Nature Medicine

Aging is the primary driver of chronic disease and mortality, requiring comprehensive frameworks for quantification of aging and nomination of longevity interventions. We developed mAge (multimodal age), a biological aging framework that integrates plasma proteomics, wearables, and mortality hazard to predict biological age, intrinsic capacity, and mortality risk. By combining proteomic and wearable data in UK Biobank samples, mAge exceeds unimodal baseline age prediction to 0.87 test R{superscript 2} and 2.3 years mean error, and reduces unimodal baseline mortality prediction error by 21%. We further constructed organ-and cell type-specific biological clocks that quantify aging across 49 distinct subsystems, revealing that cardiac, immune, and intracellular protein signatures benefit most from wearable integration. By mapping data to FDA-approved drug targets, we identified interventions, such as GLP-1 receptor agonists, gabapentin, and ACE inhibitors, that are associated with lower overall and subsystem-specific proteomic age and mortality risk or are associated with longer time-to-death and later age-at-death in longitudinal and deceased cohorts. mAge establishes a scalable framework for nominating and validating personalized longevity interventions, bridging continuous digital monitoring with molecular aging diagnostics.

Large language models (LLMs) now reach expert-level scores on medical licensing exams, encouraging the assumption that high scores imply safe medical judgment while patients increasingly use them for health advice. We show this assumption is fragile: when misleading context is injected into questions that LLMs originally answer correctly, they abandon the correct answer. We call the ability to maintain correct judgment under adversarial context epistemic resilience, and introduce MedMisBench to measure it. MedMisBench contains 10,932 medical question items and 48,889 misleading context-option pairs spanning medical reasoning, agentic capability, and patient-journey evaluation. Across 11 model configurations, mean accuracy falls from 71.1% on original questions to 38.0% under focused misleading context, with 51.5% attack success. The most damaging injections are formal, rule-like fabrications: authority-framed falsehoods reach 69.5% attack success and exception-poisoning claims reach 64.1%. A 14-member clinical panel from 7 countries identified serious potential harm in 38.2% of reviewed cases. MedMisBench exposes a structural blind spot in LLM evaluation in medical settings: existing benchmarks measure what models know, but not whether they preserve correct medical judgment under misleading context.1

Hospital antimicrobial resistance (AMR) emanates from an array of complex interactions between patient turnover, heterogeneous patient--staff contact patterns, antibiotic-driven within-host selection, and imperfect surveillance. We present a hospital AMR digital twin that combines mechanistic simulation with temporal graph learning to forecast resistance emergence from evolving daily contact networks and enable support intervention planning. Our approach is twofold: graph neural networks and transformers to model predictions and mathematical programming optimization to provide decision support. The main predictive task asks whether future spread of resistant infections is more likely to be driven by endogenous hospital transmission and selection or from importation on admission. We evaluated this task under both fully observed and partially observed settings, using baseline benchmarks together with ablations, surveillance perturbations, and distribution-shift stress tests. Under canonical conditions, the model achieved very strong predictive performance, especially when ground-truth system states were available, and remained informative under partial observation. Ablations showed that contact-weight information was relatively robust, whereas compressed node-feature representations weakened performance more noticeably when observations were incomplete. Surveillance stress tests further showed that delayed or less frequent reporting can be tolerated in some settings, but threshold calibration becomes fragile under more severe observation changes. Across broader epidemiological and surveillance shifts, the ground-truth model generally preserved strong ranking ability, while partial-observation performance was less stable. When models were trained directly in the shifted regime, performance improved compared with zero-shot transfer, indicating that the digital twin can adapt to new and previously unseen operating conditions but that portability across regimes can improve, particularly when only partial surveillance data are available. We also evaluated intervention-conditioned forecasting by branching hospital states into a small library of screening and isolation policies under a shock-and-superspreader regime. The learned models supported useful within-state action ranking and frequently identified policies that improved on the baseline containment protocols or avoided worsening outcomes. The same digital twin can also support constrained intervention selection, although reliable deployment will require careful calibration, improved robustness under partial observation, and broader policy libraries.

Biological aging is heterogeneous across organ systems, yet whether CT-derived abdominal aging provides prognostic value beyond routine clinical data and whether organ decomposition adds beyond a unified estimate remains untested. We developed and evaluated organ-specific and ensemble biological age models from radiomic features across five abdominal organs in 68,675 CT scans from 32,883 subjects, evaluated on alignment with chronological age of healthy subjects (nested cross validation: MAE=3.68 years, R^2=0.90). In sequential analyses restricted to adults aged 20-60 years which is the stratum of strongest BAG-disease association, ensemble biological age gaps provided incremental prognostic value beyond demographic covariates for all-cause disease and mortality (Delta C-index=0.141, 0.051) and beyond routine blood biomarkers (Delta C-index=0.048), confirming CT-derived aging captures structural information beyond laboratory markers. Organ-specific biological age added incremental prognostic value beyond ensemble selectively for focal diseases: cardiovascular (aorta, Delta C-index=0.091) and hepato-pancreatic (pancreas, Delta C-index=0.096). These findings establish a hierarchical organization of CT-derived biological aging, positioning routine CT as a source that adds prognostic value to existing clinical biomarkers.

Delayed diagnosis and poor antiretroviral therapy (ART) adherence remain primary drivers of HIV-related morbidity in low-resource settings, yet real-world AI validation at scale is lacking. We conducted a retrospective validation study using two publicly available, de-identified datasets: a Quality of Care cohort of 27,288 HIV-positive patients on ART across multiple healthcare facilities, and the CEPHIA multi-country assay database comprising 165,444 specimen records from six countries. Four machine learning classifiers were evaluated using 10-fold stratified cross-validation with SMOTE applied strictly to training folds. Explicit data leakage prevention, ablation analysis, calibration assessment, and bootstrap confidence intervals were applied. Economic projections used one-way sensitivity analysis. This study adheres to TRIPOD reporting guidelines. Random Forest achieved AUC-ROC of 0.9753 (95% CI: 0.970-0.975), sensitivity 87.3% (95% CI: 86.4-88.2%), specificity 95.7% (95% CI: 95.2-96.2%), and Brier score 0.079. Ablation testing confirmed robustness (AUC 0.963 without the primary predictor). Temporal validation on held-out future patients yielded AUC 0.772 (95% CI: 0.744-0.802), confirming generalisation across time. Real-world analysis revealed median diagnosis-to-ART delay of 74 days, with 47.3% of patients exceeding 90 days and 36.7% presenting with CD4 below 200 cells per microlitre. Multi-country CEPHIA analysis identified 18.6% HIV recency within the 130-day early-intervention window. Decision curve analysis confirmed net clinical benefit across threshold probabilities 0.03-0.45. Subgroup analysis demonstrated consistent AUC across sex, age, CD4 strata, and WHO staging (max difference 0.051). Economic modelling projected base-case savings of USD 415 per patient (USD 2.07 million per 5,000-patient cohort). These findings provide large-scale empirical evidence that AI-driven informatics can predict ART adherence failure and quantify systemic care gaps, offering a scalable framework for equitable HIV care delivery in resource-limited settings. Prospective external validation is required before clinical deployment.

Menopause affects over one billion women worldwide, yet remains poorly characterized at scale. We apply an ICD-10-based phenotyping algorithm to electronic health records (EHR) from an academic medical center (n=33,444 women aged 35-64) and a safety-net hospital system (n=7,041), yielding one of the most racially and socioeconomically diverse menopause cohorts in the literature. Structured EHR fields underrepresent symptom burden: only 38.8% of patients had any documented symptom via natural language processing, despite an estimated prevalence of 90%. Adverse pregnancy outcomes were associated with earlier menopause onset after adjustment ({beta}=-1.21 years, p=8.7x10-45). Menopausal women showed elevated risk for osteoporosis (hazard ratio of 12.40), rheumatoid arthritis (HR of 2.43), and mental and behavioral disorders (HR 2.38) relative to age-matched men, with divergence at menopause onset. We show that large-scale EHR can characterize menopause at a scale and diversity that prospective enrollment has not achieved.

Can we decode Alzheimers disease (AD) heterogeneity into a few portable axes that capture how amyloid-{beta}, tau and neurodegeneration (A-T-N) spatially co vary in vivo? To answer this question, we built a pipeline that harmonizes longitudinal amyloid-{beta}/tau PET and T1 MRI (gray matter) from ADNI cohort (12,430 images) with mixed effects modeling and then derived stage specific multimodal axes (mVCs) using linked component analysis, with robustness tested in simulations and external validation in the OASIS cohort (4,958 images). We identified a small set of multimodal axes that (i) recapitulate early tau weighted variation in cognitively unimpaired (CU) individuals, AD like A-T-N coupling in cognitively impaired (CI) individuals and atypical CU and CI participants with posterior (precuneus/occipitoparietal) and fronto insular/frontal weighted patterns, (ii) map onto domain specific cognition, APOE e4, and blood/CSF biomarkers of neurodegeneration, neuroaxonal injury and astrocyte activation, (iii) predict clinical transitions, (iv) generalize in an independent cohort, and (v) demonstrate modelling robustness to missing data, high dimensionality, and cross-cohort variability, enabling direct application of the extracted axes to new datasets for biomarker discovery and stratification. Multimodal axes provide a portable, interpretable layer for quantifying amyloid-{beta}-tau-neurodegeneration coupling at the individual level, complementing current biomarker-based staging frameworks based on A-T-N status and tau PET topography, and can be computed on new datasets to aid clinical assessment and trial enrichment.

Climate change is shifting infectious disease burdens1-6, but attributing transmission changes remains difficult where interventions and socioeconomic development interact with temperature-dependent signals7-11. Mechanistic models can isolate temperature-dependent signals from non-climatic influences5,12-16 but are often not tested against independent data. Here, we present a validation-first framework using a temperature-dependent R transmission model17 to detect and attribute temperature-mediated climate impacts on schistosomiasis transmission across Africa. First, semi-natural mesocosm experiments confirmed the model's biological constraints, with high temperatures suppressing the host-parasite system above ~33{degrees}C. Next, we established epidemiological relevance in the Lake Victoria Basin using 141,829 longitudinal infection records. Interannual temperature anomalies predicted infection risk, with anthropogenic warming accounting for 17.1% of observed infections relative to a natural-forcing-only counterfactual. Finally, across Africa, the mechanistic R predictor explained prevalence better than correlative climate metrics, even after accounting for intervention and socioeconomic covariates. Applying the validated framework to ensemble climate model simulations and a natural-forcing-only counterfactual (1984-2014) showed that anthropogenic warming increased transmission potential in cooler regions while suppressing it in hotter regions across Africa, a contrast projected to intensify under higher-emissions scenarios by mid-century. Climate impacts are not solely future threats, but present-day forces already reshaping transmission and disease burden.

Electronic health record (EHR) prediction models often summarize longitudinal histories as static patient-level features, which may omit potentially informative event ordering. We developed a simplified spike-timing-dependent plasticity (STDP)-inspired framework that represents asynchronous EHR data as sparse, directional transition features. The approach encodes whether one clinical event precedes another within prespecified temporal windows, preserving event identity, directionality, and approximate timing while retaining feature-level interpretability. We evaluated this framework in two retrospective prediction tasks with different temporal scales: incident acute kidney injury (AKI) prediction in 17,351 MIMIC-IV ICU stays and early postoperative recurrence prediction in 713 CUMC patients with pancreatic ductal adenocarcinoma (PDAC). Models were compared with static burden features (demographics, comorbidities, raw lab measurements) and in addition with STDP transitional feature sets using patient-level cross-validation and rolling prediction horizons. In AKI, a calibrated STDP ensemble model showed higher discrimination than static burden alone at the 24-hour decision snapshot for AKI by 72 hours, with AUROC 0.838 versus 0.800, and at 48 hours for near-term AKI prediction, with AUROC 0.868 versus 0.827. In PDAC, STDP transition features modestly improved Day -30 preoperative recurrence prediction, with AUROC 0.611 versus 0.587 and AUPRC 0.323 versus 0.318 for static burden and showed similar performance at Day 0 (7 days before recorded surgery date), with AUROC 0.681 and AUPRC 0.363. Decision-curve and feature analyses suggested that selected temporal transitions were clinically interpretable across renal, inflammatory, hepatobiliary, hematologic, glycemic, and nutritional trajectories. These findings suggest that STDP-inspired transition features may provide a practical, interpretable way to incorporate temporal ordering into EHR-based risk prediction across both acute and longitudinal settings

Of the two influenza A virus (IAV) subtypes circulating endemically in humans, A/H3N2 and A/H1N1pdm09, A/H3N2 has historically been the dominant driver of disease burden in older adults. Based on an analysis of publicly available global surveillance data from 2015 to 2025 (>300,000 subtyped, age-stratified infections), we report a substantially increased contribution of A/H1N1pdm09 to influenza morbidity in older adults since approximately 2022. Birth cohort-stratified analyses suggest elevated A/H1N1pdm09 burden among individuals born before 1955-1959, consistent with erosion of pre-existing immunity originally generated by exposure to historical A/H1N1 strains. Pooled estimates across datasets and analytical approaches indicate the increase in A/H1N1pdm09 burden rises with earlier birth year, ranging from 1.22-fold (95% CI 1.08-1.37) for the 1955-1959 birth cohort to 3.10-fold (95% CI 2.58-3.72) for the 1930-1934 cohort. These findings point to a substantial rise in the overall influenza burden among the most vulnerable age groups, with implications for vaccine policy, clinical management, and public health planning.

Multi-organ biological aging is often represented as parallel organ-specific clocks, but how age gaps should be interpreted within an anatomically coupled imaging system remains unclear. Applying end-to-end deep learning to abdominal Dixon MRI from 67,130 UK Biobank participants, we show that abdominal biological aging is hierarchically organized across eight compartments. Compartment age gaps were positively intercorrelated (mean pairwise r = 0.42), and their unweighted mean--the Overall Aging Gap (OAG)--broadly stratified all 15 prespecified prospective endpoints, including 14 incident diseases and all-cause mortality (hazard ratios 1.15-1.49 per s.d.; mortality HR = 1.41 per s.d.). After accounting for OAG, compartment-level associations became sparser and more anatomically coherent, indicating disease-specific refinement beyond the shared axis. Healthier lifestyle was associated with lower risk within accelerated-aging strata. These findings establish a hierarchical framework for interpreting abdominal MRI age gaps: OAG stratifies broad prospective risk, whereas axis-conditional compartment engagement refines disease-specific anatomical vulnerability.

We present VA-LLM, a 1.62-billion-parameter autoregressive transformer pre-trained from scratch on 1.74 trillion tokens of clinical text spanning 22 years of care for 13.8 million patients in the Veterans Health Administration, with mortality outcomes confirmed through the National Death Index for 7.8 million patients. In a retrospective-prospective evaluation on 107,555 withheld patients, VA-LLM achieved higher 5-year AUPRC than Llama-2 (7 billion parameters), BioGPT _large (1.57 billion parameters), and GatorTron (3.91 billion parameters), matching GatorTron's 100,000-patient performance with only 10,000 labeled patients. In a clinical validation against the VA's operational Care Assessment Need (CAN) score on 5.5 million patients one year beyond the pre-training corpus, VA-LLM achieved a 90-day mortality AUROC of 90.00% versus 87.74% (p < 0.001) and a 45% relative improvement in AUPRC; post-hoc recalibration recovered calibration comparable to CAN (Brier 0.0091 versus 0.0093) without sacrificing discrimination. Across 21 pre-training checkpoints, discriminative performance correlated more strongly with cumulative mortality experience (CME), the total person-years contributed by patients with confirmed deaths, than with token count ({Delta}R2 = 0.15; Williams p < 10-6). Performance plateaued once marginal cohorts added fewer confirmed deaths, even as pre-training loss continued to decrease. These findings suggest that the clinical composition of pre-training data, particularly the completeness of documented patient trajectories, correlates with predictive performance more closely than corpus size alone.

Approximately 30-40% of stroke patients retain aphasia at 12 months. Early forecasting may guide rehabilitation and prognostic enrichment of clinical trials, yet machine learning (ML) prediction of language recovery has typically relied on chronic-phase data unavailable at the acute decision point. Whether acute features predict 12-month outcomes, and whether global severity and connected-speech recovery share substrates in an ML framework, is untested. We studied 73 patients with acute left-hemisphere ischemic stroke and aphasia (mean 2.8 days post-onset). Two 12-month outcomes were defined: aphasia resolution (Western Aphasia Battery-Revised Aphasia Quotient [WAB-AQ] [&ge;]93.8) and discourse normalization (Modern Cookie Theft content units [&ge;]22.1; N=61). Four ML algorithms were trained on four hierarchical feature sets (clinical, volumetric, anatomical, network-disconnection) using nested cross-validation and SHapley Additive exPlanations (SHAP) stability analysis. Acute WAB-AQ dominated (mean |SHAP| = 13.60, ~20x the next feature). For aphasia resolution, random forest achieved F1 = 0.874 (95% CI, 0.800-0.941), Pearson r = 0.827, mean absolute error (MAE) = 7.26 WAB-AQ points; clinical features alone achieved F1 = 0.851. For discourse, support vector regression achieved F1 = 0.725 (95% CI 0.593-0.831), r = 0.617, MAE = 8.96 content units. Three predictors were shared (acute WAB-AQ, lesion volume, left pars triangularis); ventral-stream tracts were linked to aphasia resolution, whereas interhemispheric and prefrontal connectivity were linked to discourse. Both models overpredicted severe chronic outcomes. Acute-phase ML forecasts 12-month aphasia resolution accurately and discourse more modestly. Clinical features carry most predictive variance; acute imaging reveals shared and outcome-specific substrates mapping onto dual-stream architecture, supporting early stratification for rehabilitation and prognostic trial enrichment.

Mood and anxiety disorders emerge predominantly in adolescence, yet they are usually identified only once symptoms have consolidated, when intervention can only be reactive. A marker that registers the loss of healthy brain function before symptoms crystallise would allow earlier and more targeted treatment, much as caged canaries once warned miners of danger before it became apparent. Here we report such a marker using a single baseline resting-state functional MRI scan in 150 adolescents in the Human Connectome Project Boston Adolescent Neuroimaging of Depression and Anxiety (HCP BANDA) cohort, allowing us to prospectively predict depression and anxiety symptoms one year later in held-out participants at r = 0.60, substantially above the effect-size ceiling reported for functional connectivity in the same data. The marker is not computed from raw functional connectivity but read out from a whole-brain generative model fitted to each individual's dynamics, which gives access to interference structure that covariance-based features cannot represent. The regions driving the prediction, including precuneus, ventromedial prefrontal and anterior cingulate cortices, are among those previously implicated in internalising disorders, and the same signature tracks cognitive variation in healthy participants and is mechanistically linked to the efficiency of task-related computation. These findings establish a mechanistically interpretable and prospectively predictive marker of adolescent mental health and define a clear path towards external validation and clinical use.

Effective pandemic response requires accurate modeling of population compliance with non-pharmaceutical interventions (NPIs), yet most epidemic models treat behavioral change as fixed scenarios rather than an emergent process. Here, we test whether large language model (LLM)-based agents can generate individualized behavioral responses to time-varying NPIs and disease risk. We instantiate demographically representative agents in three U.S. cities (Boston, Denver, San Antonio) and condition them on evolving outbreak conditions and policies during the early COVID-19 pandemic, without fitting to observed mobility data. Across three frontier LLMs and their ensemble, agents generate zero-shot mobility changes across restaurants, retail, and entertainment venues, benchmarked against cellphone-derived foot-traffic records. The simulations recover average mobility trends across cities and venue types but exhibit overly narrow within-city variation. The three LLMs display distinct biases, while an ensemble approach improves robustness and overall performance. These findings establish LLM agents as a promising framework for modeling adherence to NPIs and highlight the need for further fine-tuning and empirical validation before they can support policy analysis.

Liver fibrosis is a major cause of death in low- and middle-income country contexts. In rural, poor areas of sub-Saharan Africa, schistosomiasis is an underestimated cause of liver fibrosis. Despite the need for increased diagnostic capacity for schistosomiasis-related liver fibrosis, there are no automated, clinically-validated tools to diagnose schistosomiasis-related liver fibrosis. We present SchistoTrackNet which is, to our knowledge, the first deep learning-based model for distinguishing distinct presentations of schistosomiasis-related liver fibrosis of varying severity. Ultrasound images from 1533 participants aged 5--84 years from three districts in rural Uganda were used to train and evaluate the presented models. The models were evaluated by assessing failure cases and by comparing results with re-readings performed by sonographers experienced in diagnosis of schistosomiasis morbidity. Our models show potential to enable automated reading of ultrasound images for schistosomiasis-related liver fibrosis to allow large-scale surveillance of schistosomiasis morbidity and contribute towards the World Health Organization target to eliminate schistosomiasis as a public health problem.

Suicidal risk may be encoded in everyday communication patterns but diluted in routine digital interactions. We introduce a method for surfacing this latent signal: training per-person language model agents on individuals' authored text (the on-screen text each participant typed, captured whenever a keyboard was visible in screenshots) and placing those agents in simulated social interactionsa communicative stress test. Using data from 79 adults with recent suicidal ideation, we ne-tuned individual LoRA adapters on Qwen3-8B using each participant's authored text, then placed agents in standardized conversations with probe personas. Agent-generated risk language was associated with EMA-measured suicidal ideation (r= .576, p < .001), with a single neutral small-talk probe performing nearly as well (r= 551). A shue control conrmed the signal is person-specic (r= .071 when adapters were mismatched), and automated descriptions of participants' general smartphone activity produced no signal, conrming specicity to interpersonal communication. A prompt ablation demonstrated partial robustness to removal of disclosure-encouraging language (r = .430). This proof-of-concept demonstrates that simulated social interaction can amplify latent vulnerability signals, bridging digital phenotyping, generative AI, andsuicide theory.

Estimates of population prevalence and genetic diversity of bloodborne viruses in healthy humans are essential to support population-scale monitoring for transfusion transmission risk. In the UK and globally, blood donations are routinely screened for a limited number of high-consequence pathogens, but the full composition of the plasma virome remains to be characterised. Using a novel quantitative targeted metagenomics sequencing approach, we analysed previously unscreened plasma donations collected by NHS Blood and Transplant in England for all major pathogenic and known commensal human bloodborne viruses, and quantified their viral burden. Here we show that in a representative sample of 5,064 UK blood donors in pools of 24 collected over a one-month period, the virome was dominated by a small number of largely persistent species, representing ~11% (12/106) of previously identified human bloodborne viruses. Anelloviruses (TTV, TTMV and TTMDV) was detected in 89.0% of pools, albeit at low read count inconsistent with measured anellovirus viral loads. In contrast, human pegivirus type 1 (HPgV-1), had estimated population prevalence of 3.7% (95% CI 3.0-4.4%), with high read count and complete genome recovery in around one half of positive pools, consistent with high titre in plasma. Estimated prevalences for less common detections included one species of gemykibovirus (0.12%), hepatitis C virus (genotype 1a, 0.04%) and various polyomaviruses and herpesviruses between 0.04% (parvovirus 4, BK polyomavirus) and 0.41% (human herpesvirus 6). Phylogenetic analyses revealed mixed TTV, TTMV and TTMDV populations and almost exclusively genotype 2 HPgV-1, consistent with known genotype distributions in Europe. Our results provide a baseline for describing the healthy plasma virome in UK blood donors.

COVID-19 risk scores developed during the pandemic relied on measurements contemporaneous with infection, leaving unresolved whether the metabolic and inflammatory vulnerability they capture pre-existed as a stable trait or was triggered by acute illness. Here, using 501,946 UK Biobank participants whose blood was drawn between 2006 and 2010---at least ten years before SARS-CoV-2 emerged---we show that baseline proteomic and metabolic profiles predict both COVID-19 hospitalization (2,783 events; C-statistic =0.676 [0.666--0.686]) and COVID-19 mortality (1,564 deaths; C-statistic =0.730 [0.701--0.760]) from parsimonious, regularized feature sets. The IL-1 pathway index (xIL1, +0.093) was independently selected for hospitalization but not mortality, while the IL-6 trans-signaling index (xIL6, + 0.040) was selected for mortality but not hospitalization---a differential pathway weighting corroborated by independent LightGBM/SHAP analysis and mirroring the subsequent success of tocilizumab (anti-IL-6R) and the limited efficacy of anakinra (anti-IL-1R) in reducing COVID-19 mortality in randomized trials conducted years later. The mortality model was additionally characterized by central adiposity (waist-hip ratio, +0.386), a respiratory compromise index (xRSP, +0.149), and prodromal cardiovascular disease (pCVD, +0.246). These findings establish that vulnerability to a novel pathogen is, in substantial part, a pre-existing and measurable prodromal state, with implications for pandemic preparedness and population-level risk stratification.

Genome-wide association studies have identified risk loci for aging brain disorders, but mechanistic interpretation depends on linking these loci to genes and to the tissues, cell types, and molecular modalities in which those genes act. Here we introduce FunGen-xQTL Multi-Brain (FGMB), a multi-context regulome-wide association atlas for transcriptome-wide association studies (TWAS) built from molecular datasets assembled by the ADSP Functional Genomics Consortium. FGMB provides cis-genetic prediction models for 17,375 protein-coding genes across 36 molecular datasets, 18 contexts, and 3 regulatory modalities, yielding more than 293,000 imputable gene-level or splice-event models. FGMB evaluates eight established and newer Bayesian or multivariate prediction methods, including cross-context models that borrow information across tissues and cell types. Applied to Alzheimer's disease, FGMB identified 327 TWAS associations and used joint fine-mapping of variants and predicted molecular traits to prioritize 146 gene--molecular-trait pairs, distinguishing regulatory associations from linkage disequilibrium (LD) hitchhiking.