AI News Archive: May 24, 2026 — Part 7
Sourced from 500+ daily AI sources, scored by relevance.
- Identification of amino acid metabolism-related biomarkers in liver fibrosis: a transcriptomic analysis with experimental validation
Background Liver fibrosis (LF) represents a pivotal pathological phase in the advancement of chronic liver disorders toward cirrhosis. Amino acid metabolism reprogramming plays a pivotal role in its pathogenesis, yet the underlying molecular mechanisms remain incompletely understood. Methods Integrating three public datasets (GSE14323, GSE84044, and GSE136103) with amino acid metabolism-related gene sets, we performed consensus clustering, machine learning algorithms, functional enrichment analysis, immune microenvironment composition, regulatory network construction, and drug prediction. Results Fibrotic samples were classified into two amino acid metabolism-related subtypes with distinct immune landscapes and functional phenotypes. Through integrated analysis of differentially expressed genes (DEGs) common to both subtypes, fibrotic versus control comparisons, and amino acid metabolism-related gene sets, four biomarkers, GSTP1, LDHB, OXCT1, and PTGDS, were identified. These biomarkers were enriched in pathways related to epithelial-mesenchymal transition, interferon responses, and TNF/NF-{kappa}B signaling. Notably, GSTP1 and LDHB positively correlated with M1 macrophage infiltration and negatively with regulatory T cell abundance. Single-cell transcriptomic analysis revealed that cholangiocytes expressed all four biomarkers with elevated levels in fibrosis and interacted with macrophages/mesenchymal cells via MIF-CD74/CXCR4. Regulatory network analysis highlighted key modulators, including MALAT1, hsa-miR-3163, OXCT1, SMAD4, and RELA. Furthermore, 5-fluorouracil was predicted as a multi-target compound, with the strongest predicted binding affinity for OXCT1. In vitro validation confirmed the upregulation of GSTP1 and LDHB, aligning with the bioinformatics findings. Conclusion This study identified four amino acid metabolism-related biomarkers, revealing immune heterogeneity and cholangiocyte-centered intercellular communication in LF. These findings establish a foundation for biomarker-based diagnosis, subtype-guided patient stratification, and the development of cell-type-specific therapeutic strategies in LF.
- Explainable AI for Data-Driven Design of High-Dimensional Predictive Studies
Predictive modelling is important for health data analysis and data-driven clinical decision-making. However, predictive studies are challenging to design optimally by hand when tens or even hundreds of features require selection, transformation, or interaction modelling. While complex machine learning models offer high performance, their "black-box" nature limits the clinical trust, transparency, and interpretability required for decision-making. We developed and evaluated an Exploratory AI Recommender that provides data-driven recommendations to improve predictive performance of existing interpretable statistical models. The developed framework uses flexible AI modelling to capture complex data patterns and explainable AI techniques to translate the patterns into three recommendation types: feature exclusion, non-linear terms, and feature interactions. We evaluated the framework by comparing predictive performance of a baseline (i.e., no interactions or non-linear terms) Cox Proportional Hazards (CPH) model against an augmented CPH incorporating recommendations suggested by our method. The primary analysis predicts the time to the first occurrence of a fall or related injury in 245,614 patients. Our method recommended excluding 23 features, including non-linear terms for two features, and including 221 suggested feature interactions. The C-index improved from 0.805 (95% CI 0.798-0.812) to 0.815 (95% CI 0.809-0.822), and so did calibration (intercept: -0.006 to 0.003; slope: 1.063 to 0.950). All recommendations were supported by existing literature. The method also proved effective on two additional public datasets, demonstrating wider applicability. The proposed Exploratory AI Recommender demonstrates the potential of explainable AI and data-driven study design to improve the process of developing, and the performance of high-dimensional transparent predictive models.
- Automated GenePy Gene-Burden Computation via a Reproducible Nextflow Workflow Integrated with the Genomics England (GEL) Lifebit Platform
Interpretation of rare-disease genomes remains constrained by variant-centric analytical frameworks that insufficiently capture the cumulative impact of multiple variants within a gene. GenePy provides an individual-level, gene-based burden metric that integrates variant consequence, allele frequency, and zygosity into a unified quantitative score, enabling a transition from discrete variant annotation to aggregated gene-level interpretation. In the context of Genomics England, this formulation supports a panel-agnostic, genotype-to-phenotype diagnostic strategy for unresolved monogenic disorders by prioritising genes with elevated mutational burden per individual. Here, we present a fully automated, containerised GenePy workflow deployed through Nextflow and integrated within the Genomics England (GEL) Research Environment via the Lifebit CloudOS platform. This implementation provides scalable, secure, and governance-compliant computation of gene-level burden scores across population-scale cohorts. The workflow harmonises variant annotation, quality control, and chunked data aggregation within modular, reproducible processes designed for high-throughput execution on cloud-native infrastructure. By enabling robust, portable, and auditable gene-level scoring across large rare-disease sequencing datasets, this framework enhances analytical resolution and supports downstream statistical prioritisation, integrative phenotype matching, and hypothesis generation within genotype-to-phenotype diagnostic workflows.
- InSleep46: Deployment of a remote monitoring device for the detection and monitoring dementia risk in older adult populations: a feasibility study
Background: Improvements in health technology offer opportunities for remote disease screening, diagnosis and monitoring. The Withings Sleep Analyzer (WSA), an under mattress ballistocardiograph sensor able to detect body movement, breathing, and cardiac ejection is a promising technology for the non-invasive detection and monitoring of neurodegenerative diseases. InSleep46 aims to evaluate whether the WSA is able to detect preclinical Alzheimer's disease in members of the 1946 British Birth cohort, now in their late 70s. Objectives: To assess feasibility of deployment of a remote sleep, circadian and physiological monitoring device in a population of older adults. Participants: 356 participants from the Insight 46 neuroimaging sub-study (1946 British Birth Cohort), all born in one week in March 1946. Methods: We describe remote recruitment, device installation, and troubleshooting protocols. Feasibility analysis examined participant characteristics associated with recruitment and successful device set-up using logistic regression. Troubleshooting events for device installation and maintenance were recorded over a mean 14-month follow-up period. Results: During the feasibility analysis period, 263 (74%) participants, mean (SD) age 77 years (0.47) agreed to take part, of whom 245 (93%) successfully set up the WSA. Recruitment and successful set up of the WSA were not dependent on cognitive ability, socioeconomic position, or educational attainment. 162 (62%) of recruited individuals required [≥]1 troubleshooting call (mean 2.3 per participant, range 0-16). 603 calls were required in total. Conclusion: Deployment of a remote sleep and physiological monitoring device in an older adult population is feasible. Most participants required individualised assistance to set up the device. For the technology to be widely implemented, the set up must be accessible, with dedicated support available.
- Using genetics to aid detection of adverse drug effects: a Mendelian randomisation analysis of genetically proxied GLP-1RA in 1,020,464 participants across three population-based cohorts
Background: GLP-1 receptor agonists (GLP1-RAs) are an established treatment for type 2 diabetes mellitus (T2DM) and obesity. Their widespread use is set to increase through both indication expansion and patent expiry. As well as efficacy, it is crucial to understand the safety of this drug class to enable optimal use. Here we demonstrate how a genetic approach can augment signal-detection and post-market authorization surveillance. Methods: We used single nucleotide polymorphisms (SNPs) in GLP1R to recapitulate the effect of agonism with GLP1RAs on circulating glucose, glycated hemoglobin (HbA1c), body mass index (BMI) and risk of type 2 diabetes (T2DM) using Mendelian randomisation. We then tested if the adverse effect highlighted by medicines regulators of pancreatitis and the emerging effect of sarcopenia were causally related to GLP1R agonism, using this approach. Analyses were conducted in UK biobank and replicated in FinnGen and All of Us, results being combined using meta-analysis. Analyses were further stratified by a priori risk factors of age and alcohol consumption. Results: Genetically proxied GLP-1R agonism was associated with a reduction in glucose (exp({beta}) = 0.95 95% CI [0.94, 0.97]), HbA1c (exp({beta}) = 0.94 95% CI [0.92, 0.95]), and BMI (exp({beta})=0.98 95% CI [0.97, 0.99]); and a reduced risk of T2DM (OR = 0.82 95% CI [0.79 to 0.86]). Risk of acute and chronic pancreatitis was however increased (OR = 1.10 95% CI [1.01 to 1.20] and OR = 1.05 95% CI [0.95, 1.17], respectively), which varied as a function of age with risk most pronounced in those aged 50-59 years-old (OR = 1.79 95% CI [1.43, 2.24], OR = 1.57 95% CI [1.16, 2.12]) and in drinkers (OR = 1.32 95% CI [1.12, 1.54], OR = 1.36 95% CI [1.12, 1.65]). Risk of sarcopenia also increased (OR 1.34; 95% CI 1.05,1,71). Conclusions: Genetically proxied agonism with GLP-1RAs recapitulated the pharmacological effects of GLP1-1RAs on glycaemic traits, BMI and T2DM risk. This approach supports a causal effect of GLP-1RAs on the well reported adverse effects of pancreatitis and further indicates age and alcohol consumption as risk modifying effects. The less well reported but emerging effect of sarcopenia appears to also be casually related to agonism at GLP-1R. These analyses suggest a genetic approach could be used as an adjunct to signal detection studies to enhance safety regulation as well as personalisation of the use of these drugs.
- Freu AI
Automate any Mac app with $0 recurring run cost
- Hotels strive to be found as AI models conduct travel search
Hotels strive to be found as AI models conduct travel search The Straits Times
- Wavelet Decomposition-Based Genomic Analysis of the Human Electrocardiogram
The electrocardiogram (ECG) encodes the electrical activity of the heart across multiple timescales, yet standard clinical analysis collapses this rich signal into a handful of scalar measurements that discard most of the waveform's structure. Whether the frequency signals lost in this reduction carry heritable biological information relevant to cardiovascular disease risk remains unclear. Here we decompose resting 12-lead ECGs from 47,052 White British UK Biobank participants into 84 frequency-specific energy features using Daubechies-6 wavelet analysis across 12 leads and 7 decomposition levels, and perform independent genome-wide association analyses on each feature. We identify 67 independent loci and refine these to 101 high-confidence causal variants (posterior inclusion probability > 0.80) through Bayesian fine-mapping; associated loci converge on genes governing cardiac conduction and myocardial integrity, including SCN5A, TTN, KCNQ1, and DSP, alongside less-characterized cardiomyopathy candidates. SNP-based heritability estimates range from 0.03 to 0.26, with the strongest signals in mid-frequency bands (D6-D4, ~4-32 Hz) of Lead I and aVR, and strong inter-lead genetic correlations indicate a coordinated genetic architecture underlying the waveform. Integrating these features with FinnGen R12 cardiovascular phenotypes reveals genetic correlations reaching 0.56 with heart failure, driven predominantly by energy in the highest-frequency band (D1, 125-250 Hz), a spectral range routinely filtered from clinical ECGs and previously regarded as acquisition noise. These results reframe the electrocardiogram as a multi-frequency genetic phenotype, expand the set of cardiac loci discoverable from ECG data, and implicate high-frequency cardiac electrical activity as an underexplored dimension of cardiovascular disease risk.
- FlipTip AI
Spot profitable flips at thrift stores instantly.