Toggle light / dark theme

AI and lab tests to predict genetic disease risk

When genetic testing reveals a rare DNA mutation, doctors and patients are frequently left in the dark about what it actually means. Now, researchers have developed a powerful new way to determine whether a patient with a mutation is likely to actually develop disease, a concept known in genetics as penetrance.

The team set out to solve this problem using artificial intelligence (AI) and routine lab tests like cholesterol, blood counts, and kidney function. Details of the findings were reported in the journal Science. Their new method combines machine learning with electronic health records to offer a more accurate, data-driven view of genetic risk.

Traditional genetic studies often rely on a simple yes/no diagnosis to classify patients. But many diseases, like high blood pressure, diabetes, or cancer, don’t fit neatly into binary categories. The researchers trained AI models to quantify disease on a spectrum, offering more nuanced insight into how disease risk plays out in real life.

Using more than 1 million electronic health records, the researchers built AI models for 10 common diseases. They then applied these models to people known to have rare genetic variants, generating a score between 0 and 1 that reflects the likelihood of developing the disease.

A higher score, closer to 1, suggests a variant may be more likely to contribute to disease, while a lower score indicates minimal or no risk. The team calculated “ML penetrance” scores for more than 1,600 genetic variants.

Some of the results were surprising, say the investigators. Variants previously labeled as “uncertain” showed clear disease signals, while others thought to cause disease had little effect in real-world data.

Association between Coffee Consumption and Brain MRI Parameters in the Hamburg City Health Study

Despite the association of regular coffee consumption with fewer neurodegenerative diseases, it remains unclear how coffee is associated with pre-clinical brain pathologies such as lesions in the white matter, degeneration of the cortex, or alterations of the microstructural integrity. White matter hyperintensities (WMH) are hyperintense lesions on T2-weighted images and are associated with an increased risk for stroke and depression, cognitive deterioration, and gait disorders [13,14,15]. As a marker of cerebral small vessel disease (CSVD) and vascular brain damage, WMH can vary in the degree of expression, depending on the age and the presence of cardiovascular risk factors, e.g., smoking or hypertension [16,17,18]. Previous studies have reported diverging results on the association of consumed coffee with imaging markers of CSVD. They found either beneficial associations of coffee with lacunar infarcts [7], beneficial [19] or detrimental [20] associations with WMH volume, or no significant associations at all [21,22].

A recently developed and valid imaging marker of microstructural integrity is the peak width of skeletonized mean diffusivity (PSMD), calculated as the distribution of the mean diffusivity (MD) between the 5th and 95th percentile in the white matter skeleton [23]. Only one study analyzed the association of coffee consumption with microstructural integrity, as quantified by fractional anisotropy, with a higher coffee consumption being associated with higher integrity of the white matter microstructure [24].

Damage to the brain structure is not restricted to white matter, but also extents to the cortex, e.g., in the form of atrophy. Except for one study focusing on the quantification of cortical thickness in regions susceptible for Alzheimer’s Disease [22], the link between coffee consumption and cortical thickness was only indirectly examined by measuring total brain volume or grey matter volume, with incongruent results between studies [7, 21,25,26]. This study aimed at investigating whether coffee consumption is associated with multiple brain MRI markers of vascular brain damage and neurodegeneration, including WMH, PSMD, and cortical thickness in a large, population-based cohort.

Comprehensive molecular atlas of human hippocampus maps cell subtypes and organization

The hippocampus is an important brain region known to support various cognitive (i.e., mental) processes, including the encoding and retrieval of memories, learning, decision-making and the regulation of emotional states. While extensive research has tried to delineate the structure, functions and organization of the hippocampus, the cell types contained within it and their connections with other neurons have not yet been fully mapped out.

Over the past decades, available methods for studying cell subpopulations, the expressions of genes within them and their connectivity have become increasingly advanced. One of these methods, known as spatially resolved transcriptomics, works by measuring the expression of genes in cells while preserving their arrangement in space. Another called single-nucleus RNA-sequencing (snRNA-seq), allows scientists to examine RNA molecules inside individual cell nuclei to detect differences between them and categorize cells into different subtypes.

Researchers at Johns Hopkins Bloomberg School of Public Health, the Lieber Institute for Brain Development and Johns Hopkins School of Medicine recently used a combination of these two experimental techniques to examine cells in tissue extracted from the hippocampus. Their paper, published in Nature Neuroscience, introduces a comprehensive molecular atlas of the hippocampus that maps different cell subtypes and their organization.

Antifungal Drug Resistance: An Emergent Health Threat

Fungal infections, named mycosis, can cause severe invasive and systemic diseases that can even lead to death. In recent years, epidemiological data have recorded an increase in cases of severe fungal infections, caused mainly by a growing number of immunocompromised patients and the emergence of fungal pathogenic forms that are increasingly resistant to antimycotic drug treatments. Consequently, an increase in the incidence of mortality due to fungal infections has also been observed. Among the most drug-resistant fungal forms are those belonging to the Candida and Aspergillus spp. Some pathogens are widespread globally, while others are endemic in some areas only. In addition, some others may represent a health threat for some specific subpopulations and not for the general public. In contrast to the extensive therapeutic armamentarium available for the antimicrobial chemotherapeutic treatment of bacteria, for fungal infections there are only a few classes of antimycotic drugs on the market, such as polyenes, azoles, echinocandins, and a few molecules are under trial. In this review, we focused on the systemic mycosis, highlighted the antifungal drug compounds available in the pipeline, and analyzed the main molecular mechanisms for the development of antifungal resistance to give a comprehensive overview and increase awareness on this growing health threat.

Triple cardiovascular disease detection with an artificial intelligence-enabled stethoscope (TRICORDER): design and rationale for a decentralised, real-world cluster-randomised controlled trial and implementation study

Introduction Early detection of cardiovascular disease in primary care is a public health priority, for which the clinical and cost-effectiveness of an artificial intelligence-enabled stethoscope that detects left ventricular systolic dysfunction, atrial fibrillation and cardiac murmurs is unproven but potentially transformative.

Methods and analysis TRICORDER is a pragmatic, two-arm, multi-centre (decentralised), cluster-randomised controlled trial and implementation study. Up to 200 primary care practices in urban North West London and rural North Wales, UK, will be randomised to usual care or to have artificial intelligence-enabled stethoscopes available for use. Primary care clinicians will use the artificial intelligence-enabled stethoscopes at their own discretion, without patient-level inclusion or exclusion criteria.

GPT-5 prompting guide

GPT-5, our newest flagship model, represents a substantial leap forward in agentic task performance, coding, raw intelligence, and steerability.

While we trust it will perform excellently “out of the box” across a wide range of domains, in this guide we’ll cover prompting tips to maximize the quality of model outputs, derived from our experience training and applying the model to real-world tasks. We discuss concepts like improving agentic task performance, ensuring instruction adherence, making use of newly API features, and optimizing coding for frontend and software engineering tasks — with key insights into AI code editor Cursor’s prompt tuning work with GPT-5.

We’ve seen significant gains from applying these best practices and adopting our canonical tools whenever possible, and we hope that this guide, along with the prompt optimizer tool we’ve built, will serve as a launchpad for your use of GPT-5. But, as always, remember that prompting is not a one-size-fits-all exercise — we encourage you to run experiments and iterate on the foundation offered here to find the best solution for your problem.

Scientists fed people a fat-filled milkshake — it disrupted blood flow to their brains within hours

A single indulgent meal may carry hidden risks. Scientists found that drinking a milkshake with 130 grams of fat impaired blood flow to the brain within hours, raising concerns about stroke, dementia, and how everyday diets shape brain health.

How the distinctive folds in the brain cortex, seen in humans, whales, other animals, form

One of the defining features of humans is our brain’s remarkable capacity for language, planning, memory, creativity, and more. These abilities stem not just from our large brain size, but also from the folded structure of the brain’s outer layer, the cerebral cortex.

A new study, published in the journal Nature Communications, offers insight into how these wrinkles form, pointing to a range of contributing factors—including the number of early-stage , how they migrate during development, and the specific types of cells involved.

These findings may help guide future research into brain development, evolution, and health.

Epitranscriptomic Regulation of Hepatitis B Virus by RNA 5-Methylcytosine: Functions, Mechanisms, and Therapeutic Potential

Hepatitis B virus (HBV) remains a major global health challenge, with over 296 million people chronically infected worldwide. Despite the availability of antiviral therapies, a functional cure is rarely achieved, highlighting the need for novel therapeutic strategies. RNA 5-methylcytosine (m5C) is a pivotal epitranscriptomic mark implicated in RNA stability, transport, and translation. Emerging evidence shows that m5C is conserved within HBV RNA and plays critical roles in the viral life cycle. This review provides a comprehensive overview of the molecular mechanisms governing m5C deposition and recognition, summarizes recent advances in m5C biology, and highlights the emerging role of epitranscriptomic m5C regulation in HBV infection.

/* */