AI model may allow doctors to detect cancer from DNA

Using AI to detect and diagnose cancer in patients may soon allow for earlier treatment, say investigators at Cambridge University and Imperial College London.

The key is in our DNA

Genetic information is encoded in DNA by patterns of the four bases (A, T, G and C) that make up its structure. However, environmental changes outside the cell can cause some DNA bases to be modified by adding a methyl group (this process is called “DNA methylation”).

Each individual cell possesses millions of these DNA methylation marks. Researchers have observed changes to these marks in early cancer development and they could assist in early diagnosis of cancer. Currently, It’s difficult to examine which bases in DNA are methylated in cancers and to what extent, compared to healthy tissue.

So investigators trained an AI model, using a combination of machine and deep learning, to look at the DNA methylation patterns. It identified 13 different cancer types (including breast, liver, lung, and prostate cancers) from non-cancerous tissue with 98.2% accuracy. They found that the model reinforces and enhances understanding of the underlying processes contributing to cancer.

Additional training and testing

However, they say this model only relies on tissue samples (not DNA fragments in the blood), so it would need additional training and testing on a more diverse collection of biopsy samples to be ready for clinical use.

Identifying these unusual methylation patterns may allow healthcare providers to detect cancer early. “Computational methods such as this model, through better training on more varied data and rigorous testing in the clinic, will eventually help doctors with early detection and screening of cancers,” said the paper’s lead author, Shamith Samarajiwa, in a statement.

Citation: Newsham, I., Sendera, M., Jammula, S. G., & Samarajiwa, S. A. (2024). Early detection and diagnosis of cancer with interpretable machine learning to uncover cancer-specific DNA methylation patterns. Biology Methods and Protocols, 9(1). https://doi.org/10.1093/biomethods/bpae028

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter

FDA approves clinical trial to test nanoscale sensors for recording brain activity

The Federal Drug Administration has approved a clinical trial to test the effectiveness of a new electronic grid that records brain activity during surgery. Developed by engineers at the University of California San Diego, the device has nanoscale sensors that record electrical signals directly from the surface of the human brain in record-breaking detail.

The grid’s breakthrough high resolution could provide better guidance for planning and performing surgeries to remove brain tumors and treat drug-resistant epilepsy and could improve neurosurgeons’ ability to minimize damage to healthy brain tissue, the researchers say. During epilepsy surgery, the novel grid could also improve the ability to precisely identify the regions of the brain where epileptic seizures originate, allowing for safe, effective treatment.

Ultra-thin brain sensor array

The new brain sensor array ( known as “platinum nanorod grid (PtNRGrid)” features a densely packed grid of a record-breaking 1,024 embedded electrocorticography (ECoG) sensors. The device rests on the surface of the brain and is approximately 6 microns thin and flexible. So it can both adhere and conform to the surface of the brain, bending as the brain moves while providing high-quality, high-resolution recordings of brain activity. 

In contrast, the ECoG grids most commonly used in surgeries today typically have between 16 and 64 sensors and are rigid, stiffer and more than 500 microns in thickness; and do not conform to the curved surface of the brain.

The PtNRGrid was invented by Shadi Dayeh, a Professor in the Department of Electrical and Computer Engineering at the University of California San Diego and members of his team. The team developed the PtNRGrid technology in collaboration with neurosurgeons and medical researchers from UC San Diego, Massachusetts General Hospital (MGH) and Oregon Health & Science University (OHSU).

Currently, Dayeh’s research group holds the world record for recording brain activity from a single cortical grid with 2,048 channels on the surface of the human brain, published in Science Translational Medicine in 2022.

The clinical trial is designed to demonstrate the effectiveness of the PtNRGrid device to map both normal and pathological brain activity. Surgeons will implant the PtNRGrid in 20 patients, then measure and compare the grid’s performance with the present state-of-the-art. The PtNRGrid will be deployed in surgeries to remove brain tumors and to remove tissue that causes epileptic seizures.

“Our goal is to provide a new atlas for understanding and treating neurological disorders, working with a network of highly experienced clinical collaborators at UC San Diego, MGH, and OHSU,” Dayeh said in a statement.

Record-breaking density

Pending the success of this staged trial, the team will transition to the next crucial step: making the PtNRGrid available for commercial use at scale. Demonstrating that ECoG grids with sensors in the thousands of channels record brain activity with high fidelity also opens new opportunities in neuroscience for uncovering a deeper understanding of how the human brain functions. Basic science advances, in turn, could lead to improved treatments grounded in an enhanced understanding of brain function.

Dayeh’s work toward the FDA approval is supported by an NIH BRAIN® Initiative award.

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter

AI model identifies tennis-player affective states

Researchers at Karlsruhe Institute of Technology and the University of Duisburg-Essen have trained an AI model to accurately identify affective states from the body language of tennis players during games.

Their study, published in the journal Knowledge-Based Systems, demonstrates that AI can assess body language and emotions with an accuracy similar to that of humans.

Accuracy comparable to that of human observers

“Our model can identify affective states with an accuracy of up to 68.9 percent, which is comparable and sometimes even superior to assessments made by both human observers and earlier automated methods,” said Professor Darko Jekauc of Karlsruhe Institute of Technology Institute of Sports and Sports Science in a statement.

“The reason [for this accuracy] could be that negative emotions are easier to identify because they’re expressed in more obvious ways,” said Jekauc. “Psychological theories suggest that people are evolutionarily better adapted to perceive negative emotional expressions, for example, because defusing conflict situations quickly is essential to social cohesion.”

Body language clues

The researchers recorded video sequences of 15 tennis players in a specific setting, focusing on the body language displayed when a point was won or lost. The videos showed players with cues including lowered head, arms raised in exultation, hanging racket, or differences in walking speed; these cues could be used to identify the players’ affective states. 

After being fed this data, the AI learned to associate the body language signals with different affective reactions and to determine whether a point had been won (positive body language) or lost (negative body language). “Training in natural contexts is a significant advance for the identification of real emotional states, and it makes predictions possible in real scenarios,” said Jekauc.

Uses of reliable emotion recognition

The researchers envision a number of sports applications for reliable emotion recognition, such as improving training methods, team dynamics and performance, and preventing burnout. Other fields, including healthcare, education, customer service and automotive safety, could also benefit from reliable early detection of emotional states.

Citation: Darko Jekauc, Diana Burkart, Julian Fritsch, Marc Hesenius, Ole Meyer, Saquib Sarfraz, Rainer Stiefelhagen. Recognizing affective states from the expressive behavior of tennis players using convolutional neural networks. Knowledge-Based Systems, Vol. 295, 2024. DOI: 10.1016/j.knosys.2024.111856 (open access)

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter

Holographic acoustic device precisely targets diseased neurons in brain

Researchers at Washington University in St. Louis have developed a noninvasive technology to treat human brain diseases, such as Parkinson’s disease, that simultaneously involve damage in various regions of the brain. The holographic acoustic device, combined with genetic engineering, precisely targets affected neurons in selected cell types at multiple diseased brain regions.

Hong Chen, associate professor of biomedical engineering in the McKelvey School of Engineering and of neurosurgery in the School of Medicine, and her team created the noninvasive “AhSonogenetic” device to alter genetically selected neurons in the brains of mice. The results of the study were published June 17 in the journal Proceedings of the National Academy of Sciences

Multiple technologies

AhSonogenetics combines several of Chen’s group’s recent advances. In 2021, she and her team launched Sonogenetics, a method that uses focused ultrasound to deliver a viral construct containing ultrasound-sensitive ion channels to genetically selected neurons in the brain. They use non-invasive low-intensity focused ultrasound to deliver a small burst of warmth, which opens ion channels and activates the neurons. Chen’s team was the first to show that sonogenetics could modulate the behavior of freely moving mice.

In 2022, the team designed and 3D-printed a flexible and versatile tool known as an Airy beam-enabled binary acoustic metasurface, which allowed them to manipulate ultrasound beams. She is currently developing Sonogenetics 2.0, which combines the advantage of ultrasound and genetic engineering to modulate defined neurons noninvasively and precisely in the brains of humans and animals. AhSonogenetics brings them together as a potential method to intervene in neurodegenerative diseases.

Sonogenetics gives researchers a way to precisely control brains, while airy-beam technology allows researchers to bend or steer the sound waves to generate arbitrary beam patterns inside the brain at high spatial resolution.

Treating Parkinson’s disease

Chen’s team tested the technique on a mouse model of Parkinson’s disease. With AhSonogenetics, they were able to stimulate two brain regions simultaneously in a single mouse, eliminating the need for multiple implants or interventions. This stimulation alleviated Parkinson’s-related motor deficits in the mouse model, including slow movements, difficulty walking and freezing behaviors.

The device, which costs roughly $50 to make, can be tailored in size to fit various brain sizes, expanding its potential applications. The design file for the Airy-beam holographic transducer is available on GitHub: https://github.com/ChenUltrasoundLabWUSTL/AiryBeam_Lens_Design. Funding was provided by the National Institutes of Health.

Citation: CitHu Z, Yang Y, Gong Y, Chukwu C, Ye D, Yue Y, Yuan J, Kravitz AV, Chen H. Airy-beam holographic sonogenetics for advancing neuromodulation precision and flexibility. Proceedings of the National Academy of Sciences June 17, 2024. DOI. 10.1073/pnas.2402200121 (open access)

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter

Amber-like polymer allows for long-term storage of DNA and digital files

MIT researchers have developed a glassy, amber-like polymer that can be used for long-term storage of DNA, including entire human genomes or digital files such as photos. 

“The rapid decline in DNA sequencing costs has fueled the demand for nucleic acid collection to unravel genomic information, develop treatments for genetic diseases, and track emerging biological threats,” the researchers say.

Most current methods for storing DNA require expensive freezing temperatures and are not feasible in many parts of the world. The new polymer can store DNA at room temperature while protecting the molecules from damage caused by heat or water. 

“Freezing DNA is the number one way to preserve it, but it’s very expensive, and it’s not scalable,” says James Banal, a former MIT postdoc. “I think our new preservation method is going to be a technology that may drive the future of storing digital information on DNA.”

The researchers showed that they could use this polymer to store DNA sequences and that the DNA can be easily removed from the polymer without damaging it.

Banal and Jeremiah Johnson, the A. Thomas Geurtin Professor of Chemistry at MIT, are the senior authors of the study, published in the Journal of the American Chemical Society.

Capturing DNA

DNA offers a way to store this digital information at very high density: a coffee mug full of DNA could store all of the world’s data. DNA is also very stable and relatively easy to synthesize and sequence.

The researchers decided to make a thermoset polymer from styrene and a cross-linker, which form an amber-like thermoset called “cross-linked polystyrene.” This thermoset is also very hydrophobic, so it can prevent moisture from getting in and damaging the DNA.

“Inspired by the millennia-long preservation of fossilized biological specimens in calcified minerals or glassy amber, we present Thermoset-REinforced Xeropreservation (T-REX): a method for storing DNA in deconstructable glassy polymer networks,” say the researchers.

Storing information

Using these polymers, the researchers showed that they could encapsulate DNA of varying length, from tens of nucleotides up to an entire human genome (more than 50,000 base pairs). After storing the DNA and then removing it, the researchers sequenced it and found that no errors had been introduced, which is a critical feature of any digital data storage system.

The researchers also showed that the thermoset polymer can protect DNA from temperatures up to 75 degrees Celsius (167 degrees Fahrenheit). They are now working on ways to streamline the process of making the polymers and forming them into capsules for long-term storage.

Storing genomes

The earliest application they envision is storing genomes for personalized medicine, and they also anticipate that these stored genomes could undergo further analysis as better technology is developed in the future. 

The research was funded by the National Science Foundation.

Citation: Elisabeth Prince, Ho Fung Cheng, James L. Banal, and Jeremiah A. Johnson. Reversible Nucleic Acid Storage in Deconstructable Glassy Polymer Networks, Journal of the American Chemical Society. 10.1021/jacs.4c01925

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter

AI-powered exoskeleton helps restore mobility immediately

Researchers have used AI and computer simulations to train robotic exoskeletons, which can help users save energy while walking, running, and climbing stairs.

As described in the journal Nature, the novel method customizes exoskeleton controllers to assist locomotion for elderly or stroke survivors without relying on lengthy human-involved experiments in a clinic.

“It can also apply to knee or ankle exoskeletons or other multi-joint exoskeletons,” said Xianlian Zhou, associate professor and director at New Jersey Institute of Technology’s BioDynamics Lab, in a statement. “It can be used in above-the-knee or below-the-knee prostheses, providing immediate benefits for millions of able-bodied and mobility-impaired individuals.”

No user training

Previously, patients had to spend hours “training” an exoskeleton so the technology knew how much force was needed (and where and when to apply it). The new method allows someone to use the exoskeleton immediately. The closed-loop simulation incorporates exoskeleton controller and physics models of musculoskeletal dynamics, human-robot interaction, and muscle reactions.

This research was supported by the National Science Foundation, the National Institutes of Health, and the National Institute on Disability, Independent Living, and Rehabilitation Research.

Citation: Luo, S., Jiang, M., Zhang, S., Zhu, J., Yu, S., Dominguez Silva, I., Wang, T., Rouse, E., Zhou, B., Yuk, H., Zhou, X., & Su, H. (2024). Experiment-free exoskeleton assistance via learning in simulation. Nature, 630(8016), 353-359. 10.1038/s41586-024-07382-4

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter

How to create ‘wired miniature brains’

Researchers at the University of California San Diego have created highly realistic brain cortical organoids—miniature artificial brains with functioning neural networks.

The new technique, published in Nature Protocols, will enable scientists to perform more advanced research, according to Alysson Muotri, senior author and director of the UC San Diego Sanford Stem Cell Institute (SSCI) Integrated Space Stem Cell Orbital Research Center, in a statement.

Neurological disorders

The new research ranges from autism to schizophrenia (and other neurological disorders in which electrical activity is altered), testing potentially therapeutic drugs and gene therapies before patient use and screening for efficacy and side effects.

Previous methods of creating brain organoids have not enabled researchers to study the brain’s electrical activity, says Muotri. The new method enables researchers to study neural networks created from the stem cells of patients with various neurodevelopmental conditions. These new tiny replicas of the human brain are so realistic they rival “the complexity of the fetal brain’s neural network,” he said.

Organoids in space

Muotri and researchers at the Federal University of Amazonas in Manaus, Amazonas, Brazil, are also teaming up to record and investigate Amazonian tribal remedies for Alzheimer’s disease.

In March, Muotri—in partnership with NASA—sent brain organoids to space. The organoids were made from the stem cells of patients with Alzheimer’s disease and ALS (amyotrophic lateral sclerosis, also known as Lou Gehrig’s disease). The payload was returned in May 2024 for testing.

Microgravity

Using microgravity (experienced by astronauts in space) the researchers plan to mimic an accelerated version of Earth-based aging, allowing the researchers to witness the effects of several years of disease progression while studying the month-long mission’s payload, including potential changes in protein production, signaling pathways, oxidative stress and epigenetics. Other research possibilities for the brain organoids include disease modeling and understanding human consciousness, says Muotri.

This work was supported by the National Institutes of Health, California Institute for Regenerative Medicine (CIRM), a grant from the Department of Defense, and a “Humans in Space” grant by Boryung in Korea.

Citation: Fitzgerald, M. Q., Chu, T., Puppo, F., Blanch, R., Chillón, M., Subramaniam, S., & Muotri, A. R. (2024). Generation of “semi-guided” cortical organoids with complex neural oscillations. Nature Protocols, 1-27. https://doi.org/10.1038/s41596-024-00994-0

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter

Good global-warming news

A new study published in the journal Nature Climate Change has revealed “significant progress” in reducing levels in the atmosphere of chemicals that destroy Earth’s protective ozone layer.

Decline of harmful hydrochlorofluorocarbons (HCFCs)

The findings show, for the first time, a notable decline in the atmospheric levels of potent “ozone-depleting substances” (ODS) called hydrochlorofluorocarbons (HCFCs). These are harmful greenhouse gases, so a reduction should also lessen global warming.

The Montreal Protocol introduced controls on the production and usage of ODS, which were once widely used in manufacturing hundreds of products, including refrigerators, aerosol sprays, foams and packaging. HCFCs were developed as replacements for chlorofluorocarbons (CFCs) and CFC production has been banned globally since 2010.

Replacement with non-ozone-depleting compounds

According to lead author Luke Western, Marie Curie Research Fellow at the University of Bristol’s School of Chemistry, “production of HCFCs is currently being phased out globally, with a completion date slated for 2040. These are being replaced by non-ozone-depleting hydrofluorocarbons (HFCs) and other compounds. By enforcing strict controls and promoting the adoption of ozone-friendly alternatives, the protocol has successfully curbed the release and levels of HCFCs into the atmosphere.”

The results rely on high-precision measurements at globally distributed atmospheric observatories, using data from the Advanced Global Atmospheric Gases Experiment (AGAGE) and the National Atmospheric and Oceanic Administration (NOAA).

Two forms of ozone

The good: In the atmosphere, good O3 (ozone) protects us from the sun’s harmful UV rays. The famous hole in the ozone layer was created by the use of CFCs. These are now banned, so the hole is shrinking.

The bad: At ground level in urban environments, bad O3 (in the form of photochemical smog) is created by the reaction of urban pollution and sunlight. It becomes a powerful urban pollutant with negative health effects, as noted in What’s Worse Than Global Warming?

Citation: Western, L. M., Daniel, J. S., Vollmer, M. K., Clingan, S., Crotwell, M., Fraser, P. J., Ganesan, A. L., Hall, B., Harth, C. M., Krummel, P. B., Mühle, J., Salameh, P. K., Stanley, K. M., Reimann, S., Vimont, I., Young, D., Rigby, M., Weiss, R. F., Prinn, R. G., . . . Montzka, S. A. (2024). A decrease in radiative forcing and equivalent effective chlorine from hydrochlorofluorocarbons. Nature Climate Change, 1-3. https://doi.org/10.1038/s41558-024-02038-7

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter

Sugar substitute xylitol linked to increased risk of heart attack and stroke

Cleveland Clinic researchers have found higher amounts of the sugar alcohol xylitol are associated with an increased risk of cardiovascular events like heart attack and stroke.  

The team, led by Stanley Hazen, M.D., Ph.D., found an association in a large-scale patient analysis, preclinical research models and a clinical intervention study. Findings were published in the European Heart Journal  

Xylitol is a common sugar substitute used in sugar-free candy, gums, baked goods and oral products like toothpaste. Over the past decade, the use of sugar substitutes has increased significantly in processed foods that are promoted as healthy alternatives, said Hazen in a statement.

Investigating sugar alcohols and artificial sweeteners

“This study again shows the immediate need for investigating sugar alcohols and artificial sweeteners, especially as they continue to be recommended in combatting conditions like obesity or diabetes,” said Dr. Hazen, Chair of Cardiovascular and Metabolic Sciences at Cleveland Clinic’s Lerner Research Institute and Co-Section Head of Preventive Cardiology in the Heart, Vascular & Thoracic Institute.

Xylitol is a sugar substitute commonly used in sugar-free candy, gums, baked goods, and oral products like toothpaste. The use of sugar substitutes has increased significantly in processed foods as these products have been promoted as healthier alternatives to sugar.

Research limitation: association, not causation

The authors note that further studies assessing the long-term cardiovascular safety of xylitol are warranted. The research had several limitations, including the fact that the studies demonstrate association and not causation. They recommend talking to your doctor or a certified dietitian to learn more about healthy food choices and for personalized recommendations.   

The study was partly supported by the National Institutes of Health and the Office of Dietary Supplements.

Citation: Marco Witkowski, Ina Nemet, Xinmin S Li, Jennifer Wilcox, Marc Ferrell, Hassan Alamri, Nilaksh Gupta, Zeneng Wang, Wai Hong Wilson Tang, Stanley L Hazen, Xylitol is prothrombotic and associated with cardiovascular risk, European Heart Journal, 2024, ehae244, https://doi.org/10.1093/eurheartj/ehae244

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter

AI-powered tool lets scientists rapidly analyze complex biological images

Scientists at Chan Zuckerberg Biohub San Francisco (CZ Biohub SF) have developed Omega, an open-source software tool that significantly advances the bioimage analysis field, according to the scientists.

Omega is integrated into large language models (LLMs), such as OpenAI’s ChatGPT, enabling scientists to process and analyze biological images using natural language conversations, rather than issuing formal commands or writing code.

Omega was created by Loïc A. Royer and his team, and documented in a paper published June 10, 2024 in Nature Methods.

A plug-in for napari, Omega is an open-source image viewer used worldwide in diverse scientific fields, especially in biomedical research. “Omega allows users to quickly generate and edit code to solve complex image processing tasks,” explained Royer, a senior group leader and director of imaging AI at CZ Biohub SF.

In this example, the user asks Omega to z-project 3D images.

Omega’s collaborative features include:

  • Interactive image analysis: Users can instruct Omega to perform specific tasks, such as segmenting cell nuclei, counting objects, and generating detailed reports, all through simple conversational prompts.
  • On-demand widget creation: Omega can create custom widgets tailored to user-defined tasks, facilitating specialized image filtering, transformations, and visualizations.
  • An AI-augmented code editor: Omega includes an intelligent code editor that enhances code management with automatic commenting, error detection, and correction features.
  • Multimodal capabilities: Beyond text, Omega can interpret visual data, integrating multiple data types to provide comprehensive

Scientific community members are already using Omega, which has been available for download from a GitHub repository since May 2023. Royer said regular updates have been posted since then.

Omega source code: GitHub repository.

Citation: Royer, L. A. (2024). Omega—Harnessing the power of large language models for bioimage analysis. Nature Methods, 1-3. https://doi.org/10.1038/s41592-024-02310-w

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter