March Selected literature from CNS journals
Social disadvantage accelerates aging
AI-driven personalized medicine: transforming clinical practice in inflammatory bowel disease
A Western dietary pattern during pregnancy is associated with neurodevelopmental disorders in childhood and adolescence
4 | Nature Communications
DrBioRight 2.0: a large language model-powered bioinformatics conversational system for large-scale cancer functional proteomics analysis
5 | Nature Communications
Prospective cohort study and genetic analysis reveal associations between 2,923 plasma proteins and the risk of inflammatory bowel disease
6 | Nature Communications
Plasma proteomics-based identification of biomarkers for the polygenic risk of type 2 diabetes and its comorbidities
7 | The Lancet Digital Health
Proteomics-based organ aging signatures predict 20-year risk of age-related disease: the Whitehall II cohort study
8 | Briefings in Bioinformatics
A cancer driver gene discovery method that fuses large language models with omics causal inference
Social disadvantage accelerates aging
Social disadvantage accelerates aging
Social disadvantage, like advanced age, is an important driver of multiple health risks, but its specific impact on the aging process has remained unclear. A research team at the UCL Queen Square Institute of Neurology, University College London, published an important study in Nature Medicine that systematically reveals how social disadvantage accelerates biological aging and thereby raises the risk of age-related diseases (ARDs). Using a multi-cohort approach, the study examined the associations between social disadvantage, age-related plasma proteins and age-related diseases. It found pronounced features of accelerated immunosenescence among socially disadvantaged groups and identified 14 age-specific proteins closely tied to social disadvantage in early and later life (including DNAJB9, F2, HSPA1A and BGN). These proteins explained 39% of the attributable risk for the elevated incidence of 66 age-related diseases in socially disadvantaged populations. Mechanistic analysis showed that aberrant activation of the pro-inflammatory regulator NF-κB and its downstream effector interleukin-8 is the core pathway. The study confirms that social disadvantage accelerates aging across the entire life course, highlights the critical influence of social inequality on healthy aging, and offers potential biological targets for intervention.
▼
Scan the QR code to read the original article
AI-driven personalized medicine: transforming clinical practice in inflammatory bowel disease
AI-driven Personalized Medicine: Transforming Clinical Practice in Inflammatory Bowel Disease
The marked clinical heterogeneity of inflammatory bowel disease (IBD) poses a major challenge to precise diagnosis and treatment, and conventional endoscopy and histological analysis fall well short when it comes to predicting medium- and long-term outcomes. Recent work by the team of Professor Marietta Iacucci at University College Cork, Ireland, shows that artificial intelligence is reshaping the IBD care paradigm. The study developed a pioneering AI-driven standardized assessment system capable of real-time, precise evaluation of disease state, with distinctive strengths in predicting treatment response. Using deep learning algorithms, the team quantified features of intestinal barrier healing for the first time, providing important clues for the discovery of new therapeutic targets. Even more striking is the "Endo-Histo-OMICs" integrative approach proposed in the study, which seamlessly fuses endoscopic, histological and molecular data and opens up a new model of multi-dimensional patient stratification. Although several challenges remain to be overcome before clinical translation, this AI-enabled strategy marks an important shift in IBD management from empirical medicine to precision medicine, offering innovative solutions for improving long-term outcomes and quality of life and representing a significant breakthrough in gastroenterology.
▼
Scan the QR code to read the original article
A Western dietary pattern during pregnancy is associated with neurodevelopmental disorders in childhood and adolescence
A western dietary pattern during pregnancy is associated with neurodevelopmental disorders in childhood and adolescenceThe University of Copenhagen and partner institutions carried out a groundbreaking study that systematically clarified the biological link between a Western dietary pattern during pregnancy and neurodevelopmental disorders in offspring. Working with the COPSAC2010 cohort (n=508), the team analyzed dietary questionnaire data from pregnant women at 24 weeks of gestation together with clinical neurodevelopmental assessments of their children at age 10, finding that a Western dietary pattern in pregnancy was significantly and positively associated with the risk of attention deficit hyperactivity disorder (ADHD) and autism. Integrating data from three independent large mother-and-child cohorts (n=59,725; n=656; n=348) and combining self-reported dietary patterns with maternal blood metabolomics and fetal cord blood metabolomics, the researchers not only confirmed the association with ADHD but also identified 15 mediating metabolites that significantly improved ADHD prediction. Of particular importance, longitudinal metabolomic analysis revealed a persistent and stable pattern of association between Western diet-related metabolic features from early to mid-pregnancy and later neurodevelopmental outcomes. The study indicates that targeted prenatal dietary intervention may help prevent neurodevelopmental disorders, underlining the importance of early intervention, and offers new perspectives for future prevention and treatment strategies.
▼
Scan the QR code to read the original article
DrBioRight 2.0: a large language model-powered bioinformatics conversational system for large-scale cancer functional proteomics analysis
DrBioRight 2.0: an LLM-powered bioinformatics chatbot for large-scale cancer functional proteomics analysis
In cancer research, large-scale cancer functional proteomics analysis is essential for understanding the mechanisms of tumor initiation and progression and for identifying potential therapeutic targets. The team of Han Liang at the University of Texas MD Anderson Cancer Center used RPPA technology to analyze 8,000 tumor samples from TCGA patients and 900 CCLE cell line samples, building a comprehensive, high-quality pan-cancer functional proteomics database, RPPA 500, comprising nearly 500 high-quality antibodies covering all major cancer hallmark pathways. To improve the accessibility and analytical power of this resource, the team also launched DrBioRight 2.0, a bioinformatics platform powered by large language models (LLMs) that lets users explore protein-centric cancer omics data in natural language, run advanced analyses, visualize results and take part in interactive discussions — all through natural language. By simplifying complex proteogenomic analysis, the tool helps accelerate the translation of large-scale functional proteomics data into meaningful biomedical insights.
▼
Scan the QR code to read the original article
Prospective cohort study and genetic analysis reveal associations between 2,923 plasma proteins and the risk of inflammatory bowel disease
Associations of 2923 plasma proteins with incident inflammatory bowel disease in a prospective cohort study and genetic analysis
Proteomics holds broad but still underexploited promise in inflammatory bowel disease (IBD) research. Researchers from the Clinical Epidemiology Unit of Qilu Hospital of Shandong University and other institutions carried out an in-depth study using data from 48,800 IBD-free participants in the UK Biobank Pharma Proteomics Project (UKB-PPP), applying Cox regression to assess the associations between 2,923 plasma proteins and the risk of incident IBD. The team also performed a Mendelian randomization (MR) meta-analysis, integrating cis-protein quantitative trait locus (pQTL) data from UKB-PPP with IBD genome-wide association study (GWAS) data from the International Inflammatory Bowel Disease Genetics Consortium and the FinnGen study to identify causal proteins, while colocalization analysis further strengthened the evidence for shared causal genetic variants. The results showed that 673, 295 and 125 proteins were associated with the risk of IBD, Crohn's disease (CD) and ulcerative colitis (UC), respectively. Through MR and colocalization analysis, eight proteins (IL12B, CD6, MXRA8, CXCL9, IFNG, CCN3, RSPO3 and IL18) showed strong causal and colocalization evidence for IBD or its subtypes. These findings deepen our understanding of the molecular etiology of IBD and provide potential therapeutic targets for future intervention strategies.
▼
Scan the QR code to read the original article
Plasma proteomics-based identification of biomarkers for the polygenic risk of type 2 diabetes and its comorbidities
Identification of plasma proteomic markers underlying polygenic risk of type 2 diabetes and related comorbidities
A collaboration between AstraZeneca and the University of Cambridge has systematically characterized the plasma proteomic profile associated with the polygenic risk of type 2 diabetes (T2D) and its clinical significance. Using UK Biobank data and a genome-wide polygenic score (PGS) approach, the team analyzed 2,922 circulating proteins and constructed a molecular profile of T2D and its cardiometabolic complications. The genome-wide T2D-PGS was significantly associated with 617 plasma proteins, 75% of which were also linked to other metabolic traits. Disease-specific stratified analysis further identified 342 associated proteins, including 20% unique targets. In-depth functional analysis highlighted key pathways such as the complement cascade, and identified potential therapeutic targets such as FAM3D alongside comorbidity biomarkers including EFEMP1 and IGFBP2. To support translation of the findings, the team developed a dedicated interactive data portal (https://public.cgr.astrazeneca.com/t2d-pgs/v1/), integrating complete association analyses, causal inference, pathway enrichment and clinical validation data, providing a valuable research resource for scientists worldwide. The study not only deepens understanding of T2D pathogenesis but also lays important groundwork for new diagnostic and therapeutic strategies.
▼
Scan the QR code to read the original article
Proteomics-based organ aging signatures predict 20-year risk of age-related disease: the Whitehall II cohort study
Proteomic organ-specific ageing signatures and 20-year risk of age-related diseases: the Whitehall II observational cohort studyAccurately assessing organ-specific aging and its impact on disease risk is essential for building a precise system of prevention and control for age-related diseases. A team from the UCL Queen Square Institute of Neurology, University College London, together with the University of Helsinki and other partners, analyzed plasma samples from 6,235 middle-aged participants in a large prospective cohort, using proteomics to determine the biological age gaps of nine key organs, and followed these participants for up to 20 years to examine how organ aging relates to 45 age-related diseases and to multimorbidity. Individuals with larger organ age gaps showed significantly increased risk of 30 diseases, six of which were directly linked to accelerated aging of the corresponding organ: liver failure, dilated cardiomyopathy, chronic heart failure, lung cancer, agranulocytosis and lymph node metastasis. Larger organ age gaps were also associated with an increased risk of multi-organ multimorbidity. The study reveals the link between accelerated aging of specific organs and elevated risk of multiple diseases, underscores the importance of the biology of organ aging in disease development, and opens new perspectives for personalized prevention and treatment strategies.
▼
Scan the QR code to read the original article
A cancer driver gene discovery method that fuses large language models with omics causal inference
Cancer gene identification through integrating causal prompting large language model with omics data-driven causal inferenceMulti-omics identification of cancer driver genes is of key scientific value for revealing the mechanisms of tumorigenesis and developing precision treatment strategies. Researcher Huiyan Sun and her team at the School of Artificial Intelligence, Jilin University, developed a new framework called ICGI (Integrative Causal Gene Identification) to identify cancer-related genes from multi-omics data. The framework combines a large language model (LLM) with data-driven causal feature selection, introducing contextual prompts about causal relationships into the LLM to improve both the accuracy and the interpretability of cancer gene identification. To validate its effectiveness, the team applied ICGI to transcriptomic data from six different cancer types in The Cancer Genome Atlas and demonstrated superior performance over existing state-of-the-art methods in identifying the key genes that distinguish cancerous from normal samples. The team also built an online service platform: users enter a gene of interest and a specific cancer type, and the platform reports whether that gene plays an important role in the cancer, together with an easy-to-understand explanation and a summary of the inferences drawn from the data-driven causal learning method. This work provides a powerful new tool for cancer gene identification and advances precision medicine based on multi-omics data analysis.
▼
Scan the QR code to read the original article
