research
-
KAIST Research Team Breaks Down Musical Instincts with AI
Music, often referred to as the universal language, is known to be a common component in all cultures. Then, could ‘musical instinct’ be something that is shared to some degree despite the extensive environmental differences amongst cultures?
On January 16, a KAIST research team led by Professor Hawoong Jung from the Department of Physics announced to have identified the principle by which musical instincts emerge from the human brain without special learning using an artificial neural network model.
Previously, many researchers have attempted to identify the similarities and differences between the music that exist in various different cultures, and tried to understand the origin of the universality. A paper published in Science in 2019 had revealed that music is produced in all ethnographically distinct cultures, and that similar forms of beats and tunes are used. Neuroscientist have also previously found out that a specific part of the human brain, namely the auditory cortex, is responsible for processing musical information.
Professor Jung’s team used an artificial neural network model to show that cognitive functions for music forms spontaneously as a result of processing auditory information received from nature, without being taught music. The research team utilized AudioSet, a large-scale collection of sound data provided by Google, and taught the artificial neural network to learn the various sounds. Interestingly, the research team discovered that certain neurons within the network model would respond selectively to music. In other words, they observed the spontaneous generation of neurons that reacted minimally to various other sounds like those of animals, nature, or machines, but showed high levels of response to various forms of music including both instrumental and vocal.
The neurons in the artificial neural network model showed similar reactive behaviours to those in the auditory cortex of a real brain. For example, artificial neurons responded less to the sound of music that was cropped into short intervals and were rearranged. This indicates that the spontaneously-generated music-selective neurons encode the temporal structure of music. This property was not limited to a specific genre of music, but emerged across 25 different genres including classic, pop, rock, jazz, and electronic.
< Figure 1. Illustration of the musicality of the brain and artificial neural network (created with DALL·E3 AI based on the paper content) >
Furthermore, suppressing the activity of the music-selective neurons was found to greatly impede the cognitive accuracy for other natural sounds. That is to say, the neural function that processes musical information helps process other sounds, and that ‘musical ability’ may be an instinct formed as a result of an evolutionary adaptation acquired to better process sounds from nature.
Professor Hawoong Jung, who advised the research, said, “The results of our study imply that evolutionary pressure has contributed to forming the universal basis for processing musical information in various cultures.” As for the significance of the research, he explained, “We look forward for this artificially built model with human-like musicality to become an original model for various applications including AI music generation, musical therapy, and for research in musical cognition.” He also commented on its limitations, adding, “This research however does not take into consideration the developmental process that follows the learning of music, and it must be noted that this is a study on the foundation of processing musical information in early development.”
< Figure 2. The artificial neural network that learned to recognize non-musical natural sounds in the cyber space distinguishes between music and non-music. >
This research, conducted by first author Dr. Gwangsu Kim of the KAIST Department of Physics (current affiliation: MIT Department of Brain and Cognitive Sciences) and Dr. Dong-Kyum Kim (current affiliation: IBS) was published in Nature Communications under the title, “Spontaneous emergence of rudimentary music detectors in deep neural networks”.
This research was supported by the National Research Foundation of Korea.
2024.01.23 View 3653 -
A KAIST Research Team Observes the Processes of Memory and Cognition in Real Time
The human brain contains approximately 86 billion neurons and 600 trillion synapses that exchange signals between the neurons to help us control the various functions of the brain including cognition, emotion, and memory. Interestingly, the number of synapses decrease with age or as a result of diseases like Alzheimer’s, and research on synapses thus attracts a lot of attention. However, limitations have existed in observing the dynamics of synapse structures in real time.
On January 9, a joint research team led by Professor Won Do Heo from the KAIST Department of Biological Sciences, Professor Hyung-Bae Kwon from Johns Hopkins School of Medicine, and Professor Sangkyu Lee from the Institute for Basic Science (IBS) revealed that they have developed the world’s first technique to allow a real-time observation of synapse formation, extinction, and alterations.
Professor Heo’s team conjugated dimerization-dependent fluorescent proteins (ddFP) to synapses in order to observe the process in which synapses create connections between neurons in real time. The team named this technique SynapShot, by combining the words ‘synapse’ and snapshot’, and successfully tracked and observed the live formation and extinction processes of synapses as well as their dynamic changes.
< Figure 1. To observe dynamically changing synapses, dimerization-dependent fluorescent protein (ddFP) was expressed to observe flourescent signals upon synapse formation as ddFP enables fluorescence detection through reversible binding to pre- and postsynaptic terminals. >
Through a joint research project, the teams led by Professor Heo and Professor Sangkyu Lee at IBS together designed a SynapShot with green and red fluorescence, and were able to easily distinguish the synapse connecting two different neurons. Additionally, by combining an optogenetic technique that can control the function of a molecule using light, the team was able to observe the changes in the synapses while simultaneously inducing certain functions of the neurons using light.
Through more joint research with the team led by Professor Hyung-Bae Kwon at the Johns Hopkins School of Medicine, Professor Heo’s team induced several situations on live mice, including visual discrimination training, exercise, and anaesthesia, and used SynapShot to observe the changes in the synapses during each situation in real time. The observations revealed that each synapse could change fairly quickly and dynamically. This was the first-ever case in which the changes in synapses were observed in a live mammal.
< Figure 2. Microscopic photos observed through changes of the flourescence of the synapse sensor (SynapShot) by cultivating the neurons of an experimental rat and expressing the SynapShot. The changes in the synapse that is created when the pre- and post-synaptic terminals come into contact and the synapse that disappears after a certain period of time are measured by the fluorescence of the SynapShot. >
Professor Heo said, “Our group developed SynapShot through a collaboration with domestic and international research teams, and have opened up the possibility for first-hand live observations of the quick and dynamic changes of synapses, which was previously difficult to do. We expect this technique to revolutionize research methodology in the neurological field, and play an important role in brightening the future of brain science.”
This research, conducted by co-first authors Seungkyu Son (Ph.D. candidate), Jinsu Lee (Ph.D. candidate) and Dr. Kanghoon Jung from Johns Hopkins, was published in the online edition of Nature Methods on January 8 under the title “Real-time visualization of structural dynamics of synapses in live cells in vivo”, and will be printed in the February volume.
< Figure 3. Simultaneous use of green-SynapShot and red-SynapShot to distinguish and observe synapses with one post-terminal and different pre-terminals. >
< Figure 4. Dimer-dependent fluorescent protein (ddFP) exists as a green fluorescent protein as well as a red fluorescent protein, and can be applied together with blue light-activated optogenetic technology. After activating Tropomyosin receptor kinase B (TrkB) by blue light using optogenetic technology, the strengthening of synaptic connections through signals of brain-derived neurotrophic factor is observed using red-SynapShot. >
< Figure 5. Micrographs showing real-time changing synapses in the visual cortex of mice trained through visual training using in vivo imaging techniques such as two-photon microscopy as well as at the cellular level. >
This research was supported by Mid-Sized Research Funds and the Singularity Project from KAIST, and by IBS.
2024.01.18 View 3736 -
KAIST Research team develops anti-icing film that only requires sunlight
A KAIST research team has developed an anti-icing and de-icing film coating technology that can apply the photothermal effect of gold nanoparticles to industrial sites without the need for heating wires, periodic spray or oil coating of anti-freeze substances, and substrate design alterations.
The group led by Professor Hyoungsoo Kim from the Department of Mechanical Engineering (Fluid & Interface Laboratory) and Professor Dong Ki Yoon from the Department of Chemistry (Soft Material Assembly Group) revealed on January 3 to have together developed an original technique that can uniformly pattern gold nanorod (GNR) particles in quadrants through simple evaporation, and have used this to develop an anti-icing and de-icing surface.
Many scientists in recent years have tried to control substrate surfaces through various coating techniques, and those involving the patterning of functional nanomaterials have gained special attention. In particular, GNR is considered a promising candidate nanomaterial for its biocompatibility, chemical stability, relatively simple synthesis, and its stable and unique property of surface plasmon resonance. To maximize the performance of GNR, it is important to achieve a high uniformity during film deposition, and a high level of rod alignment. However, achieving both criteria has thus far been a difficult challenge.
< Figure 1. Conceptual image to display Hydrodynamic mechanisms for the formation of a homogeneous quadrant cellulose nanocrystal(CNC) matrix. >
To solve this, the joint research team utilized cellulose nanocrystal (CNC), a next-generation functional nanomaterial that can easily be extracted from nature. By co-assembling GNR on CNC quadrant templates, the team could uniformly dry the film and successfully obtain a GNR film with a uniform alignment in a ring-shape. Compared to existing coffee-ring films, the highly uniform and aligned GNR film developed through this research showed enhanced plasmonic photothermal properties, and the team showed that it could carry out anti-icing and de-icing functions by simply irradiating light in the visible wavelength range.
< Figure 2. Optical and thermal performance evaluation results of gold nanorod film and demonstration of plasmonic heater for anti-icing and de-icing. >
Professor Hyoungsoo Kim said, “This technique can be applied to plastic, as well as flexible surfaces. By using it on exterior materials and films, it can generate its own heat energy, which would greatly save energy through voluntary thermal energy harvesting across various applications including cars, aircrafts, and windows in residential or commercial spaces, where frosting becomes a serious issue in the winter.” Professor Dong Ki Yoon added, “This research is significant in that we can now freely pattern the CNC-GNR composite, which was previously difficult to create into films, over a large area. We can utilize this as an anti-icing material, and if we were to take advantage of the plasmonic properties of gold, we can also use it like stained-glass to decorate glass surfaces.”
This research was conducted by Ph.D. candidate Jeongsu Pyeon from the Department of Mechanical Engineering, and his co-first author Dr. Soon Mo Park (a KAIST graduate, currently a post-doctoral associate at Cornell University), and was pushed in the online volume of Nature Communication on December 8, 2023 under the title “Plasmonic Metasurfaces of Cellulose Nanocrystal Matrices with Quadrants of Aligned Gold Nanorods for Photothermal Anti-Icing." Recognized for its achievement, the research was also selected as an editor’s highlight for the journals Materials Science and Chemistry, and Inorganic and Physical Chemistry.
This research was supported by the Individual Basic Mid-Sized Research Fund from the National Research Foundation of Korea and the Center for Multiscale Chiral Architectures.
2024.01.16 View 6391 -
KAIST develops an artificial muscle device that produces force 34 times its weight
- Professor IlKwon Oh’s research team in KAIST’s Department of Mechanical Engineering developed a soft fluidic switch using an ionic polymer artificial muscle that runs with ultra-low power to lift objects 34 times greater than its weight.
- Its light weight and small size make it applicable to various industrial fields such as soft electronics, smart textiles, and biomedical devices by controlling fluid flow with high precision, even in narrow spaces.
Soft robots, medical devices, and wearable devices have permeated our daily lives. KAIST researchers have developed a fluid switch using ionic polymer artificial muscles that operates at ultra-low power and produces a force 34 times greater than its weight. Fluid switches control fluid flow, causing the fluid to flow in a specific direction to invoke various movements.
KAIST (President Kwang-Hyung Lee) announced on the 4th of January that a research team under Professor IlKwon Oh from the Department of Mechanical Engineering has developed a soft fluidic switch that operates at ultra-low voltage and can be used in narrow spaces.
Artificial muscles imitate human muscles and provide flexible and natural movements compared to traditional motors, making them one of the basic elements used in soft robots, medical devices, and wearable devices. These artificial muscles create movements in response to external stimuli such as electricity, air pressure, and temperature changes, and in order to utilize artificial muscles, it is important to control these movements precisely.
Switches based on existing motors were difficult to use within limited spaces due to their rigidity and large size. In order to address these issues, the research team developed an electro-ionic soft actuator that can control fluid flow while producing large amounts of force, even in a narrow pipe, and used it as a soft fluidic switch.
< Figure 1. The separation of fluid droplets using a soft fluid switch at ultra-low voltage. >
The ionic polymer artificial muscle developed by the research team is composed of metal electrodes and ionic polymers, and it generates force and movement in response to electricity. A polysulfonated covalent organic framework (pS-COF) made by combining organic molecules on the surface of the artificial muscle electrode was used to generate an impressive amount of force relative to its weight with ultra-low power (~0.01V).
As a result, the artificial muscle, which was manufactured to be as thin as a hair with a thickness of 180 µm, produced a force more than 34 times greater than its light weight of 10 mg to initiate smooth movement. Through this, the research team was able to precisely control the direction of fluid flow with low power.
< Figure 2. The synthesis and use of pS-COF as a common electrode-electrolyte host for electroactive soft fluid switches. A) The synthesis schematic of pS-COF. B) The schematic diagram of the operating principle of the electrochemical soft switch. C) The schematic diagram of using a pS-COF-based electrochemical soft switch to control fluid flow in dynamic operation. >
Professor IlKwon Oh, who led this research, said, “The electrochemical soft fluidic switch that operate at ultra-low power can open up many possibilities in the fields of soft robots, soft electronics, and microfluidics based on fluid control.” He added, “From smart fibers to biomedical devices, this technology has the potential to be immediately put to use in a variety of industrial settings as it can be easily applied to ultra-small electronic systems in our daily lives.”
The results of this study, in which Dr. Manmatha Mahato, a research professor in the Department of Mechanical Engineering at KAIST, participated as the first author, were published in the international academic journal Science Advances on December 13, 2023. (Paper title: Polysulfonated Covalent Organic Framework as Active Electrode Host for Mobile Cation Guests in Electrochemical Soft Actuator)
This research was conducted with support from the National Research Foundation of Korea's Leader Scientist Support Project (Creative Research Group) and Future Convergence Pioneer Project.
* Paper DOI: https://www.science.org/doi/abs/10.1126/sciadv.adk9752
2024.01.11 View 6494 -
A KAIST Research Team Develops High-Performance Stretchable Solar Cells
With the market for wearable electric devices growing rapidly, stretchable solar cells that can function under strain have received considerable attention as an energy source. To build such solar cells, it is necessary that their photoactive layer, which converts light into electricity, shows high electrical performance while possessing mechanical elasticity. However, satisfying both of these two requirements is challenging, making stretchable solar cells difficult to develop.
On December 26, a KAIST research team from the Department of Chemical and Biomolecular Engineering (CBE) led by Professor Bumjoon Kim announced the development of a new conductive polymer material that achieved both high electrical performance and elasticity while introducing the world’s highest-performing stretchable organic solar cell.
Organic solar cells are devices whose photoactive layer, which is responsible for the conversion of light into electricity, is composed of organic materials. Compared to existing non-organic material-based solar cells, they are lighter and flexible, making them highly applicable for wearable electrical devices. Solar cells as an energy source are particularly important for building electrical devices, but high-efficiency solar cells often lack flexibility, and their application in wearable devices have therefore been limited to this point.
The team led by Professor Kim conjugated a highly stretchable polymer to an electrically conductive polymer with excellent electrical properties through chemical bonding, and developed a new conductive polymer with both electrical conductivity and mechanical stretchability. This polymer meets the highest reported level of photovoltaic conversion efficiency (19%) using organic solar cells, while also showing 10 times the stretchability of existing devices. The team thereby built the world’s highest performing stretchable solar cell that can be stretched up to 40% during operation, and demonstrated its applicability for wearable devices.
< Figure 1. Chemical structure of the newly developed conductive polymer and performance of stretchable organic solar cells using the material. >
Professor Kim said, “Through this research, we not only developed the world’s best performing stretchable organic solar cell, but it is also significant that we developed a new polymer that can be applicable as a base material for various electronic devices that needs to be malleable and/or elastic.”
< Figure 2. Photovoltaic efficiency and mechanical stretchability of newly developed polymers compared to existing polymers. >
This research, conducted by KAIST researchers Jin-Woo Lee and Heung-Goo Lee as first co-authors in cooperation with teams led by Professor Taek-Soo Kim from the Department of Mechanical Engineering and Professor Sheng Li from the Department of CBE, was published in Joule on December 1 (Paper Title: Rigid and Soft Block-Copolymerized Conjugated Polymers Enable High-Performance Intrinsically-Stretchable Organic Solar Cells).
This research was supported by the National Research Foundation of Korea.
2024.01.04 View 5110 -
KAIST presents strategies for environmentally friendly and sustainable polyamides production
- Provides current research trends in bio-based polyamide production
- Research on bio-based polyamides production gains importance for achieving a carbon-neutral society
Global industries focused on carbon neutrality, under the slogan "Net-Zero," are gaining increasing attention. In particular, research on microbial production of polymers, replacing traditional chemical methods with biological approaches, is actively progressing.
Polyamides, represented by nylon, are linear polymers widely used in various industries such as automotive, electronics, textiles, and medical fields. They possess beneficial properties such as high tensile strength, electrical insulation, heat resistance, wear resistance, and biocompatibility. Since the commercialization of nylon in 1938, approximately 7 million tons of polyamides are produced worldwide annually. Considering their broad applications and significance, producing polyamides through bio-based methods holds considerable environmental and industrial importance.
KAIST (President Kwang-Hyung Lee) announced that a research team led by Distinguished Professor Sang Yup Lee, including Dr. Jong An Lee and doctoral candidate Ji Yeon Kim from the Department of Chemical and Biomolecular Engineering, published a paper titled "Current Advancements in Bio-Based Production of Polyamides”. The paper was featured on the cover of the monthly issue of "Trends in Chemistry” by Cell Press.
As part of climate change response technologies, bio-refineries involve using biotechnological and chemical methods to produce industrially important chemicals and biofuels from renewable biomass without relying on fossil resources. Notably, systems metabolic engineering, pioneered by KAIST's Distinguished Professor Sang Yup Lee, is a research field that effectively manipulates microbial metabolic pathways to produce valuable chemicals, forming the core technology for bio-refineries. The research team has successfully developed high-performance strains producing a variety of compounds, including succinic acid, biodegradable plastics, biofuels, and natural products, using systems metabolic engineering tools and strategies.
The research team predicted that if bio-based polyamide production technology, which is widely used in the production of clothing and textiles, becomes widespread, it will attract attention as a future technology that can respond to the climate crisis due to its environment-friendly production technology. In this study, the research team comprehensively reviewed the bio-based polyamide production strategies. They provided insights into the advancements in polyamide monomer production using metabolically engineered microorganisms and highlighted the recent trends in bio-based polyamide advancements utilizing these monomers. Additionally, they reviewed the strategies for synthesizing bio-based polyamides through chemical conversion of natural oils and discussed the biodegradability and recycling of the polyamides. Furthermore, the paper presented the future direction in which metabolic engineering can be applied for the bio-based polyamide production, contributing to environmentally friendly and sustainable society.
Ji Yeon Kim, the co-first author of this paper from KAIST, stated "The importance of utilizing systems metabolic engineering tools and strategies for bio-based polyamides production is becoming increasingly prominent in achieving carbon neutrality."
Professor Sang Yup Lee emphasized, "Amid growing concerns about climate change, the significance of environmentally friendly and sustainable industrial development is greater than ever. Systems metabolic engineering is expected to have a significant impact not only on the chemical industry but also in various fields."
< [Figure 1] A schematic overview of the overall process for polyamides production >
This paper by Dr. Jong An Lee, PhD student Ji Yeon Kim, Dr. Jung Ho Ahn, and Master Yeah-Ji Ahn from the Department of Chemical and Biomolecular Engineering at KAIST was published in the December issue of 'Trends in Chemistry', an authoritative review journal in the field of chemistry published by Cell. It was published on December 7 as the cover paper and featured review.
※ Paper title: Current advancements in the bio-based production of polyamides
※ Author information: Jong An Lee, Ji Yeon Kim, Jung Ho Ahn, Yeah-Ji Ahn, and Sang Yup Lee
This research was conducted with the support from the development of platform technologies of microbial cell factories for the next-generation biorefineries project and C1 gas refinery program by Korean Ministry of Science and ICT.
< [Figure 2] Cover paper of the December issue of Trends in Chemistry >
2023.12.21 View 3879 -
A KAIST Team Develops Selective Transfer Printing Technology for MicroLEDs
- A KAIST research team led by Professor Keon Jae Lee demonstrates the transfer printing of a large number of micro-sized inorganic semiconductor chips via the selective modulation of micro-vacuum force.
MicroLEDs are a light source for next-generation displays that utilize inorganic LED chips with a size of less than 100 μm. MicroLEDs have attracted a great deal of attention due to their superior electrical/optical properties, reliability, and stability compared to conventional displays such as LCD, OLED, and QD. To commercialize microLEDs, transfer printing technology is essential for rearranging microLED dies from a growth substrate onto the final substrate with a desired layout and precise alignment. However, previous transfer methods still have many challenges such as the need for additional adhesives, misalignment, low transfer yield, and chip damage.
Professor Lee’s research team has developed a micro-vacuum assisted selective transfer printing (µVAST) technology to transfer a large number of microLED chips by adjusting the micro-vacuum suction force.
The key technology relies on a laser-induced etching (LIE) method for forming 20 μm-sized micro-hole arrays with a high aspect ratio on glass substrates at fabrication speed of up to 7,000 holes per second. The LIE-drilled glass is connected to the vacuum channels, controlling the micro-vacuum force at desired hole arrays to selectively pick up and release the microLEDs. The micro-vacuum assisted transfer printing accomplishes a higher adhesion switchability compared to previous transfer methods, enabling the assembly of micro-sized semiconductors with various heterogeneous materials, sizes, shapes, and thicknesses onto arbitrary substrates with high transfer yields.
< Figure 01. Concept of micro-vacuum assisted selective transfer printing (μVAST). >
Professor Keon Jae Lee said, “The micro-vacuum assisted transfer provides an interesting tool for large-scale, selective integration of microscale high-performance inorganic semiconductors. Currently, we are investigating the transfer printing of commercial microLED chips with an ejector system for commercializing next-generation displays (Large screen TVs, flexible/stretchable devices) and wearable phototherapy patches.”
This result titled “Universal selective transfer printing via micro-vacuum force” was published in Nature Communications on November 26th, 2023. (DOI: 10.1038/S41467-023-43342-8)
< Figure 02. Universal transfer printing of thin-film semiconductors via μVAST. >
< Figure 03. Flexible devices fabricated by μVAST. >
Title: Entire process including LIE and µVAST
Vimeo link: https://vimeo.com/894430416?share=copy
2023.12.19 View 2671 -
KAIST introduces eco-friendly technologies for plastic production and biodegradation
- A research team under Distinguished Professor Sang Yup Lee of the Department of Chemical and Biomolecular Engineering published a paper in Nature Microbiology on the overview and trends of plastic production and degradation technology using microorganisms.
- Eco-friendly and sustainable plastic production and degradation technology using microorganisms as a core technology to achieve a plastic circular economy was presented.
Plastic is one of the important materials in modern society, with approximately 460 million tons produced annually and with expected production reaching approximately 1.23 billion tons in 2060. However, since 1950, plastic waste totaling more than 6.3 billion tons has been generated, and it is believed that more than 140 million tons of plastic waste has accumulated in the aquatic environment. Recently, the seriousness of microplastic pollution has emerged, not only posing a risk to the marine ecosystem and human health, but also worsening global warming by inhibiting the activity of marine plankton, which play an important role in lowering the Earth's carbon dioxide concentration.
KAIST President Kwang-Hyung Lee announced on December 11 that a research team under Distinguished Professor Sang Yup Lee of the Department of Chemical and Biomolecular Engineering had published a paper titled 'Sustainable production and degradation of plastics using microbes', which covers the latest technologies for producing plastics and processing waste plastics in an eco-friendly manner using microorganisms.
As the international community moves to solve this plastic problem, various efforts are being made, including 175 countries participating to conclude a legally binding agreement with the goal of ending plastic pollution by 2024. Various technologies are being developed for sustainable plastic production and processing, and among them, biotechnology using microorganisms is attracting attention.
Microorganisms have the ability to naturally produce or decompose certain compounds, and this ability is maximized through biotechnologies such as metabolic engineering and enzyme engineering to produce plastics from renewable biomass resources instead of fossil raw materials and to decompose waste plastics.
Accordingly, the research team comprehensively analyzed the latest microorganism-based technologies for the sustainable production and decomposition of plastics and presented how they actually contribute to solving the plastic problem. Based on this, they presented limitations, prospects, and research directions of the technologies for achieving a circular economy for plastics.
Microorganism-based technologies for various plastics range from widely used synthetic plastics such as polyethylene (PE) to promising bioplastics such as natural polymers derived from microorganisms (polyhydroxyalkanoate (PHA)) that are completely biodegradable in the natural environment and do not pose a risk of microplastic generation. Commercialization statuses and latest technologies were also discussed. In addition, the technology to decompose these plastics using microorganisms and their enzymes and the upcycling technology to convert them into other useful compounds after decomposition were introduced, highlighting the competitiveness and potential of technology using microorganisms.
First author So Young Choi, a research assistant professor in the Department of Chemical and Biomolecular Engineering at KAIST, said, “In the future, we will be able to easily find eco-friendly plastics made using microorganisms all around us,” and corresponding author Distinguished Professor Sang Yup Lee said, “Plastic can be made more sustainable. It is important to use plastics responsibly to protect the environment and simultaneously achieve economic and social development through the new plastics industry, and we look forward to the improved performance of microbial metabolic engineering technology.”
This paper was published on November 30th in the online edition of Nature Microbiology.
※ Paper Title : Sustainable production and degradation of plastics using microbes
Authors: So Young Choi, Youngjoon Lee, Hye Eun Yu, In Jin Cho, Minju Kang & Sang Yup Lee
2023.12.11 View 3973 -
North Korea and Beyond: AI-Powered Satellite Analysis Reveals the Unseen Economic Landscape of Underdeveloped Nations
- A joint research team in computer science, economics, and geography has developed an artificial intelligence (AI) technology to measure grid-level economic development within six-square-kilometer regions.
- This AI technology is applicable in regions with limited statistical data (e.g., North Korea), supporting international efforts to propose policies for economic growth and poverty reduction in underdeveloped countries.
- The research team plans to make this technology freely available for use to contribute to the United Nations' Sustainable Development Goals (SDGs).
The United Nations reports that more than 700 million people are in extreme poverty, earning less than two dollars a day. However, an accurate assessment of poverty remains a global challenge. For example, 53 countries have not conducted agricultural surveys in the past 15 years, and 17 countries have not published a population census. To fill this data gap, new technologies are being explored to estimate poverty using alternative sources such as street views, aerial photos, and satellite images.
The paper published in Nature Communications demonstrates how artificial intelligence (AI) can help analyze economic conditions from daytime satellite imagery. This new technology can even apply to the least developed countries - such as North Korea - that do not have reliable statistical data for typical machine learning training.
The researchers used Sentinel-2 satellite images from the European Space Agency (ESA) that are publicly available. They split these images into small six-square-kilometer grids. At this zoom level, visual information such as buildings, roads, and greenery can be used to quantify economic indicators. As a result, the team obtained the first ever fine-grained economic map of regions like North Korea. The same algorithm was applied to other underdeveloped countries in Asia: North Korea, Nepal, Laos, Myanmar, Bangladesh, and Cambodia (see Image 1).
The key feature of their research model is the "human-machine collaborative approach," which lets researchers combine human input with AI predictions for areas with scarce data. In this research, ten human experts compared satellite images and judged the economic conditions in the area, with the AI learning from this human data and giving economic scores to each image. The results showed that the Human-AI collaborative approach outperformed machine-only learning algorithms.
< Image 1. Nightlight satellite images of North Korea (Top-left: Background photo provided by NASA's Earth Observatory). South Korea appears brightly lit compared to North Korea, which is mostly dark except for Pyongyang. In contrast, the model developed by the research team uses daytime satellite imagery to predict more detailed economic predictions for North Korea (top-right) and five Asian countries (Bottom: Background photo from Google Earth). >
The research was led by an interdisciplinary team of computer scientists, economists, and a geographer from KAIST & IBS (Donghyun Ahn, Meeyoung Cha, Jihee Kim), Sogang University (Hyunjoo Yang), HKUST (Sangyoon Park), and NUS (Jeasurk Yang). Dr Charles Axelsson, Associate Editor at Nature Communications, handled this paper during the peer review process at the journal.
The research team found that the scores showed a strong correlation with traditional socio-economic metrics such as population density, employment, and number of businesses. This demonstrates the wide applicability and scalability of the approach, particularly in data-scarce countries. Furthermore, the model's strength lies in its ability to detect annual changes in economic conditions at a more detailed geospatial level without using any survey data (see Image 2).
< Image 2. Differences in satellite imagery and economic scores in North Korea between 2016 and 2019. Significant development was found in the Wonsan Kalma area (top), one of the tourist development zones, but no changes were observed in the Wiwon Industrial Development Zone (bottom). (Background photo: Sentinel-2 satellite imagery provided by the European Space Agency (ESA)). >
This model would be especially valuable for rapidly monitoring the progress of Sustainable Development Goals such as reducing poverty and promoting more equitable and sustainable growth on an international scale. The model can also be adapted to measure various social and environmental indicators. For example, it can be trained to identify regions with high vulnerability to climate change and disasters to provide timely guidance on disaster relief efforts.
As an example, the researchers explored how North Korea changed before and after the United Nations sanctions against the country. By applying the model to satellite images of North Korea both in 2016 and in 2019, the researchers discovered three key trends in the country's economic development between 2016 and 2019. First, economic growth in North Korea became more concentrated in Pyongyang and major cities, exacerbating the urban-rural divide. Second, satellite imagery revealed significant changes in areas designated for tourism and economic development, such as new building construction and other meaningful alterations. Third, traditional industrial and export development zones showed relatively minor changes.
Meeyoung Cha, a data scientist in the team explained, "This is an important interdisciplinary effort to address global challenges like poverty. We plan to apply our AI algorithm to other international issues, such as monitoring carbon emissions, disaster damage detection, and the impact of climate change."
An economist on the research team, Jihee Kim, commented that this approach would enable detailed examinations of economic conditions in the developing world at a low cost, reducing data disparities between developed and developing nations. She further emphasized that this is most essential because many public policies require economic measurements to achieve their goals, whether they are for growth, equality, or sustainability.
The research team has made the source code publicly available via GitHub and plans to continue improving the technology, applying it to new satellite images updated annually. The results of this study, with Ph.D. candidate Donghyun Ahn at KAIST and Ph.D. candidate Jeasurk Yang at NUS as joint first authors, were published in Nature Communications under the title "A human-machine collaborative approach measures economic development using satellite imagery."
< Photos of the main authors. 1. Donghyun Ahn, PhD candidate at KAIST School of Computing 2. Jeasurk Yang, PhD candidate at the Department of Geography of National University of Singapore 3. Meeyoung Cha, Professor of KAIST School of Computing and CI at IBS 4. Jihee Kim, Professor of KAIST School of Business and Technology Management 5. Sangyoon Park, Professor of the Division of Social Science at Hong Kong University of Science and Technology 6. Hyunjoo Yang, Professor of the Department of Economics at Sogang University >
2023.12.07 View 4684 -
The Relentless Rain: East Asia's Recent Floods and What Lies Beneath
In just a month's time, East Asia witnessed torrential downpours that would usually span an entire season. Japan, battered by three times its usual monthly rainfall, faced landslides and flooding that claimed over 200 lives. Meanwhile, South Korea grappled with its longest monsoon in seven years, leaving more than 40 individuals dead or missing. But these events, as harrowing as they sound, are more than just weather anomalies. They're telltale signs, symptoms of a larger malaise that has gripped our planet.
Diving deep into these rain-soaked mysteries, a recently published paper in the journal Science Advances offers a fresh perspective. Led by a research team at the Korea Advanced Institute of Science and Technology (KAIST), Korea, the research unpacks the influence of human-induced climate changes on the East Asia Summer Monsoon frontal system.
Heavy summer rain has a significant impact on agriculture and industry, and can be said to be one of the greatest threats to human society by causing disasters such as floods and landslides, affecting the local ecosystem. It has been reported from all over the world that the intensity of summer heavy rain has changed over the past few decades. However, summer rain in East Asia is caused by various forms such as typhoons, extratropical cyclones, and fronts, and efforts to study heavy frontal rain, which account for more than 40% of summer rainfall, is still insufficient. In addition, because heavy rain is also influenced by natural fluctuations or coincidences in the climate system, it is not yet known to what extent warming due to human activities affects the intensity of heavy frontal precipitation.
An international joint research team consisting of eight institutions from Korea, the United States, and Japan, including KAIST, Tokyo University, Tokyo Institute of Technology, Chonnam National University, GIST, and Utah State University, confirmed the intensity of heavy rain caused by the weather fronts in East Asia using observation data for the past 60 years and found that the coast of southeastern China. It was found that the intensity of heavy rain increased by about 17% across the Korean Peninsula and Japan. To investigate the cause of these changes, the research team used the Earth Metaverse experiment, which simulated Earth with and without greenhouse gas emissions due to human activities, and found that heavy rain intensity was strengthened by about 6% due to greenhouse gas emissions, and the changes discovered were has succeeded for the first time in the world in showing that warming cannot be explained without the effects of human activities.
< Figure 1. (Left) Observed difference in frontal rainfall intensity (FRI) from the later (1991–2015) to the earlier periods (1958–1982) (Right) Visualization of the impact of anthropogenic warming on the intensity of heavy frontal rain analyzed using the Earth Metaverse experiment. >
"It's not just about connecting the dots," said Moon, the first author of the paper, "it's about seeing the larger pattern. Our data analysis reveals a clear and intensified trend in East Asia's frontal rainfall, one that's intertwined with human actions and increasingly harmful for lives and property."
One of the intriguing finds from the study is the mechanics behind this intensification. The team found increased moisture transport due to a warmer climate, which, when coupled with the strengthening of a gigantic weather system called the West North Pacific Subtropical High, results in enhanced frontal rainfall. It’s akin to the climate dialing up the volume on rain events. As the atmosphere warms, it holds more moisture, leading to heavier downpours when conditions are right.
Nobuyuki Utsumi, another voice from the team, makes the science accessible for all, saying, "Monsoon rain isn't just rain anymore. The frequency, the intensity, it's changing. And our actions, our carbon footprint, are casting a larger shadow than we anticipated."
While the devastating news of floods fills headlines, Professor Simon Wang of Utah State University, reminds us of the underlying importance of their study. "It's like reading a detective novel. To solve the mystery of these floods, one has to trace them back to their roots. This work hints at a future where such intense rain events aren't the exception but might become an everyday reality."
Hyungjun Kim, the principal investigator of the team throws in a note of caution, "Understanding this is just the first step. Predicting and preparing for these extremes is the real challenge. Every amplified rainfall event is a message from the future, urging us to adapt." So far, predicting rainfall intensity and locations remains a challenging task for even the most sophisticated weather models.
< Figure 2. Comparison of rates of change in Anthropocene fingerprints. The horizontal axis shows the long-term change slope of the Anthropocene fingerprint signal (1958 to 2015). Shows the probability distribution of slopes extracted from the non-warming experiment (blue) and the warming experiment (red). The vertical solid lines are the slope of the Anthropocene fingerprint signal extracted from observational data. >
The researchers say, “Facing climate change, the narrative of this new study is more than mere numbers and data. It's a story of our planet, our actions, and the rain-soaked repercussions we're beginning to face. As we mop up the aftermath of another flood, research like Moon's beckons us to look deeper, understand better, and act wiser.”
< Figure 3. Comparison of water vapor convergence and rate of change of the western North Pacific high pressure. Shows the gradient of change in water vapor convergence (horizontal axis) and the Northwestern Pacific-East Asia pressure gradient (vertical axis) extracted from warming (red) and non-warming (blue) experiments. Shows the distribution of slope changes of the two indices during the period 1958 to 1982 (P1) and 1991 to 2015 (P2). >
The results of this study were published on November 24 in Science Advances. (Paper title: Anthropogenic warming induced intensification of summer monsoon frontal precipitation over East Asia)
This research was conducted with support from the National Research Foundation of Korea's Overseas Scientist Attraction Project (BP+) and the Anthropocene Research Center.
2023.12.05 View 3182 -
KAIST-UCSD researchers build an enzyme discovering AI
- A joint research team led by Distinguished Professor Sang Yup Lee of the Department of Chemical and Biomolecular Engineering and Bernhard Palsson of UCSD developed ‘DeepECtransformer’, an artificial intelligence that can predict Enzyme Commission (EC) number of proteins.
- The AI is tasked to discover new enzymes that have not been discovered yet, which would allow prediction for a total of 5,360 types of Enzyme Commission (EC) numbers
- It is expected to be used in the development of microbial cell factories that produce environmentally friendly chemicals as a core technology for analyzing the metabolic network of a genome.
While E. coli is one of the most studied organisms, the function of 30% of proteins that make up E. coli has not yet been clearly revealed. For this, an artificial intelligence was used to discover 464 types of enzymes from the proteins that were unknown, and the researchers went on to verify the predictions of 3 types of proteins were successfully identified through in vitro enzyme assay.
KAIST (President Kwang-Hyung Lee) announced on the 24th that a joint research team comprised of Gi Bae Kim, Ji Yeon Kim, Dr. Jong An Lee and Distinguished Professor Sang Yup Lee of the Department of Chemical and Biomolecular Engineering at KAIST, and Dr. Charles J. Norsigian and Professor Bernhard O. Palsson of the Department of Bioengineering at UCSD has developed DeepECtransformer, an artificial intelligence that can predict the enzyme functions from the protein sequence, and has established a prediction system by utilizing the AI to quickly and accurately identify the enzyme function.
Enzymes are proteins that catalyze biological reactions, and identifying the function of each enzyme is essential to understanding the various chemical reactions that exist in living organisms and the metabolic characteristics of those organisms. Enzyme Commission (EC) number is an enzyme function classification system designed by the International Union of Biochemistry and Molecular Biology, and in order to understand the metabolic characteristics of various organisms, it is necessary to develop a technology that can quickly analyze enzymes and EC numbers of the enzymes present in the genome.
Various methodologies based on deep learning have been developed to analyze the features of biological sequences, including protein function prediction, but most of them have a problem of a black box, where the inference process of AI cannot be interpreted. Various prediction systems that utilize AI for enzyme function prediction have also been reported, but they do not solve this black box problem, or cannot interpret the reasoning process in fine-grained level (e.g., the level of amino acid residues in the enzyme sequence).
The joint team developed DeepECtransformer, an AI that utilizes deep learning and a protein homology analysis module to predict the enzyme function of a given protein sequence. To better understand the features of protein sequences, the transformer architecture, which is commonly used in natural language processing, was additionally used to extract important features about enzyme functions in the context of the entire protein sequence, which enabled the team to accurately predict the EC number of the enzyme. The developed DeepECtransformer can predict a total of 5360 EC numbers.
The joint team further analyzed the transformer architecture to understand the inference process of DeepECtransformer, and found that in the inference process, the AI utilizes information on catalytic active sites and/or the cofactor binding sites which are important for enzyme function. By analyzing the black box of DeepECtransformer, it was confirmed that the AI was able to identify the features that are important for enzyme function on its own during the learning process.
"By utilizing the prediction system we developed, we were able to predict the functions of enzymes that had not yet been identified and verify them experimentally," said Gi Bae Kim, the first author of the paper. "By using DeepECtransformer to identify previously unknown enzymes in living organisms, we will be able to more accurately analyze various facets involved in the metabolic processes of organisms, such as the enzymes needed to biosynthesize various useful compounds or the enzymes needed to biodegrade plastics." he added.
"DeepECtransformer, which quickly and accurately predicts enzyme functions, is a key technology in functional genomics, enabling us to analyze the function of entire enzymes at the systems level," said Professor Sang Yup Lee. He added, “We will be able to use it to develop eco-friendly microbial factories based on comprehensive genome-scale metabolic models, potentially minimizing missing information of metabolism.”
The joint team’s work on DeepECtransformer is described in the paper titled "Functional annotation of enzyme-encoding genes using deep learning with transformer layers" written by Gi Bae Kim, Professor Sang Yup Lee of the Department of Chemical and Biomolecular Engineering of KAIST and their colleagues. The paper was published via peer-review on the 14th of November on “Nature Communications”.
This research was conducted with the support by “the Development of next-generation biorefinery platform technologies for leading bio-based chemicals industry project (2022M3J5A1056072)” and by “Development of platform technologies of microbial cell factories for the next-generation biorefineries project (2022M3J5A1056117)” from National Research Foundation supported by the Korean Ministry of Science and ICT (Project Leader: Distinguished Professor Sang Yup Lee, KAIST).
< Figure 1. The structure of DeepECtransformer's artificial neural network >
2023.11.24 View 3201 -
An intravenous needle that irreversibly softens via body temperature on insertion
- A joint research team at KAIST developed an intravenous (IV) needle that softens upon insertion, minimizing risk of damage to blood vessels and tissues.
- Once used, it remains soft even at room temperature, preventing accidental needle stick injuries and unethical multiple use of needle.
- A thin-film temperature sensor can be embedded with this needle, enabling real-time monitoring of the patient's core body temperature, or detection of unintended fluid leakage, during IV medication.
Intravenous (IV) injection is a method commonly used in patient’s treatment worldwide as it induces rapid effects and allows treatment through continuous administration of medication by directly injecting drugs into the blood vessel. However, medical IV needles, made of hard materials such as stainless steel or plastic which do not mechanically match the soft biological tissues of the body, can cause critical problems in healthcare settings, starting from minor tissue damages in the injection sites to serious inflammations.
The structure and dexterity of rigid medical IV devices also enable unethical reuse of needles for reduction of injection costs, leading to transmission of deadly blood-borne disease infections such as human immunodeficiency virus (HIV) and hepatitis B/C viruses. Furthermore, unintended needlestick injuries are frequently occurring in medical settings worldwide, that are viable sources of such infections, with IV needles having the greatest susceptibility of being the medium of transmissible diseases. For these reasons, the World Health Organization (WHO) in 2015 launched a policy on safe injection practices to encourage the development and use of “smart” syringes that have features to prevent re-use, after a tremendous increase in the number of deadly infectious disease worldwide due to medical-sharps related issues.
KAIST announced on the 13th that Professor Jae-Woong Jeong and his research team of its School of Electrical Engineering succeeded in developing the Phase-Convertible, Adapting and non-REusable (P-CARE) needle with variable stiffness that can improve patient health and ensure the safety of medical staff through convergent joint research with another team led by Professor Won-Il Jeong of the Graduate School of Medical Sciences.
The new technology is expected to allow patients to move without worrying about pain at the injection site as it reduces the risk of damage to the wall of the blood vessel as patients receive IV medication. This is possible with the needle’s stiffness-tunable characteristics which will make it soft and flexible upon insertion into the body due to increased temperature, adapting to the movement of thin-walled vein. It is also expected to prevent blood-borne disease infections caused by accidental needlestick injuries or unethical re-using of syringes as the deformed needle remains perpetually soft even after it is retracted from the injection site.
The results of this research, in which Karen-Christian Agno, a doctoral researcher of the School of Electrical Engineering at and Dr. Keungmo Yang of the Graduate School of Medical Sciences participated as co-first authors, was published in Nature Biomedical Engineering on October 30. (Paper title: A temperature-responsive intravenous needle that irreversibly softens on insertion)
< Figure 1. Disposable variable stiffness intravenous needle. (a) Conceptual illustration of the key features of the P-CARE needle whose mechanical properties can be changed by body temperature, (b) Photograph of commonly used IV access devices and the P-CARE needle, (c) Performance of common IV access devices and the P-CARE needle >
“We’ve developed this special needle using advanced materials and micro/nano engineering techniques, and it can solve many global problems related to conventional medical needles used in healthcare worldwide”, said Jae-Woong Jeong, Ph.D., an associate professor of Electrical Engineering at KAIST and a lead senior author of the study.
The softening IV needle created by the research team is made up of liquid metal gallium that forms the hollow, mechanical needle frame encapsulated within an ultra-soft silicone material. In its solid state, gallium has sufficient hardness that enables puncturing of soft biological tissues. However, gallium melts when it is exposed to body temperature upon insertion, and changes it into a soft state like the surrounding tissue, enabling stable delivery of the drug without damaging blood vessels. Once used, a needle remains soft even at room temperature due to the supercooling phenomenon of gallium, fundamentally preventing needlestick accidents and reuse problems.
Biocompatibility of the softening IV needle was validated through in vivo studies in mice. The studies showed that implanted needles caused significantly less inflammation relative to the standard IV access devices of similar size made of metal needles or plastic catheters. The study also confirmed the new needle was able to deliver medications as reliably as commercial injection needles.
< Photo 1. Photo of the P-CARE needle that softens with body temperature. >
Researchers also showed possibility of integrating a customized ultra-thin temperature sensor with the softening IV needle to measure the on-site temperature which can further enhance patient’s well-being. The single assembly of sensor-needle device can be used to monitor the core body temperature, or even detect if there is a fluid leakage on-site during indwelling use, eliminating the need for additional medical tools or procedures to provide the patients with better health care services.
The researchers believe that this transformative IV needle can open new opportunities for wide range of applications particularly in clinical setups, in terms of redesigning other medical needles and sharp medical tools to reduce muscle tissue injury during indwelling use. The softening IV needle may become even more valuable in the present times as there is an estimated 16 billion medical injections administered annually in a global scale, yet not all needles are disposed of properly, based on a 2018 WHO report.
< Figure 2. Biocompatibility test for P-CARE needle: Images of H&E stained histology (the area inside the dashed box on the left is provided in an expanded view in the right), TUNEL staining (green), DAPI staining of nuclei (blue) and co-staining (TUNEL and DAPI) of muscle tissue from different organs. >
< Figure 3. Conceptual images of potential utilization for temperature monitoring function of P-CARE needle integrated with a temperature sensor. >
(a) Schematic diagram of injecting a drug through intravenous injection into the abdomen of a laboratory mouse (b) Change of body temperature upon injection of drug (c) Conceptual illustration of normal intravenous drug injection (top) and fluid leakage (bottom) (d) Comparison of body temperature during normal drug injection and fluid leakage: when the fluid leak occur due to incorrect insertion, a sudden drop of temperature is detected.
This work was supported by grants from the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT.
2023.11.13 View 5875