Nat Methods. 2026 Jul 31. doi: 10.1038/s41592-026-03187-7. Online ahead of print.
NO ABSTRACT
PMID:42538469 | DOI:10.1038/s41592-026-03187-7
Tag: statistics
Nat Methods. 2026 Jul 31. doi: 10.1038/s41592-026-03140-8. Online ahead of print.
ABSTRACT
Spatially resolved multimodal data enable the exploration of transcriptional, proteomic and metabolic regulation, yet analytical tools to integrate these spatial omics modalities, particularly spatial metabolomics, remain limited. We developed SpaMTP, an end-to-end framework that implements functions within a common Seurat architecture. It introduces analyses for metabolite annotation, joint clustering, enrichment tests, spatial alignment, multimodal integration, visualization and seamless software interoperability. Its utility is demonstrated across different biological systems.
PMID:42538468 | DOI:10.1038/s41592-026-03140-8
Environ Sci Pollut Res Int. 2026 Aug 1. doi: 10.1007/s11356-026-38089-w. Online ahead of print.
ABSTRACT
Air pollution is a well-established risk factor for physical illness and premature mortality, yet its relationship with suicide remains less well understood, particularly at the national level and over longer periods of exposure. Although previous studies have reported positive associations between particulate matter and suicide, evidence remains heterogeneous, and national longitudinal analyses are limited. This study examines the association between fine particulate matter (PM₂.₅) concentrations and suicide mortality across local authorities in the UK between 2015 and 2023. Suicide data obtained from the UK Office for National Statistics were combined with modelled PM₂.₅ estimates from the Department for Environment, Food and Rural Affairs, resulting in a balanced panel of more than 2200 local authority-year observations. Fixed-effects panel regression models were employed to account for unobserved time-invariant local characteristics and common year-specific shocks, while nonlinear specifications were estimated to examine potential threshold effects. The descriptive analysis revealed a negative cross-sectional association between PM₂.₅ concentrations and suicide rates. However, after controlling for local authority and year fixed effects, PM₂.₅ was positively and significantly associated with suicide mortality (β = 0.221, p = 0.007). The nonlinear model further identified a statistically significant U-shaped relationship, with an estimated turning point at approximately 7.8 µg/m3, suggesting that higher PM₂.₅ concentrations are associated with increased suicide mortality beyond this threshold. Although the nonlinear specification modestly improved model fit, the findings should be interpreted cautiously given the observational design and the absence of certain time-varying socioeconomic controls. Overall, the results suggest that the relationship between PM₂.₅ and suicide mortality is conditional rather than uniform, varying across pollution exposure levels and geographic contexts. While the findings provide national-scale evidence on the association between long-term PM₂.₅ exposure and suicide mortality in the UK, they should be interpreted as exploratory associations rather than evidence of causality. Future research incorporating additional socioeconomic variables, longer time series, and higher temporal resolution data is needed to further evaluate these relationships.
PMID:42538457 | DOI:10.1007/s11356-026-38089-w
Nat Comput Sci. 2026 Jul 31. doi: 10.1038/s43588-026-01030-9. Online ahead of print.
ABSTRACT
Single-cell sequencing has transformed our understanding of cellular heterogeneity, enabling the construction of multi-omics atlases through data integration. However, conventional atlas updates require full reintegration of all datasets, creating scalability challenges that limit the timeliness and adaptability of biomedical research. Here we present multimodal integration with continual learning (MIRACLE), an online learning framework for scalable multimodal integration. Using dynamic architecture adaptation and data rehearsal, MIRACLE continually integrates diverse datasets while preserving biological fidelity. Across evaluations, MIRACLE achieves accurate online integration with substantially improved efficiency, refining and expanding atlases with new cross-modal, cross-tissue and cross-disease data. Applied to respiratory infections, it reveals both shared and pathogen-specific immune mechanisms in coronavirus disease 2019, influenza A and tuberculosis. Overall, MIRACLE provides an efficient and collaborative solution for the continual integration, sharing and exploration of biological knowledge.
PMID:42538449 | DOI:10.1038/s43588-026-01030-9
Nat Med. 2026 Jul 31. doi: 10.1038/s41591-026-04521-4. Online ahead of print.
ABSTRACT
Recent rapid progress in the field of computational pathology has been enabled by foundation models. These models are beginning to move beyond encoding image patches toward whole-slide understanding, but their clinical utility remains limited. Here we present PRISM2, a multimodal slide-level foundation model trained on 2.3 million whole-slide images and 14 million question-answer pairs derived from 700,000 pathology reports. Through clinical dialogue supervision, PRISM2 aligns histomorphology with diagnostic reasoning, yielding representations that support both prompt-based inference and transferable embeddings for downstream tasks. With prompt-based inference, PRISM2 achieves or exceeds (P < 0.05) the balanced accuracy of clinical-grade products calibrated for cancer detection in the prostate, breast and breast lymph node. Additionally, across comprehensive diagnostic, biomarker and survival benchmarks, PRISM2 embeddings never statistically underperform previous foundation models via linear probing (P < 0.05). Furthermore, task-specific fine-tuning on survival prediction outperforms training from scratch on the same large survival dataset. PRISM2 demonstrates how language-supervised pretraining provides a scalable, clinically grounded signal for generalizable pathology representations, bridging human diagnostic reasoning and foundation model performance.
PMID:42538427 | DOI:10.1038/s41591-026-04521-4
Int J Legal Med. 2026 Aug 1. doi: 10.1007/s00414-026-03935-6. Online ahead of print.
ABSTRACT
BACKGROUND: Molecular autopsy is increasingly used in unexplained sudden death, but reported diagnostic yield varies across sequencing eras, cohort types, and variant-classification frameworks. This review evaluated ACMG-corrected diagnostic yield, genomic architecture, VUS burden, and downstream family translation.
METHODS: This PRISMA-based systematic review and meta-analysis was registered in PROSPERO (CRD420251082728). Studies reporting postmortem genetic testing in sudden unexplained death were included. Quantitative diagnostic-yield synthesis included 44 core molecular autopsy studies comprising 4,041 index cases; 35 studies contributed family-translation data. Random-effects meta-analysis was performed, and subgroup analyses were conducted by sequencing strategy and age group.
RESULTS: The pooled ACMG-corrected diagnostic yield was approximately 13%, with substantial heterogeneity. Broader sequencing approaches showed modestly higher yield than candidate/restricted panels (OR 1.36, 95% CI 1.09-1.70; p = 0.006), but with greater VUS burden. Young adult/adult cohorts showed the highest pooled yield, while infant/pediatric cohorts demonstrated broader multisystem genomic architecture and higher VUS burden. Recurrent substrates included RYR2, SCN5A, KCNQ1, KCNH2, TTN, and MYBPC3. Family-translation data were available from 35 studies, including 25 overlapping index-case cohorts; genotype-positive relatives, phenotype-positive relatives, segregation analysis, and clinically actionable interventions were frequently reported.
CONCLUSIONS: Contemporary ACMG-compatible molecular autopsy provides modest but clinically meaningful diagnostic yield. Its major value extends beyond postmortem diagnosis, supporting family-based preventive genomic medicine through cascade screening, phenotype correlation, and targeted risk stratification.
PMID:42538417 | DOI:10.1007/s00414-026-03935-6
Eur J Clin Nutr. 2026 Jul 31. doi: 10.1038/s41430-026-01796-1. Online ahead of print.
ABSTRACT
BACKGROUND: The impact of meal timing and eating window on mortality and life expectancy remains unclear.
METHODS: This cohort study included 31,044 adults aged 40 years and older from the National Health and Nutrition Examination Survey 1999-2018. The meal timing and the eating window were estimated from a 24-h dietary recall.
RESULTS: During a median (IQR) follow-up of 8.6 (4.5-13.0) years, 7129 deaths were documented. Compared with a first meal within 7:00-8:00, the hazard ratios (HRs) and 95% CIs of all-cause mortality were 1.10 (1.02-1.18), 1.19 (1.05-1.34), and 1.29 (1.10-1.51) for a first meal within 8:00-10:00, 10:00-12:00, and after 12:00, respectively. Compared with a last meal within 19:00-20:00, the HR of all-cause mortality was 1.13 (1.03-1.25) for a last meal before 19:00, and 1.27 (1.06-1.54) for eating after midnight. The associations were also observed for cardiovascular mortality. The association between eating window and cardiovascular mortality risk varied by first meal timing. At the age of 50 years, the estimated life expectancy was on average 2.46 (0.61-4.28) years shorter among participants having a first meal after 12:00 and 2.35 (0.82-5.14) years shorter among those having a last meal after midnight compared with reference groups.
CONCLUSIONS: Later first meals and earlier or later last meals were associated with higher mortality risks and shorter life expectancy. These findings suggest that meal timing may be a modifiable behavioral factor associated with mortality risk and life expectancy.
PMID:42538400 | DOI:10.1038/s41430-026-01796-1
Mol Psychiatry. 2026 Jul 31. doi: 10.1038/s41380-026-03771-5. Online ahead of print.
ABSTRACT
Objective neural markers that reflect the underlying pathophysiological mechanisms of affective disorders are needed to facilitate early identification of individuals most at risk of future affective disorders and ultimately provide neural targets to guide therapeutic interventions. Using an emotional n-back paradigm designed to examine working memory (WM) and emotional regulation (ER) capacity, we previously showed that WM-related elevated left dlPFC activity (a key node of the central executive network (CEN)) and elevated right precuneus activity (a key node of the default mode network (DMN)), as well as ER-related elevated left dlPFC activity were positively associated with future depression severity in young adults at risk for affective disorders. We now aimed to replicate and extend these previous longitudinal findings by examining relationships among right precuneus activity and left dlPFC activity during WM and ER tasks and future depression severity in a new independent young adult sample (n = 77: 50 female, age = 24.68), and a larger combined sample (n = 121: 83 female, age = 23.81) comprising the original and new samples. The Hamilton Rating Scale for Depression (HAM-D) and Young Mania Rating Scale (YMRS) were measured at 12 months post scan to assess future depression and mania/hypomania severity respectively. In the new sample, we showed patterns of left dlPFC activity and right precuneus activity during WM, and left dlPFC activity during ER that were consistent with the original sample. In both the new and combined samples, future depression severity was robustly predicted by WM-related left dlPFC activity and right precuneus activity, and ER-related left dlPFC activity (all ps < 0.05 qFDR). These findings were specific to future depression severity. The effect sizes (pseudo R-squared values) for the full models including all IVs in the new and combined samples ranged from approximately 25-43%, with left dlPFC and right precuneus activity during WM explaining 15.59% of variance in future depression severity in the new sample; and left dlPFC activity during ER explaining 14.63% of variance in future depression severity in the combined sample. These replicated, longitudinal findings provide candidate neural markers to guide risk identification and targeting of new interventions for individuals with and those at risk for future affective disorders.
PMID:42538394 | DOI:10.1038/s41380-026-03771-5
Cancer Epidemiol. 2026 Jul 31;104:103166. doi: 10.1016/j.canep.2026.103166. Online ahead of print.
ABSTRACT
BACKGROUND: Alcohol and tobacco are among the leading preventable causes of cancer worldwide and frequently co-occur. While both are well-established independent carcinogens, their combined effects remain insufficiently characterized and inconsistently interpreted across cancer types.
OBJECTIVE: This review synthesizes epidemiological and mechanistic evidence on alcohol-tobacco interaction in carcinogenesis, distinguishing statistical interaction from evidence of biological mechanism.
METHODS: We examined interaction patterns across major cancer sites and assess the extent to which combined effects exceed individual risks. Particular attention is given to methodological approaches used to quantify interaction, including additive and multiplicative models and their implications for causal interpretation.
RESULTS: Current evidence supports a heterogeneous landscape of interaction. Strong and consistent synergistic effects are observed for upper aerodigestive tract cancers, whereas evidence is heterogeneous for hepatocellular carcinoma and context-dependent for breast cancer. By contrast, lung, pancreatic, and early-onset colorectal cancers show predominantly independent effects, with tobacco remaining the principal driver of lung cancer. Mendelian randomization studies support independent causal effects of smoking and, for some cancers, alcohol, but have not assessed their joint interaction.
CONCLUSION: To reconcile these findings, we propose a unified conceptual framework in which alcohol-tobacco interaction is viewed as a continuum ranging from strong biological synergy to functional independence, depending on tissue-specific vulnerability and exposure context. Finally, we identify key gaps in the field, including the need for standardized interaction metrics, prospective cohorts capturing joint exposure trajectories, integrative multi-omics approaches, and combined Mendelian randomization studies. Addressing these challenges will be essential to improve risk stratification and inform both precision oncology strategies and targeted, combined prevention efforts.
PMID:42537280 | DOI:10.1016/j.canep.2026.103166
J Psychosom Res. 2026 Jul 17;211:112931. doi: 10.1016/j.jpsychores.2026.112931. Online ahead of print.
ABSTRACT
BACKGROUND: Fatigue is a debilitating symptom in autoimmune liver disease, but prospective research on fatigue in Primary Sclerosing Cholangitis (PSC) remains limited. This study aimed to estimate fatigue frequency in patients with PSC and explore cross-sectional and prospective biopsychosocial associations with fatigue severity.
METHODS: This prospective study included 226 patients with PSC. Fatigue severity was assessed using the PBC-40 fatigue domain at baseline and six-month follow-up. Biomedical, psychological and social variables were assessed using clinical records, blood biomarkers and self-report measures. Descriptive statistics and multivariable linear regression models were used to examine cross-sectional and prospective associations with fatigue severity.
RESULTS: At baseline, 19% of patients reported clinically significant fatigue. Patients with fatigue reported greater psychobehavioral burden and impairment across several domains. Cross-sectionally, greater fatigue severity was significantly associated with higher depressive symptom severity (p < 0.001) and greater fear-avoidance behavior (p < 0.001), whereas biomedical disease severity markers were not independently associated with fatigue severity. At six months, most patients without baseline fatigue remained below the clinical threshold (92%), while fatigue persisted in 79% of patients with baseline fatigue. Prospectively, greater fear-avoidance behavior (p < 0.001) and detectable IL-6 serum level (p = 0.02) were associated with greater fatigue persistence over time.
DISCUSSION: Fatigue represents a clinically relevant and persistent symptom burden in PSC. Findings support a multidimensional biopsychosocial conceptualization of fatigue involving psychobehavioral and inflammatory processes and may provide useful directions for future mechanistic and interventional research.
PMID:42537269 | DOI:10.1016/j.jpsychores.2026.112931