The Expanding Role of AI in Science

Credit: iStock THE EXPANDING ROLE OF AI IN SCIENCE How Will AI Change the Food and Drink Industry? Could AI Help Predict the Next Pandemic? How Is AI Accelerating the Discovery of New Materials? CONTENTS 04 Could AI Help Predict the Next Pandemic? 08 How AI Is Transforming Cancer Prevention, Diagnosis and Treatment 12 How Is AI Shaping Proteomics and Multiomics? 16 How Will AI Change the Food and Drink Industry? 19 How Is AI Accelerating the Discovery of New Materials? 23 How AI Tools Are Shaping the Future of Neuroscience THE EXPANDING ROLE OF AI IN SCIENCE 3 TECHNOLOGYNETWORKS.COM FOREWORD Artificial intelligence (AI) is steadily becoming an integral part of scientific research and applied innovation. Across a wide range of disciplines, AI is being adopted as a tool to support data analysis, streamline complex processes and improve decision-making. This eBook brings together a series of articles that examine how AI is contributing to progress in varied fields such as cancer research, neuroscience, materials science, proteomics and the food and drink industry. The examples presented here illustrate how AI is being used to improve early disease detection, inform treatment planning, identify new materials, better understand biological systems and much more. This collection highlights practical applications, emerging challenges and areas of ongoing research, encouraging consideration of where and how AI can be most effectively used in science. The Technology Networks editorial team 4 THE EXPANDING ROLE OF AI IN SCIENCE Could AI Help Predict the Next Pandemic? Artificial intelligence can monitor disease outbreaks and could help prepare us for future pandemics Blake Forman The COVID-19 pandemic highlighted the speed at which infectious diseases can spread — and the importance of an equally agile and robust array of tools to predict, monitor and control their prolifera tion. New and long-standing artificial intelligence (AI) tools were deployed during the pandemic to help fill this role. Lessons learned from this period have shown that AI can be successfully utilized in early-warning systems for infection, outbreak detection, epidemiological forecasting and resource allocation. With new pathogens of concern and new strains of old viruses a constant threat, building on these AI-powered tools and incorporating them into public health is a key priority. This article outlines examples of where AI has been utilized to predict disease outbreaks and how AI models could help inform future strategies for controlling the spread of infectious diseases to prevent possible pandemics. AI’s contribution to pandemic preparedness In August 2024, the World Health Organization (WHO) updated its list of pathogens that could spark the next pandemic, which grew to include more than 30 pathogens. The microorganisms were selected based on available evidence showing them to be highly transmissible and virulent, with limited access to vaccines and treatments. While some pathogens on the list may never cause an epidemic, the growing number of pathogens of concern highlights the need for new tools to help predict and control the spread of infectious diseases. Recognizing the utility of AI in preparing for future pandemics, the US Centers for Disease Control and Prevention (CDC) Center for Forecasting and Outbreak Analytics launched Insight Net in 2023. The US network hopes to transform the analytic capacities Credit: iStock TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 5 for infectious disease outbreaks by combining machine learning and AI with the best available technologies and academic research. Similarly, the WHO Hub for Pandemic and Epidemic Intelligence is working towards implementing AI in surveillance programs. A key lesson from the COVID-19 pandemic was that effective preparedness relies on monitoring known pathogens and anticipating viral mutations that can evade host immune responses. To address this, researchers at Harvard Medical School (HMS) and the University of Oxford have developed an AI tool named EVEscape. To build the tool, the researchers took their existing generative model EVE – which can predict mutations in viral proteins that won’t interfere with the virus’s function – and added biological and structural details about the virus. Together, this data allows EVEscape to predict the variants most likely to occur as a virus evolves. In a study, published in the journal Nature, the researchers demonstrated that EVEscape is as accurate as high-throughput experimental scans at anticipating variations for SARS-CoV-2 and is generalizable to other viruses including influenza, HIV and understudied viruses with pandemic potential. The researchers continue to utilize EVEscape to predict future variants of SARS-CoV-2 and publish a biweekly variant report. They are now working on broadening this work to include other pathogens with pandemic potential. “We want to know if we can anticipate the variation in viruses and forecast new variants — because if we can, that’s going to be extremely important for designing vaccines and therapies,” said Debora Marks, professor of systems biology at the Blavatnik Institute at HMS. Disease surveillance in disaster contexts In addition to predicting how diseases may evolve, AI can also help track and contain epidemics. Earlywarning systems for disease surveillance have greatly benefited from incorporating AI algorithms that can analyze text for signals of infectious disease events with high accuracy and at unprecedented speeds. A study, published in the journal Emerging Infectious Disease, utilized open-source data from EPIWATCH – an AI early-warning system – to analyze the effects of the Russia-Ukraine war on infectious disease epidemiology. Conflict situations can increase the risk of epidemics, and disruptions to public health surveillance create extra challenges in tracking them. In the study, researchers demonstrated the value of using AIpowered open-source intelligence to gather information about unfolding epidemics in a conflict zone where formal surveillance was reduced. The researchers analyzed patterns of infectious diseases and syndromes before (November 1, 2021–February 23, 2022) and during (February 24–July 31, 2022) the conflict. Case numbers for the most frequently "We want to know if we can anticipate the variation in viruses and forecast new variants — because if we can, that’s going to be extremely important for designing vaccines and therapies." TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 6 reported diseases were compared with numbers from formal sources. The researchers found increases in overall infectious disease reports. In addition, compared with formal surveillance, the researchers were able to extract more complete case data for the eight most reported infectious diseases. While the study has some limitations, such as the lack of data from smaller regions in Ukraine, it demonstrates how open-source health intelligence systems can be valuable for making real-time public health decisions during disasters. Informing strategies against future pandemics AI-driven approaches complement human-curated ones, providing new insights to help health professionals make more informed decisions during outbreaks. A team of engineers at the University of Houston recently developed an AI tool to identify hotspots of infection linked to air traffic. This model could help policymakers decide on air traffic controls during pandemics. The study was published in the journal Scientific Reports. To explore how air traffic impacts the spread of disease, the researchers developed a graph neural network (GNN)-based framework called Dynamic Weighted GraphSAGE. “The uniqueness of this work is that the graph can accommodate dynamic weights, reflecting aviation pattern changes over time. We also used directed graphs to accurately capture the directionality and asymmetry of flight traffic between regions,” lead researcher and associate professor at the University of Houston Dr. Hien Van Nguyen, told Technology Networks. “These unique features make our GNN very suitable for modeling the spatial and temporal changes in air traffic, and from this, we can predict the spreading of infectious disease cases via air travel.” The researcher’s analysis found that air traffic significantly drove COVID-19 infection during the pandemic. By performing sensitivity analysis on the model, the researchers could observe changes in the spreading patterns. From this analysis, they found that Western Europe, the Middle East and North America were regions with the highest sensitivity to these changes and therefore concluded they had a disproportionately large impact on the spread of the virus. “This highlights the importance of looking at air travel patterns to potentially curb the spreading of airborne diseases,” said Nguyen. “In the past, governments have tried this, but I believe our model provides a more systematic, data-driven way to decide how much air traffic we want to cut and to predict how that would likely impact the spreading pattern of disease." “The model we’ve developed can be used as a data-driven tool for policymakers to evaluate the effectiveness of travel restrictions on the spread of airborne diseases,” said Nguyen. The next steps for this research will be to verify the model on other types of infectious diseases with different spreading rates and other properties that might impact spreading predictions. “Various infectious diseases can be applicable here because our framework "The model we’ve developed can be used as a data-driven tool for policymakers to evaluate the effectiveness of travel restrictions on the spread of airborne diseases." TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 7 is not restricted. It has no assumptions related to COVID-19, and we can apply this model to influenza or any airborne disease influenced by human travel migration patterns,” explained Nguyen. “In the future, we could potentially use this framework for early warning, not just predicting the spread of a disease but to detect spikes and unusual patterns.” Insights from the study were used to search through restriction policies on air traffic for controlling the pandemic, which identified policies and strategies that reduced predicted global COVID cases effectively with lesser air traffic reductions. Nguyen concluded, “We can apply the tool for other interventions such as where to increase health infrastructure, and where to deploy resources to anticipate the increased number of patients. So, in addition to helping with decisions on air travel, we can identify high-impact regions and quantify the potential outcomes of strategies aimed at helping these regions.” Future outlooks for AI in infectious disease monitoring AI has become an established technology in many areas of medicine. Emerging technologies such as quantum computing, biosensors, augmented intelligence and large language models are all predicted to play an increasing role in infectious disease surveillance in the future. While AI continues to improve surveillance infrastructures, limitations such as the prevalence of databases that underrepresent select populations and the potential for predictions that aren’t generalizable still need to be overcome. In addition, data privacy is a significant concern regarding using AI in infectious disease surveillance. As models incorporate data streams from sources such as wearable health technology, connected health devices and smartphones that may be linked to open social media, approaches to preserve privacy will be a priority. AI will likely continue to improve disease surveillance infrastructure, but future pandemics remain a possibility. Experiences with AI during the COVID-19 pandemic have shown its utility in this space, however, it still cannot replace the collective intelligence required to prevent emerging infectious diseases. Pandemic preparedness continues to require the combined efforts of collaborative surveillance networks. MEET THE INTERVIEWEE Dr. Hien Van Nguyen is an associate professor in the Department of Electrical and Computer Engineering at the University of Houston. His research focuses on the intersection of artificial intelligence, computer vision and biomedical image analysis. 8 THE EXPANDING ROLE OF AI IN SCIENCE Credit: iStock Artificial intelligence (AI) is rapidly transforming cancer research, driving innovations from early detection and diagnosis to treatment planning and drug discovery. AI houses the ability to process vast datasets at unprecedented speed and accuracy, allowing enhancements to image interpretation, uncovering novel biomarkers, predicting treatment responses and supporting personalized medicine. Recent collaborations between research institutes and tech companies, alongside AI-powered diagnostic approvals from regulatory bodies, reflect the field’s momentum. As these technologies evolve, they hold the promise of improving patient outcomes, streamlining workflows and redefining how clinicians approach cancer care. The use of AI in cancer prevention Prevention and early detection remain the most effective strategies for reducing cancer-related mortality. However, achieving widespread early detection is challenging due to the multitude of cancer risk factors, the variability in when and how early warning signs present and difficulties in knowing when to initiate screening. Access to cancer screening is also hindered by health inequalities such as proximity to healthcare providers, socioeconomic barriers and medical literacy. The application of AI holds promise for overcoming these obstacles, potentially transforming cancer prevention and early detection. For example, researchers developed an AI tool – encompassing data from 6.2 million patients over 41 years – capable of How AI Is Transforming Cancer Prevention, Diagnosis and Treatment Isabel Ely, PhD TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 9 identifying individuals at the highest risk of developing pancreatic cancer up to 3 years before diagnosis. Another study focused on leveraging large language models to monitor electronic health records with the goal of better understanding the social determinants of health – factors that can play a critical role in cancer prevention, early detection and treatment outcomes. By extracting and analyzing nuanced information embedded in clinical notes and patient histories, the large language models could help clinicians proactively address risk factors that traditional models might overlook. The integration of AI tools that can pinpoint individuals at heightened risk for certain cancers represents a significant advancement in precision medicine. Such tools enable clinicians to target diagnostic testing toward high-risk populations, reducing unnecessary procedures and associated anxiety for lower-risk individuals. AI applications in cancer imaging and analysis Technological advances in medical imaging and minimally invasive biomarkers hold promise in addressing challenges across the spectrum of cancer detection, treatment and monitoring. However, the interpretation of the large volume of data that is generated by these advancements presents a barrage of new potential challenges. Recent advances in AI methodologies have made great strides in automatically quantifying radiographic patterns in medical imaging data. For example, in 2021, the US Food and Drug Administration authorized the marketing of software designed to assist pathologists in identifying areas suspicious for cancer, supplementing the review of digitally scanned slide images from prostate biopsies. AI is also enhancing the processing of medical images, such as mammograms. A recent study demonstrated that AI imaging algorithms not only improve breast cancer detection from mammograms but also predict the long-term risk of invasive breast cancer. More recently, a study found that AI-assisted interpretation of brain scans may help improve care for children with gliomas – tumors that are typically treatable but vary widely in their risk of recurrence. Investigators trained deep learning algorithms to analyze sequential post-treatment brain scans and flag patients at risk of cancer recurrence. Uniquely, the researchers used a technique called temporal learning, which trains the model to synthesize information from multiple brain scans taken over several months after surgery. This approach differs from traditional models, which typically draw conclusions from single imaging snapshots. The temporal learning model predicted the recurrence of either low- or high-grade glioma within one year posttreatment with an accuracy of 75–89%, substantially higher than the ~50% accuracy associated with predictions based on single scans. While providing AI models with more post-treatment images improved prediction accuracy, the benefit plateaued after four to six images. Although further validation of AI models is needed before they can be applied clinically, researchers hope to launch clinical trials to determine whether AI-informed "The integration of AI tools that can pinpoint individuals at heightened risk for certain cancers represents a significant advancement in precision medicine." TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 10 risk predictions can improve patient care – ultimately the driving goal behind developing such tools. AI in cancer treatment development and scheduling Alongside aiding cancer detection and diagnosis, AI is becoming increasingly vital in developing new cancer treatments. For instance, researchers have used AI to identify activation patterns and predicted T-cell behavior to improve immunotherapy outcomes. Further, AI is also helping scientists uncover the biological mechanisms that underlie drug responses, with studies employing deep learning models to map shared drug response pathways, providing predictive insights that could inform future therapeutic strategies. Recent advancements also highlight the potential of deep reinforcement learning (DRL) frameworks in oncology. DRL frameworks have successfully tackled a variety of drug scheduling challenges, including managing immune responses after transplant surgery and controlling bacterial drug resistance. For example, a recent study applied a DRL network to optimize the treatment schedule in metastatic prostate cancer. “These ‘deep’ methods use artificial neural networks with many intricately connected layers that enable them to learn highly complex relationships between system variables,” Kit Gallagher, first author and PhD student at the University of Oxford and Moffitt Cancer Center, told Technology Networks. Gallagher explained how DRL frameworks operate: “At each time step, a deep learning agent receives information about the system’s state (e.g., tumor size) and selects an action (e.g., treat or not treat) from a set of options. The agent learns its strategy through trial and error, aiming to maximize a reward function that incentivizes positive outcomes, such as tumor shrinkage or cure, and penalizes negative events like excessive drug toxicity. DRL is especially well suited for these tasks because it can account for the long-term effects of actions, even when the relationship between actions and outcomes is not fully understood.” The DRL is then paired with a virtual patient, powered by a mathematical tumor model that simulates treatment responses. “This approach offers two major advantages: The tumor model can quickly generate the vast amount of data needed to train a deep learning system and it allows us to explore novel treatment strategies that have not yet been tested in clinical settings,” Gallagher said. By combining deep learning with mathematical modeling, researchers can design more adaptable and individualized cancer therapies. “The deep learning model allows us to generate personalized treatment protocols that account for variations in cancer dynamics between individual patients,” said Gallagher. “It can also be used to explore the best treatment scheduling approaches in more complex treatment settings where traditional analytic approaches are not possible, such as when multiple drugs may be scheduled simultaneously or the drug dose may be varied,” he added. "Incorporating patientreported outcomes that track symptoms and functional status in real time, beyond the clinic walls, may enhance the accuracy and clinical utility of prognostic AI models." TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 11 Challenges and future directions of AI in cancer care AI holds immense potential to revolutionize cancer care by enabling earlier diagnoses, generating more accurate risk assessments, guiding effective treatment strategies and freeing up clinicians to focus on patient-centered care. However, several challenges must be addressed to fully realize this potential. “The clinical applicability of machine learning approaches, such as deep reinforcement learning, can be limited by their ‘black-box’ nature – if we cannot rationalize treatment protocols recommended by the AI algorithm, then we cannot expect clinicians to place trust in these recommendations when treating their patients,” said Gallagher. Physician hesitancy in adopting AI tools often stems from this limited explainability, as well as concerns around medical liability, the financial consequences of errors and a general lack of familiarity with these technologies. To ensure the safe and effective use of AI in clinical settings, randomized clinical trials are needed to validate its applications. Another critical issue is bias in AI models. If the data used to train AI systems are not sufficiently diverse or representative of the broader population, these tools risk perpetuating existing medical biases. As such, the development and adoption of AI and machine learning models must be grounded in accepted standards that prioritize bias mitigation and reproducibility. It is also important to recognize that AI systems cannot fully replicate the nuance of human clinical decisionmaking. Factors such as patient demeanor, cognitive state and subtle clinical cues – while vital to assessing risk – are not always well-captured in datasets. Incorporating patient-reported outcomes that track symptoms and functional status in real time, beyond the clinic walls, may enhance the accuracy and clinical utility of prognostic AI models. Looking ahead, researchers are expanding the scope of precision medicine beyond drug selection to optimize treatment strategies. “The increasing potential of precision medicine has primarily focused on personalizing the choice of drug for each patient; however, the increasing integration of mathematical models into clinical care will also unlock the potential of personalized treatment schedules, allowing us to achieve more effective treatments with existing drugs,” Gallagher concluded. MEET THE INTERVIEWEE Kit Gallagher, MSc, is a mathematics PhD student at the University of Oxford. His current work focuses on treatment-resistant prostate and ovarian cancers but develops themes applicable to a range of disease and treatment settings. 12 THE EXPANDING ROLE OF AI IN SCIENCE How Is AI Shaping Proteomics and Multiomics? Molly Coddington Artificial intelligence (AI) has emerged as a powerful toolset that could create new opportunities and help overcome hurdles in proteomics and wider omics disciplines. Bolstered by AI, these fields of research could have a profound impact on science and society. At the Children's Medical Research Institute, the University of Sydney, Associate Professor Qing Zhong’s research interests span big data analysis, machine learning and computational biology. His work involves mining and managing large-scale proteomics and multiomics datasets. He aims to advance cancer research and implement big datadriven, evidence-based computational tools to enable predictive, preventive and personalized medicine, among other projects. Zhong recently joined Technology Networks for a conversation on AI’s progress in proteomics and multiomics, barriers to its widespread implementation and his vision for a “continuous, high-resolution lens on biology.” DIA-NN, DeeProM and AlphaFold “AI applications in proteomics have gained significant traction recently,” Zhong said. “Particularly with the emergence of data-independent acquisition neural networks (DIA-NN) for streamlined DIA analysis, DeeProM for predicting cancer cell vulnerabilities and AlphaFold for protein structure prediction.” Pioneered by the laboratory of Professor Ruedi Aebersold, DIA is considered a breakthrough technique in mass spectrometry (MS)-based proteomics. Unlike data-dependent acquisition (DDA), DIA offers unbiased analysis with larger proteome coverage and higher reproducibility, making it a useful method for discovery Credit: iStock TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 13 proteomics. Discovery research is incredibly important for interrogating the underpinning mechanisms of biological states, such as health and disease. There’s one drawback, however; DIA generates large amounts of data, which creates a bottleneck. “DIA-NN uses deep neural networks to handle large volumes of DIA data, simplifying peptide identification and quantitation,” Zhong explained. DIA-NN is also free to use, contributing to its growing popularity in highthroughput proteomics. In 2022, researchers – including Zhong – published a pan-cancer proteomic map of 949 human cell lines. The team developed a deep learning-based computational pipeline, named Deep Proteomic Marker, or DeeProM. “DeeProM enabled the full integration of proteomic data with drug responses and CRISPR-Cas9 gene essentiality screens to build a comprehensive map of protein-specific biomarkers of cancer vulnerabilities that are essential for cancer cell survival and growth,” Zhong said. A significant challenge in the study of proteins is their versatility, which is also why they’re so useful in biology. A protein’s function is closely related to its structure. For decades, scientists have worked to develop methods capable of deciphering protein structure. The issue, however, is that the number of different configurations a protein could adopt is enormous. Enter AlphaFold, an AI program developed by Google’s DeepMind that is trained on mass amounts of data from the Protein Data Bank to predict protein structure. “AlphaFold has revolutionized structural proteomics by accurately predicting protein folding, offering vital clues about protein function and interaction networks,” Zhong said. AlphaFold can also design de novo proteins – a longstanding challenge in the field –  for a wide variety of applications, including the development of novel therapeutics, diagnostics and imaging reagents. An estimated 2 million researchers across 190 countries are using AlphaFold to inform their research across several applications, from accelerating drug discovery, identifying protein structural alterations associated with diseases such as Alzheimer’s to generating plastic-eating enzymes. The model’s significant impact on science and society earned Google DeepMind’s Demis Hassabis and John M. Jumper onehalf of the 2024 Nobel Prize for Chemistry. “Together, these AI-driven approaches accelerate discoveries in disease mechanisms and therapeutic development, pushing proteomics beyond traditional experimental limits,” Zhong said. AI hurdles that are yet to be surmounted Though AI’s transformative potential in proteomics is being realized to some degree, its integration still faces significant challenges.  "Collaborative data-sharing frameworks, uniform standardization efforts and privacy-preserving technologies are urgently needed to accelerate AI-driven breakthroughs in proteomics and wider biomedical fields." TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 14 Data volume, quality and privacy A common misconception about AI’s position in proteomics and multiomics research, according to Zhong, is that there’s already enough data to drive AI research at the same speed as fields such as natural language processing or computer vision. “In reality, although biomedical experiments generate vast quantities of raw data, only a fraction of these datasets are well annotated, standardized and of high quality,” he said. “Unlike the billions of labeled texts or images available for training large language or vision models, biomedical data often remain scattered and behind institutional firewalls, limiting opportunities for building equally powerful AI systems,” Zhong continued. Collaborative data-sharing frameworks, uniform standardization efforts and privacy-preserving technologies are urgently needed to accelerate AI-driven breakthroughs in proteomics and wider biomedical fields. Privacy-preserving technologies will be integral to AI’s widespread adoption in healthcare research, where patient confidentiality is paramount. At the Human Proteome Organization 2024 World Congress, Zhong presented his recent pre-print research that seeks to address this challenge.* Zhong and colleagues developed a federated deep learning (FDL) approach, called ProCanFDL. FDL is a technique used to train AI models without sending raw data to the model itself – instead, the model is brought to the data. “Our system enables AI to learn from individual cancer proteomic data securely, behind local firewalls. In this system, each local computer trains its own AI model on private data, and only the updated local model parameters are aggregated to create a single, more robust global model,” Zhong explained. Local models were trained on simulated sites that contained data from a pan-cancer cohort and 29 cohorts that were held behind firewalls, representing 8 countries and 19,930 DIA-MS runs. “This global AI model demonstrated a significant improvement in accuracy for cancer subtyping tasks, highlighting its potential to uncover valuable insights into tumors and inform potential treatments – all while maintaining data security and privacy,” Zhong said. The researchers predict their approach could enable the development of large-scale, privacy-compliant proteomics AI models across institutions globally, advancing digital health. Funding for AI in omics Headlines are frequently dominated by sizable AI investment announcements. Though reports suggest venture capital deal activity in AI for healthcare has flourished over the last five years, Zhong believes there is a funding disparity.  “While massive investments – sometimes running into the billions – fuel the development of large language models (LLMs) in the tech sector, the same level of financial backing remains scarce for omics research,” he said. The impact? There are fewer opportunities to build “Large Omics Models” to a scale and size that compares to contemporary LLMs. “Limited funding slows the creation of foundational datasets, impedes the development of cutting-edge analytic tools and ultimately restricts the field’s growth,” Zhong emphasized, adding that there is an urgent need for greater philanthropic, governmental and industrial investment in omics-focused AI initiatives. An omics version of ImageNet Lastly, Zhong highlighted the pressing need for data standards and reproducibility in this line of research, especially as AI models in proteomics and wider omics studies become “increasingly data-hungry”, he said. “Much like ImageNet transformed the field of computer vision – and large, standardized corpora such as Wikipedia dumps did for language models – omics TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 15 studies need a well-curated, widely accessible reference dataset,” Zhong continued. The omics version of ImageNet, or “omics ImageNet”, as he described it, would help to unify metadata protocols, file formats and quality checks across different labs. Subsequently, this would enable reproducible, transparent benchmarking and foster collaboration. “Establishing such a foundation could dramatically accelerate AI-driven discoveries, making it easier for teams around the world to contribute to – and build upon – the same high-quality datasets,” Zhong said. A future without limits In a future without barriers, Zhong believes that AI could transform proteomics and multiomics into a “continuous, high-resolution lens on biology”, one that operates at a massive scale, which “might even dwarf the data used to train LLMs, like ChatGPT,” he said.  “Much as LLMs have reshaped the way we interact with technology, ‘Large Omics Models’ would seamlessly integrate proteomic, genomic and other molecular data, revealing complex cellular processes and disease pathways in real-time,” Zhong said. “By predicting how proteins and other biomolecules evolve, interact and respond under diverse conditions, these models would drive breakthroughs from new diagnostics to highly tailored therapies.” In this world, scientists could be freed from laborious data management issues. They could pursue creative research projects at an unprecedented pace. “Meanwhile, the broader public would reap the benefits of earlier disease detection, more precise interventions and a deeper comprehension of health that shapes public policy and healthcare worldwide,” Zhong said. Zhong paints a compelling picture of a future where AI, powered by unified data standards and transformative “Large Omics Models,” revolutionizes proteomics and omics research, delivering profound benefits to science, medicine and society. Given the rapid pace of current advancements, it may not be too long before we see whether this vision can become a reality. *This article is based on research findings that are yet to be peer-reviewed. Results are therefore regarded as preliminary and should be interpreted as such. Find out about the role of the peer review process in research here. For further information, please contact the cited source. MEET THE INTERVIEWEE Dr. Qing Zhong is an associate professor at Children's Medical Research Institute (CMRI) at The University of Sydney. His research interests encompass big cancer data analysis, machine learning and computational biology. 16 THE EXPANDING ROLE OF AI IN SCIENCE AI is now inescapable. What once was an acronym reserved for science fiction has become a buzzword in almost every industry. From aerospace to agriculture, banking to biotechnology, seemingly every sector is now scrabbling to think of ways to maximize their use of machine learning. And the food and beverage industry is no exception. Mondelez (the manufacturer of Oreos and many other products) says it has developed an AI tool to optimize the flavors of new products. Unilever is relying on AI analysis of weather data to help the company adjust ice cream sales forecasts to cut waste. Nestlé has announced it will create AI-powered “digital twins” of products like Nespresso coffee machines for future marketing materials. Many of these initiatives have come from food companies themselves, which have mountains of data on their own processes and projects. What the companies may lack, however, is AI expertise. This, says Dr. Nicholas Watson, a professor of AI in food at the University of Leeds, is where academia can help guide this emerging technology. AI and food production: greater connectivity required “They’ve [companies] got these large datasets that they can use, and it is your data, it's my data and it's personal data,” Watson told Technology Networks. Once partnered with a food company, Watson and his team focus on combining low-cost sensors with machine learning models to monitor and optimize production processes and predict food properties. “Some of the companies we work with make millions of their products a day,” said Watson. “That’s millions of How Will AI Change the Food and Drink Industry? Leo Bear-McGuinness Credit: iStock TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 17 samples for our training data set. So, we need to work with industry partners. They've got the data; we've got ideas and know-how.” In food factories, a lot of these data can be captured with low-cost sensor tools like weight scales and cameras. “You can imagine it's almost trivial to put a camera in a pizza factory and record every pizza,” said Watson, “but a lot of machine learning uses a technique called supervised machine learning, so you have lots of data, and then you label that data. So, for the pizza example, we would have good quality, bad quality or we could have a quality score from 0–10.” Even this set-up, however, still requires some mundane human-led work and the time that comes with it. For now, at least. “Capturing a million images of a million pizzas is just a case of putting a camera there and leaving it,” Watson said. “But someone has to look at every image and say, ‘Good quality, bad quality, 0, 3, 4, 8, 10…’ and that labeling of data is often what takes the time. We’ve used methods such as transfer learning, active learning and semi-supervised learning, where we label some of those edge cases which are quite hard to determine, to overcome that data labeling burden.” AI and food safety Aside from product quality, many companies are also interested in using AI to improve product safety. One microbiology startup, Spore.Bio, is planning to use AIpowered pathogen detection technology to reduce food testing time from days to minutes. However, caution in this area should be exercised, says Watson. Boosting food safety is a noble aim but one that can lead food producers – and AI researchers – to fruitlessly chase unobtainable goals. “One of my PhD students has been working all week to get his model accuracy from 99.5% to 99.55%,” Watson told Technology Networks, “and I’m like, ‘What does that actually mean in terms of the problem you’re solving?’” “Because if we’re looking at, say, allergens in food, 99% accuracy – which is generally good in most training models – means 1 in 100 is wrong. And if you’re making hundreds of thousands, if not millions, of products a day, that's a lot of potential risk you're making. So, I think it's about how do we manage that? Maybe we have this AI system as an early warning indicator, but we still continue with the very robust food safety protocols we have taking samples and sending them off to the lab.” A waste of wasted data Another promising use of AI is applying it to tackle the food and beverage industry’s gigantic waste issue. In 2019, the US Environmental Protection Agency estimated that 66 million tons of wasted food was generated in the industry’s retail and service sectors. An additional 40 million tons was thought to be generated in the manufacturing and processing of food and beverages. To cut down this secondary figure, many food manufacturers have recently incorporated AI into their workflow processes in order to identify areas where food waste could be prevented. Nestle, among other companies, recently announced they had joined Innovate UK’s BridgeAI initiative, an AI consortium that aims to redistribute up to 700 tons of quality surplus food – the equivalent of up to 1.5 million meals – by the end of its project. For their own part in this frugal goal, Watson and his team at the University of Leeds are collaborating with Australia’s Commonwealth Scientific and Industrial Research Organisation on an AI project to turn food waste into edible, sustainable proteins. “Typically, if you want to reuse food waste, you can turn it into a liquid media, then you could ferment that with yeast,” Watson explained to Technology Networks. “Then the yeast will procreate and you'll get a microbial protein, which you can then process into food.” “But there are lots and lots of different parameters. How do you treat the waste initially? How do you understand its varying composition? And then you go to TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 18 fermentation and you've got things like temperature, the type of organisms you use, pH, how long you ferment it for. This is just a big problem,” he continued. “With AI, we can actually reduce the time and cost from maybe 25 experiments to about 5 experiments. We use some of our own data, but then we go to the literature and use tools to extract all the information and data in that literature to augment what we're doing,” Watson added. From food waste to food taste Most foods, of course, have been cooked and eaten for centuries. Newer edible creations, however – such as processed plant-based meat alternatives – haven’t had the benefit of countless refinements over generations. Fortunately, says Watson, machine learning can condense these centuries of iterations into mere minutes, if given the right prompts. “One of the challenges you have with plant-based proteins is the bitterness and the taste,” Watson said. “They’re just not that enjoyable to eat. Some people might like that, some maybe don't.” “One of our research fellows has started a very prestigious scholarship where he will look at how you can select and process different proteins to reduce that bitterness level. He will collect some data, try different types of plant-based proteins and different processing techniques, and measure their bitterness; then we can build a model that will say, ‘OK, for this protein and this processing route you'll get to this bitterness level.’ But what we actually want to do is try and run some optimization on that, because we don't want to predict bitterness; we want to reduce it. That's generally what most people want.” This battle against bitterness can come into conflict with cultural tastes, however. “I just came back from a few weeks in China, visiting a university for a conference, and we had the pleasure of eating something called stinky tofu. It's one of the most interesting things I've ever eaten. I straight away spoke to some of my Chinese friends and said, ‘How can you possibly eat that?’ And they went, ‘What are you talking about? It's lovely. That blue cheese stuff you guys eat, what is that all about? That just tastes like sweaty socks.’ It’s just that cultural thing.” Whether such regional palettes can be programmed into AI algorithms remains to be seen. In the meantime, it seems a little human subjectivity can still go a long way in food research. MEET THE INTERVIEWEE Dr. Nicholas Watson is a professor of artificial intelligence in food at the University of Leeds. His research focuses on developing digital solutions to address environmental sustainability, food safety and health challenges in food production systems. . 19 THE EXPANDING ROLE OF AI IN SCIENCE How Is AI Accelerating the Discovery of New Materials? Alexander Beadle Batteries, solar panels, computer chips, carbon capture systems. All these innovative technologies, and others like them, are the result of serious breakthroughs in materials science – driven by the discovery and synthesis of novel inorganic materials. For decades, the discovery of new inorganic materials with more favorable properties was a daunting task of trial and error, with scientists forced to conduct hundreds upon hundreds of hours of painstaking experimentation to identify and synthesize just a handful of potential new materials. Computational chemistry was a revolution in the world of materials science when it was first introduced. With the increased availability of supercomputers and the combined efforts of physicists, chemists and computer scientists, researchers could simulate the behavior of molecules and materials at the atomic scale. This helped scientists to accurately predict the properties of new materials without the need for such repetitive physical experimentation, shedding a significant amount of the “trial and error” baggage. Today, with the advent of machine learning (ML) and artificial intelligence (AI), there is a sense that another materials revolution could be on the way, with AIguided materials discovery set to accelerate these computational approaches even further. Better materials for tackling the climate crisis “Let me just say that I am not a material scientist, I am a theoretical physicist,” Dr. Eliu A. Huerta began our conversation. Huerta is the lead for translational AI in the Data Science and Learning Division at Argonne National Laboratory, US, and has been working at the intersection of AI and scientific research for a decade Credit: iStock TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 20 already. Under his guidance, researchers at Argonne have been applying AI and advanced computational techniques to tackle grand challenges in astrophysics, cosmology, materials science and biophysics. “One thing I learned when I was being trained as a theoretical physicist is the value of being an outsider. You look at things in a completely different way and this allows you to propose new ideas,” Huerta told Technology Networks. “I am really excited about solving problems that AI alone cannot solve, that domain knowledge alone cannot solve, that supercomputing alone cannot solve — but where a mix of all of these can provide new approaches and new opportunities to understand science in a way you couldn’t do with these separate tools,” said Huerta. Huerta recently helped lead a project that sought to design new materials for carbon capture. Specifically, the group was interested in a class of compounds known as metal-organic frameworks (MOFs). MOFs are highly porous materials, made up of metal ion clusters and organic ligands which function as the network’s nodes and linkers respectively. Published in Communications Chemistry, the researchers used a generative AI diffusion model to suggest unique and chemically diverse linkers that could be used to make novel MOFs. These were screened using a modified neural network that would select the MOFs with the best theoretical carbon capture performance. The final structures were then validated using traditional computational chemistry methods, including molecular dynamics and grand canonical Monte Carlo simulations, to compute more credible CO2 absorption capabilities and make the final selection of best performers. “Typically, when you do computational chemistry, you know the structure that you want to validate. You know that the molecule already exists and you want to go and measure some properties. But we were interested in doing this and borrowing ideas from drug design and discovery,” Huerta explained. “So here we use the diffusion model, which is a generative AI model, and we expose that model to different molecular structures so that the model would learn not only about metal-organic frameworks, but about physics broadly speaking. I think this was the key to allowing the model to propose some chemical structures that were entirely novel." The strength of this kind of approach is its speed, Huerta continued. Creating new MOFs has been the focus of researchers for several decades, yet the number of exceptionally high-performing MOF materials is still relatively low. “Experimental science is needed, but it maybe is not the optimal way to go and discover new materials,” Huerta said. “Then again, if you are only using computational chemistry methods, it is very challenging to create a new material from the ground up! Now, with the method that we are proposing, we are learning from experimental chemistry and computational chemistry, but we are allowing AI to go and explore this vast chemical design "With the team’s tool having found success in suggesting novel, high-performance carbon capture materials, Huerta believes that similar workflows with altered parameters could help with accelerating the design of advanced materials for other applications." TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 21 space and find new things that we did not know about in the past.” In the study, Huerta’s team was able to generate over 120,000 MOF candidates in 33 minutes using a supercomputer at the Argonne Leadership Computing Facility. This was whittled down by the modified neural network to 364 AI-generated MOFs that were believed to be high-performing. In total, this process took just over five hours. Further computational analysis, which took only a few days to complete, found 102 stable MOFs in this dataset, of which 6 had a CO2 capacity that ranked in the top 5% of materials in the popular hMOF database. With the team’s tool having found success in suggesting novel, high-performance carbon capture materials, Huerta believes that similar workflows with altered parameters could help with accelerating the design of advanced materials for other applications. “When you have these tools, you want to develop the ability to go and tackle similar problems that you can apply MOFs in, for example, hydrogen storage. This is just another parameter that you can use to fine-tune your generative AI model. Going beyond this, there is also methane capture, with methane being another gas that is responsible for environmental pollution,” Huerta said. “There are many applications where we can use this software to go and explore the properties of new materials.” Improving batteries with AIgenerated materials Machine learning and generative AI are useful for more than just MOFs. As tools, they can be applied in a similar way to accelerate the design and discovery of other classes of material for different applications. “Each generation faces a defining technical challenge, and for ours, that challenge is climate change. It’s urgent and requires immediate action,” Dr. Austin Sendek, an adjunct professor of materials science & engineering at Stanford University, told Technology Networks. “AI has the potential to accelerate fundamental scientific processes. My focus on energy technologies is driven by both this scientific potential and the broader global impact of solving this issue,” Sendek said. At Stanford, Sendek’s research focuses on harnessing the power of machine learning and AI to design new materials that can support the decarbonization of the global economy. The main focus of this research is on batteries and electrochemistry – using machine learning to assist in the discovery of better electrolytes to support high-performance batteries. “Electrochemistry offers a new modality through which we can achieve many of the same outcomes as burning fuels, but by using clean, renewable electrons instead,” Sendek explained. “While powering vehicles through electricity via batteries is a major application, the scope of next-generation electrochemistry extends to areas like cement production, grid power backup and the creation of various materials, for instance.” The search for new and improved battery materials and electrolytes is a core part of the decarbonization effort. Higher-performance batteries can help store more energy from the grid when renewable energy production is high, meaning that more green energy can still be released even when production is low. Battery improvements could also lead to better electric vehicles with larger ranges or faster charging times – again helping to support the transition away from fossil fuels. In a 2018 paper, published in the journal Chemistry of Materials, Sendek and co-authors demonstrated the use of a machine learning-based prediction model to generate novel lithium ion conductors for use in allsolid-state batteries. “Electrolyte components heavily impact the performance and properties of these cells,” Sendek noted. “There’s a vast chemical space to explore, with over 10 billion commercially available molecules that can be used to modify electrolytes in various ways.” The team found that this machine learning-assisted approach was 2.7 times more likely to identify fast lithium conductors than a random search, in addition to TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 22 performing well in a head-to-head competition against six PhD students with experience in the field. In a review published in Advanced Energy Materials, Sendek and his colleagues also highlight the ability of machine learning-based approaches to assist in other areas, such as process optimization, cell lifetime prediction and battery modeling, in addition to accelerating materials discovery. To further explain why AI and machine learning are such helpful tools, Sendek strips it back to a matter of depth and breadth. “On the depth side, AI and machine learning help us identify patterns and relationships that would typically require scientific intuition and principles to uncover. For example, it can reveal how different electrolytes affect battery performance or predict the properties of new materials — tasks that, in the past, might have taken years of experimentation,” he said. “On the breadth side, once we develop predictive models, machine learning allows us to apply them to a much larger design space at incredible speeds. These models can rapidly evaluate vast numbers of molecules, often identifying potential candidates that would have taken traditional scientific methods much longer to find.” The future outlook for materials and AI AI and machine learning are quickly solidifying themselves as novel, useful tools in many areas of science – including medicine, proteomics, drug discovery and more. In materials science, the benefit of these tools is their speed. With today’s computing power, these tools can generate realistic novel materials at a blistering pace, freeing scientists from extensive loops of “trial and error” and allowing them to put their expertise to better use further down the development process. In the context of the global climate crisis, the search for innovative new materials for carbon capture and green energy storage cannot happen quickly enough. However, the application of AI and machine learning to materials science is still a relatively nascent field, with several limitations needing to be addressed as the field develops. For example, much of AI’s use in materials discovery and design relies on the use of prebuilt datasets, therefore it is key that these datasets are properly assessed to ensure their reliability and quality. Materials scientists have also begun to discuss the various ethical considerations that may come with incorporating AI into their work. This includes the importance of recognizing and mitigating any bias in an AI model’s training data, as well as remaining vigilant in recognizing AI-generated misinformation through the application of rigorous validation practices. MEET THE INTERVIEWEES Dr. Austin Sendek is an adjunct professor of materials science and engineering at Stanford University. His research and teaching focuses on harnessing the power of machine learning and AI to accelerate the design and discovery of new materials for decarbonizing the global economy. Dr. Eliu A. Huerta is a theoretical astrophysicist, mathematician and computer scientist whose research lies at the interface of physics, AI and computational science. His interdisciplinary work focuses on applying advanced computational techniques and AI methodologies to tackle grand challenges in astrophysics, cosmology, observational astronomy and complex systems.. 23 THE EXPANDING ROLE OF AI IN SCIENCE How AI Tools Are Shaping the Future of Neuroscience Rhianna-lily Smith AI has become a focal point of scientific inquiry and innovation, finding applications in fields as diverse as medicine, engineering and environmental science. In neuroscience, its potential is particularly intriguing. The brain is often described as one of the most complex systems in nature. Decoding its vast networks of neurons and understanding how they produce thoughts, emotions and behaviors requires interpreting immense datasets and conducting intricate experiments. AI is increasingly being used as a critical tool in neuroscience, helping researchers tackle complex challenges in understanding brain function. Large language models (LLMs) can process vast amounts of data while identifying patterns across scientific literature. This enables researchers to generate new hypotheses and explore potential outcomes in ways that were previously unattainable. Dr. Xiaoliang “Ken” Luo, a computational neuroscience and machine learning researcher at the Foresight Institute, has been at the forefront of this effort. In a recent study published in Nature Human Behaviour, Dr. Luo and his team demonstrated how LLMs could surpass human experts in predicting neuroscience research results. Their work introduced BrainBench, a benchmarking tool, and BrainGPT, a specialized LLM fine-tuned on neuroscience literature, which achieved an 86% accuracy in predicting experimental outcomes. These advancements highlight the potential of AI to accelerate scientific discovery and refine research methodologies. In this Q&A, Dr. Luo discusses the broader implications of AI in neuroscience, ethical considerations and how tools like BrainGPT may shape the future of research in the field. Credit: iStock TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 24 Q: What do you see as the most promising applications of AI in neuroscience? A: I, personally, see two main applications of AI in neuroscience that could offer great potential benefits. The first is a data-driven approach, where AI models like neural networks serve as powerful tools for building mechanistic understandings of the brain. These models can help explain the mechanisms behind neural activity related to various cognitive functions. The second, which is more relevant to the central focus of our publication, is a more meta-level application. Given the power of LLMs to synthesize vast amounts of information, I believe there's huge potential in leveraging generative AI to help neuroscientists more efficiently digest the literature, understand trends in understudied problems in the field and even inspire future research directions. Q: Neuroscience often inspires advances in AI, and vice versa. How do you think this interplay between studying the brain and building AI systems will shape the future of both fields? A: The relationship between neuroscience and AI has evolved in fascinating ways. While neural networks were initially inspired by the brain's architecture, recent AI advances have largely been driven by engineering breakthroughs in computing power and data processing rather than biological insights. However, I believe we're entering an exciting new phase of convergence between these fields. Current research comparing artificial and biological systems has revealed intriguing similarities in information processing and learning patterns. This bidirectional exchange offers unique opportunities: AI models can serve as testable mechanistic models of brain function, while neuroscience principles could help us develop more interpretable and robust AI systems. That said, this bio-inspired approach is just one of many valuable paths forward in AI development. The key is finding the right balance between learning from biology and pursuing purely engineering-based solutions. Q: Your study demonstrates that LLMs can outperform human experts in predicting study results. Does this suggest that AI might develop a form of "scientific intuition," and how might that differ from human intuition? A: That's an interesting perspective. While “scientific intuition” is challenging to define precisely, my guess is it stems from years of research experience and synthesizing connections across literature. The fact that LLMs trained on scientific papers can outperform human experts at prediction tasks suggests they may develop knowledge synthesis capabilities that differ from human approaches. An interesting research direction would be investigating how these models integrate information across neuroscience subdomains, which could reveal underlying patterns in the field and inspire new scientific connections. Q: How could tools like BrainGPT help address fundamental questions about brain function and cognition, especially in areas where direct experimentation is challenging? A: I should clarify a few things about BrainGPT. As we show in the paper, you could further fine-tune pretrained LLMs, at a relatively small cost, on neuroscience publications to build a model – which we call BrainGPT – that is better at predicting which study result is more likely. This success suggests the potential for LLMs to synthesize scientific literature and we hope they might eventually help identify novel connections and TECHNOLOGYNETWORKS.COM THE EXPANDING ROLE OF AI IN SCIENCE 25 suggest novel theories about the brain and cognition. We are actively exploring how BrainGPT might serve as a stepping stone toward developing systems that could help scientists navigate unexplored theoretical possibilities in neuroscience. I would say that direct experimental evidence remains irreplaceable and fundamental to scientific progress. Generative models like BrainGPT must be grounded in concrete experimental data. However, these models could help scientists explore potential research outcomes more efficiently. In an ideal world, scientists would test every possible hypothesis. But with limited resources and time, testing all possibilities becomes impractical. We envision that systems built upon BrainGPT could help researchers explore alternative scenarios and outcomes without conducting every conceivable experiment. By suggesting which experiments might be most promising and predicting possible results, such systems could help optimize resource allocation in scientific research. Q: Do you see parallels between how LLMs process information and how the brain does, or are these fundamentally different systems? A: Definitely. I think LLMs can provide inspiration for how human cognition works but I would be cautious in interpreting the success of LLMs as direct evidence of human-like processing mechanisms. There is a growing body of research on whether LLMs learn like humans and opinions are mixed. We have a recent paper out (a follow-up work from the Nature Human Behaviour paper) that shows LLMs trained on both forward and backward text perform equivalently on the BrainBench task. This is particularly telling since no human language has evolved to run backward (e.g., apple → elppa). The main takeaway from that paper is that LLMs are more general patternlearning machines than human brains. LLMs are excellent at extracting predictive patterns in sufficiently structured input – even reversed text – but it doesn’t mean they employ human-like information processing. Q: Do you think there are any ethical concerns researchers should be aware of, particularly regarding the over-reliance on AI predictions? A: While LLMs have achieved remarkable capabilities, they remain tools to enhance scientific work rather than replace human judgment. These systems excel at processing vast literature and potentially exploring possibilities, helping scientists work more efficiently. However, scientists must maintain their critical thinking and decision-making autonomy, knowing when to accept AI predictions and when to challenge them. Given the inherent biases in AI training data, these tools should augment rather than override human expertise. Q: While your study focuses on neuroscience, you mention that the methodology could be applied universally across sciences. What fields do you think might benefit most from this approach, and why? A: This approach would be particularly valuable in complex, interconnected fields like biology, where discoveries often require synthesizing information across multiple domains. As we show in the current paper, LLMs excel at identifying patterns and connections across diverse bodies of knowledge, making them especially useful for researchers navigating interdisciplinary challenges. MEET THE INTERVIEWEE Dr. Xiaoliang “Ken” Luo is a computational neuroscience and machine learning researcher at the Foresight Institute, where his work bridges neural networks and brain function. THE EXPANDING ROLE OF AI IN SCIENCE 26 TECHNOLOGYNETWORKS.COM CONTRIBUTORS Blake Forman Blake pens and edits breaking news, articles and features on a broad range of scientific topics with a focus on drug discovery and biopharma. Blake earned an honors degree in chemistry from the University of Surrey, which involved a placement year at the Medicines and Healthcare products Regulatory Agency (MHRA) laboratory, where he developed new pharmaceutical testing methods. Blake also holds an MSc in chemistry from the University of Southampton. His research project focused on the synthesis of novel fluorescent dyes often used as chemical/bio-sensors and as photosensitizers in photodynamic therapy. Blake held several editorial-based roles before joining Technology Networks as Senior Science Writer in 2024. Isabel Ely, PhD Isabel is a Science Writer and Editor at Technology Networks. She holds a BSc in exercise and sport science from the University of Exeter, a MRes in medicine and health and a PhD in medicine from the University of Nottingham. Her doctoral research explored the role of dietary protein and exercise in optimizing muscle health as we age. Molly Coddington Molly is a Senior Writer and Newsroom Team Lead at Technology Networks. Molly reports on various scientific topics, covering the latest breaking news and writing longform pieces. Before joining Technology Networks in 2019, Molly worked as a clinical research associate in the NHS and as a freelance science writer. She has a first-class honors degree in neuroscience from the University of Leeds and received a Partnership Award for her efforts in science communication. Leo Bear-McGuinness Leo is a Science Writer at Technology Networks where he focuses on environmental and food research. He holds a bachelor's degree in biology from Newcastle University and a master's degree in science communication from the University of Edinburgh. Alexander Beadle Alexander is a Science Writer and Editor for Technology Networks. He writes news and features for the Applied Sciences section, leading the site's coverage of topics relating to materials science and engineering. Before joining Technology Networks in 2023, Alexander worked as a freelance science writer, reporting on a broad range of topics including cannabis science and policy, psychedelic drug research and environmental science. He holds a masters degree in Materials Chemistry from the University of St Andrews, Scotland. Rhianna-lily Smith Rhianna-lily graduated from the University of East Anglia with a BSc in biomedicine and completed her MSc by Research in microbiology at the Quadram Institute Bioscience in 2023. Her research primarily focused on the gut microbiome in pregnant women throughout gestation. During her MSc, she developed a passion for science communication and later joined Technology Networks as an Editorial Assistant.