng said. “If we don’t use this environmental energy, it will be wasted. We can collect this energy and transform it into strength. “

Cheng said this technology is a building block for him and his team. Combining it with their new wireless communication device will provide a critical component that will work with existing command sensor modules.

“Our next steps will be to investigate miniature versions of these circuits and work to improve the extensibility of the rectifier,” Cheng said. “This is a platform on which we can easily combine and apply this technology with other modules that we have created in the past. It is easy to extend or adapt for other applications, and we plan to explore these possibilities. “

Link: Stretchable broadband dipole antennas and rectennas for collecting RF energy, by:
Jia Zhu, Zhihui Hu, Chaoyun Song, Ning Yi, Zhaozheng Yu, Zhendong Liu, Shangbin Liu, Mengjun Wang, Michael Gregory Dexheimer, Jian Yang and Huanyu Cheng, March 5, 2021, Materials Today Physics
DOI: 10.1016 / j.mtphys.2021.100377

This article was co-authored by Jia Zhu, who received his Ph.D. in Engineering and Mechanics from Pennsylvania State University in 2020; Zhihui Hu, former visiting professor of engineering and mechanics at the University of Pennsylvania and current associate professor at Wuhan University of Technology in China; Chaoyun Song, Assistant Professor in the Faculty of Engineering and Physical Sciences, Heriot-Watt University in Scotland; Ning Yi, who received her PhD in Engineering and Mechanics from Pennsylvania State University in 2020; Zhaozheng Yu, who received his MS in Engineering and Mechanics from Pennsylvania State University in 2019; Zhendong Liu, Former Visiting Graduate Student, Department of Engineering and Mechanics, University of Pennsylvania; Shangbin Liu, Graduate Student, Department of Engineering and Mechanics, University of Pennsylvania; Mengjun Wang, assistant professor at the School of Electronics and Computer Science, Hebei University of Technology in China; Michael Gregory Dexheimer, who received his MS in Engineering and Mechanics from Pennsylvania State University in 2020; and Jian Yang, professor of biomedical engineering in Pennsylvania.

Support for this work was provided by the National Science Foundation; National Heart, Lung, and Blood Institute, National Institutes of Health; and Penn State.





Originally reported by Source link

The featured images are, as they appear on the original report.

Continue Reading

Tech

Breakthrough in new materials could be the key to revolutionary transparent electronics

Published

on

Flexible, Transparent Electronics


From

The optical transparency of the new materials can enable futuristic, flexible and transparent electronics. Credit: RMIT University.

Filling a critical gap in the material spectrum

A new study published this week could pave the way for the next generation of transparent electronics.

Such transparent devices could potentially be embedded in glass, flexible displays, and smart contact lenses to bring futuristic, sci-fi-like devices to life.

For several decades, researchers have been looking for a new class of electronics based on semiconductor oxides, the optical transparency of which could allow this fully transparent electronics to be used.

Oxide devices can also find applications in power electronics and communications technology, reducing the carbon footprint of our utilities.

A team led by RMIT has introduced ultrafine beta-tellurite in a family of two-dimensional (2D) semiconductor materials, providing an answer to this long-term search for a highly mobile p-type oxide.

“This new, highly mobile p-type oxide fills a critical gap in the material spectrum, enabling fast and transparent circuits,” says Team Leader Dr. Torben Daenecke, who led the collaborative work on the three FLEET nodes.

Other key advantages of long-sought oxide-based semiconductors are their stability in air, less stringent purity requirements, low cost, and ease of deposition.

“The missing link in our advance was finding the right, ‘positive’ approach,” says Torben.

Positivity was lacking

There are two types of semiconductor materials. “N-type” materials contain a large number of negatively charged electrons, while “p-type” semiconductors contain many positively charged holes.

It is the amalgamation of n-type and p-type complementary materials that allows electronic devices such as diodes, rectifiers, and logic circuits to be created.

Deposition of molten metal

The molten mixture of tellurium and selenium, rolled over the surface, precipitates an atomically thin sheet of beta-tellurite. Credit: FLEET

Modern life is critically dependent on these materials as they are the building blocks of every computer and smartphone.

An obstacle to oxide devices has been that although many high performance n-type oxides are known, there is a significant lack of high quality p-type oxides.

Theory prompts action

However, in 2018, a computational study found that beta tellurite (β-TeO2) could be an attractive candidate for p-type oxide, with tellurium’s special place on the periodic table means it can behave as a metal or non-metal, providing it oxide with unique beneficial properties.

“This prediction prompted our team at RMIT University to study its properties and applications,” says Dr. Torben Daenecke, FLEET associate researcher.

Liquid metal – the way to explore 2D materials

Dr. Daenecke’s team demonstrated the separation of beta-tellurite using a specially developed synthesis technology based on liquid metal chemistry.

“A molten mixture of tellurium (Te) and selenium (Se) is prepared and allowed to roll on the surface,” explains one of the authors of the article, Patjari Aukaraserinont.

“Due to the presence of oxygen in the ambient air, the melt drop naturally forms a thin surface oxide layer of beta-tellurite. When a drop of liquid rolls over the surface, this oxide layer sticks to it, depositing atomically thin sheets of oxide on its way. “

“The process is similar to drawing: you use a glass rod as a pen, and liquid metal is your ink,” explains Ms. Aucaraserenont, FLEET PhD student at RMIT.

Ali Zawabeti, Patjari Aukaraserinont and Torben Daeneke

RMIT Team (from left to right): Ali Zawabeti, Patjari Aukaraserinont and Torben Daeneke with transparent electronics. Credit: FLEET

While the desired β-phase of tellurite rises below 300 ° C, pure tellurium has a high melting point, above 500 ° C. Therefore, selenium was added to develop alloy which has a lower melting point, which makes synthesis possible.

“The ultra-thin sheets we got are only 1.5 nanometers thick, which equates to just a few atoms. The material was very transparent in the visible spectrum with a band gap of 3.7 eV, which means they are virtually invisible to the human eye, ”explains co-author Dr. Ali Zawabeti.

Beta Tellurite Evaluation: Up to 100X Faster

To evaluate the electronic properties of the developed materials, field effect transistors (FET) were manufactured.

“These devices showed characteristic p-type switching as well as high hole mobility (about 140 cm2V-1s-1), showing that beta-tellurite is ten to one hundred times faster than existing p-type oxide semiconductors. The excellent on / off ratio (over 106) also proves that this material is suitable for energy efficient and fast devices, ”said Ms. Patjari Aukaraserinont.

“The findings fill a major gap in the digital library of materials,” said Dr. Ali Zawabeti.

“Having a fast transparent p-type semiconductor at our disposal could revolutionize transparent electronics, as well as improve displays and improve energy efficient devices.”

The team plans to further explore the potential of this new semiconductor. “Our further research into this exciting material will focus on integration into existing and next generation consumer electronics,” says Dr. Torben Daenecke.

Link: “Highly mobile semiconductor two-dimensional β-TeO2 p-type” April 5, 2021, Nature Electronics
DOI: 10.1038 / s41928-021-00561-5

FLEET researchers from RMIT, ANU and UNSW collaborated with colleagues from Deakin University and the University of Melbourne. Matthias Wurdak of FLEET (ANU) conducted 2D nanosheets transfer experiments, and Kurosh Kalantarzade (UNSW) helped with material and device characteristics analysis.

This project was supported by the Australian Research Council (Center of Excellence and DECRA programs), the authors also acknowledge support from the RMIT University Microscopy and Microanalysis Foundation (RMMF), RMIT University Micronano-Nanotechnology Research Center (MNRF), and funding obtained through postdoctoral MacKenzie program. University of Melbourne Scholarship Program.





Originally reported by Source link

The featured images are, as they appear on the original report.

Continue Reading

Tech

Terahertz image of graphene paves the way for optimization and industrialization

Published

on

Terahertz image of graphene paves the way for optimization and industrialization


Graphene Flagship researchers have developed a new measurement standard for the analysis of graphene and layered materials that can speed up production and optimize device fabrication. Credit: Graphene Flagship.

Graphene Flagship Researchers Develop a New Measurement Standard for Analysis graphene and multilayer materials that can speed up production and optimize device fabrication.

X-ray scans have revolutionized medical treatment by allowing us to see a person from the inside without surgery. Likewise, terahertz spectroscopy penetrates graphene films, allowing scientists to draw detailed maps of their electrical quality without damaging or contaminating the material. Graphene Flagship has brought together researchers from academia and industry to develop and refine this analytical method, and a new measurement tool is now ready to characterize graphene.

These efforts were made possible by a collaborative environment created by the European consortium Graphene Flagship, with the participation of scientists from Graphene Flagship partners, DTU, Denmark, IIT, Italy, Aalto University, Finland, AIXTRON, UK, imec, Belgium, Graphenea, Spain, University of Warsaw, Poland, and Thales R&T, France, as well as employees in China, Korea and the USA.

Graphene is often sandwiched between many different layers and materials used in electronic and photonic devices. This complicates the quality assessment process. Terahertz spectroscopy simplifies the task. It displays encapsulated materials and shows the quality of the graphene underneath, exposing flaws at critical points in the manufacturing process. It is a fast, non-destructive technology that explores the electrical properties of graphene and laminates without the need for direct contact.

The development of characterization techniques such as terahertz spectroscopy is fundamental to accelerating large-scale production as they ensure that graphene devices are produced consistently and predictably, without drawbacks. Quality control precedes trust. With other developments pioneered by Graphene Flagship, such as roll-to-roll graphene and laminates, fabrication technology is poised to take the next step. Terahertz spectroscopy allows us to ramp up graphene production without losing quality.

Terahertz image of graphene opens the way to industrialization

Terahertz spectroscopy penetrates graphene films, allowing scientists to draw detailed maps of their electrical quality without damaging or contaminating the material. Credit: Peter Baggild (Graphene Flagship / DTU)

“This is the method we need to meet the high performance production levels provided by Graphene Flagship,” explains Peter Baggild of DTU, Graphene Flagship partner. “We are confident that terahertz spectroscopy in graphene production will become as routine an X-ray scan in hospitals,” he adds. “In fact, thanks to terahertz spectroscopy, you can easily map even meter-scale samples of graphene without touching them, which is not possible with some other modern methods.” In addition, Graphene Flagship is currently exploring how to apply terahertz spectroscopy directly to graphene roll-to-roll production lines and accelerate imaging.

Collaboration was key to this achievement. Graphene Flagship researchers in academia have worked closely with leading graphene manufacturers such as Graphene Flagship partners AIXTRON, Graphenea, and IMEC. “This is the best way to ensure that our solution is relevant to our end users, companies that produce graphene and laminates on an industrial scale,” says Böggild. “Our publication is a comprehensive case study that highlights the versatility and reliability of terahertz spectroscopy for quality control and should help our colleagues apply this technique to many industrially important substrates such as silicon, sapphire, silicon carbide and polymers.” he adds.

Setting standards is an important step in the development of any new material to ensure that it is safe, authentic, and offers reliable and consistent performance. This is why Graphene Flagship has a dedicated working group dedicated to graphene standardization, measurement and analytical methods, and manufacturing processes. The newly developed terahertz spectroscopy method will soon become a standard technical specification thanks to the work of the Flagship Committee on Graphene Standardization. “This will undoubtedly accelerate the adoption of this new technology as it will show how the analysis and comparison of graphene samples can be performed in a reproducible manner,” explains Peter Jepsen of Graphene Flagship Partner DTU, co-author of the study. “Terahertz spectroscopy is another step towards increasing confidence in products containing graphene,” he concludes.

Amaya Zurutuza, article co-author and scientific director of Graphene Flagship, a Graphenea partner, says: “At Graphenea, we are convinced that terahertz imaging can help develop quality control methods that can meet manufacturing requirements and provide relevant information on graphene quality. what is needed on our way to the successful industrialization of graphene. “

Turid Gspann, Chair of the Flagship Graphene Standardization Committee, says: “This terahertz [spectroscopy] It is expected that this technology will become widespread in industry. It does not require any special sample preparation and is a mapping method that allows you to analyze large areas with a minimum of time. “

Marco Romagnoli, Head of Electronics and Photonics Integration, Graphene Flagship, adds: “The THz Spectroscopy Instrument for wafer scale applications is a state of the art high TRL system for characterizing multilayer stacks on wafers containing CVD graphene. Works in a short time and with good accuracy, and provides the main parameters of interest such as carrier mobility, conductivity, scattering time, and carrier density. This valuable technical achievement also exemplifies the benefits of being part of a large collaborative project such as the Graphene Flagship. “

Andrea C. Ferrari, Graphene Flagship Science and Technology Officer and Chair of its Management Group, adds: “Once again, Graphene Flagship researchers are pioneering new characterization techniques to drive the development of graphene technology. This helps us move steadily along our innovation and technology roadmap and will drive the commercialization of graphene in a wide range of applications. ”

Reference: “Case Studies of the Electrical Characteristics of Graphene Using Terahertz Time Domain Spectroscopy” by Patrick R. Welan, Bingbin Zhou, Odile Bezensnet, Abhaya Shivayogimatha, Niraja Mishra, Qian Shen, Bjarke S. Jessen, Iwona Pasternak, David MA Mackenzie, G.G. , Kunzhi Sun, Pierre Seneor, Bruno Dlubak, Birong Luo, Frederic Vosterberg, Depin Huang, Haofei Shi, Da Luo, Meihui Wang, Rodney Su Ruoff, Ben R. Conran, Clifford McAlees, Cedric Huygebaert, Stephen Brems, Ilarothy J. Attack, Wlodek Strupinski, Dirch H. Petersen, Stephen Forti, Camilla Coletti, Alexander Juvre, Kenneth B.K. Theo, Alba Centeno, Amaya Zurutuza, Pierre Leganier, Peter U Jepsen and Peter Baggild, February 17, 2021, 2D materials
DOI: 10.1088 / 2053-1583 / abdbcb





Originally reported by Source link

The featured images are, as they appear on the original report.

Continue Reading

Trending