Researchers at UCLA have developed an artificial intelligence tool that can use electronic health records to identify patients with undiagnosed Alzheimer’s disease, addressing a critical gap in Alzheimer’s care: significant underdiagnosis, particularly among underrepresented communities.
The study appears in the journal npj Digital Medicine.
University of California San Diego researchers have discovered the enzyme responsible for chromothripsis, a process in which a single chromosome is shattered into pieces and rearranged in a scrambled order, allowing cancer cells to rapidly evolve and become resistant to treatment.
Since its discovery more than a decade ago, chromothripsis has emerged as a major driver of cancer progression and treatment resistance, but scientists haven’t learned what causes it. Now, UC San Diego scientists have solved this longstanding mystery in cancer biology, opening up new possibilities for treating the most aggressive cancers. The results are published in Science.
An interdisciplinary research team from Dresden University of Technology (TUD), Rostock University Medical Center (UMR) and Dresden University Hospital has developed an innovative, implantable and fully absorbable sensor film. For the first time, it enables reliable early detection of circulatory disorders in intestinal anastomoses—one of the riskiest surgical procedures in the abdominal cavity. The results have now been presented in the journal Advanced Science.
Intestinal anastomoses, which is the surgical connection of two sections of the intestine after the removal of diseased tissue, carry a considerable risk of post-operative complications. In particular, circulatory disorders or immunological reactions can lead to serious consequential damage or even death within a short period of time. However, direct monitoring of the suture site has not been possible until now, which often entails corresponding risks for patients as well as considerable costs due to follow-up operations and long hospital stays.
Based on this specific medical need, the interdisciplinary network of the Else Kröner Fresenius Center (EKFZ) for Digital Health at TUD and Dresden University Hospital brought together key experts from Dresden and Rostock.
After reviewing thousands of benchmarks used in AI development, a Stanford team found that 5% could have serious flaws with far-reaching ramifications.
Each time an AI researcher trains a new model to understand language, recognize images, or solve a medical riddle, one big question remains: Is this model better than what went before? To answer that question, AI researchers rely on batteries of benchmarks, or tests to measure and assess a new model’s capabilities. Benchmark scores can make or break a model.
But there are tens of thousands of benchmarks spread across several datasets. Which one should developers use, and are all of equal worth?
Scientists have discovered that beneficial root-dwelling fungi boost plant resilience to disease by remodeling the plant cell membrane at pathogen infection sites—offering critical new insights into how plants coordinate defenses in complex natural environments.
This new research reveals that the membrane interface between plant cells and invading pathogen microbes is not fixed. Instead, it can be reshaped by co-colonizing symbionts, fundamentally altering how plants interact with pathogens and potentially improving resistance to disease.
The study is published in the journal Cell Reports.
Predictive computational methods for drug discovery have typically relied on models that incorporate three-dimensional information about protein structure. But these modeling methods face limitations due to high computational costs, expensive training data, and inability to fully capture protein dynamics.
Ainnocence develops predictive AI models based on target protein sequence. By bypassing 3D structural information entirely, sequence-based AI models can screen billions of drug candidates in hours or days. Ainnocence uses amino acid sequence data from target proteins and wet lab data to predict drug binding and other biological effects. They have demonstrated success in discovering COVID-19 antibodies and their platform can be used to discover other biomolecules, small molecules, cell therapies, and mRNA vaccines.
Current genetic medicines are limited by tolerability, scalability, and immunogenicity issues. Utilizing components from viral and non-viral delivery platforms, we developed a lipid-based delivery vehicle formulated with a chimeric fusion protein that delivers nucleic acid cargo inside cells effectively and with broad distribution and low immunogenicity. This proteolipid vehicle platform is suitable for safe and effective repeat dosing of DNA and/or RNA in vivo.
Here, Conny Gysemans & team find variable patient responses align with specific immune gene signatures, offering a tool to predict treatment success or resistance.
Address correspondence to: Conny Gysemans, Leuven Diabetes Lab, Clinical and Experimental Endocrinology (CEE), CHROMETA, KU Leuven, Leuven, Belgium. Phone: 32.16.377454; Email: conny.gysemans@kuleuven.be.
Heart muscle cells grown from patient stem cells—known as human induced pluripotent stem cell–derived cardiomyocytes, or hiPSC-CMs—are a promising way to repair hearts damaged by heart attacks and heart failure. But transplanted hiPSC-CMs often have trouble syncing to the rhythm of native heart cells, which can cause dangerous arrhythmias after transplantation.
For years, stem cell biologists and cardiac researchers have been looking for ways to improve how implanted hiPSC-CMs mature and integrate into the heart. The challenge is that once the hiPSC-CMs are implanted in vivo, it’s hard to monitor how they integrate.
Now, Harvard University researchers have developed the first platform capable of continuously monitoring how transplanted cells mature, communicate, and synchronize with native tissue inside the body. Using this system, the researchers identified a self-assembling peptide that accelerated the maturation of hiPSC-CMs and improved the electrical coupling of the transplanted cardiac organoids. The research is published in Science.