AgeML: Age modelling with Machine Learning DOI Creative Commons
Jorge Garcia Condado, Iñigo Tellaetxe Elorriaga, Jesús M. Cortés

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: May 5, 2024

An approach to age modeling involves the supervised prediction of using machine learning from subject features. The derived metrics are used study relationship between healthy and pathological aging in multiple body systems, as well interactions them. We lack a standard for this type modeling. In work we developed AgeML, an OpenSource software age-prediction any tabular clinical data following well-established tested methodologies. objective is set standards reproducibility standardization reporting tasks. AgeML does modeling, calculates deltas, difference predicted chronological age, measures correlations deltas factors, visualizes differences different populations classifies based on deltas. With able reproduce published unveil novel relationships organs polygenetic risk scores. made easy reproducibility.

Language: Английский

The Genetic Architecture of Biological Age in Nine Human Organ Systems DOI Creative Commons
Junhao Wen, Ye Tian,

Ioanna Skampardoni

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: June 12, 2023

Abstract Understanding the genetic basis of biological aging in multi-organ systems is vital for elucidating age-related disease mechanisms and identifying therapeutic interventions. This study characterized architecture age gap (BAG) across nine human organ 377,028 individuals European ancestry from UK Biobank. We discovered 393 genomic loci-BAG pairs (P-value<5×10 -8 ) linked to brain, eye, cardiovascular, hepatic, immune, metabolic, musculoskeletal, pulmonary, renal systems. observed BAG-organ specificity inter-organ connections. Genetic variants associated with BAGs are predominantly specific respective system while exerting pleiotropic effects on traits multiple A gene-drug-disease network confirmed involvement metabolic BAG-associated genes drugs targeting various disorders. correlation analyses supported Cheverud’s Conjecture 1 – between mirrors their phenotypic correlation. causal revealed potential linking chronic diseases (e.g., Alzheimer’s disease), body weight, sleep duration BAG Our findings shed light promising interventions enhance health within a complex network, including lifestyle modifications drug repositioning strategies treating diseases. All results publicly available at https://labs-laboratory.com/medicine .

Language: Английский

Citations

12
Unraveling the genetic architecture of blood unfolded p-53 among non-demented elderlies: novel candidate genes for early Alzheimer's disease DOI Creative Commons
Arash Yaghoobi, Seyed Amir Malekpour

BMC Genomics, Journal Year: 2024, Volume and Issue: 25(1)

Published: May 3, 2024

Abstract Background Alzheimer's disease (AD) is a heritable neurodegenerative whose long asymptomatic phase makes the early diagnosis of it pivotal. Blood U-p53 has recently emerged as superior predictive biomarker for AD in stages. We hypothesized that genetic variants associated with blood could reveal novel loci and pathways involved stages AD. Results performed Genome-wide association study (GWAS) on 484 healthy mild cognitively impaired subjects from ADNI cohort using 612,843 Single nucleotide polymorphisms (SNPs). pathway analysis prioritized candidate genes an single-cell gene program. fine-mapped intergenic SNPs by leveraging cell-type-specific enhancer-to-gene linking strategy brain multimodal dataset. validated independent RNA-seq transcriptome datasets. The rs279686 between AASS FEZF1 was most significant SNP ( p -value = 4.82 × 10 –7 ). Suggestive were related to immune nervous systems. Twenty-three at 27 suggestive loci. Fine-mapping 5 yielded nine cell-specific genes. Finally, 15 dataset, five Conclusions underlined importance performing GWAS early-stage functional omics datasets pinpointing causal Our (SORCS1, KIF5C, TMEFF2, TMEM63C, HLA-E, ATAT1, TUBB, ARID1B, RUNX1) strongly implicated

Language: Английский

Citations

2
Neuroimaging-AI Endophenotypes of Brain Diseases in the General Population: Towards a Dimensional System of Vulnerability DOI Creative Commons
Junhao Wen,

Ioanna Skampardoni,

Ye Tian

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: Aug. 24, 2023

Abstract Disease heterogeneity poses a significant challenge for precision diagnostics in both clinical and sub-clinical stages. Recent work leveraging artificial intelligence (AI) has offered promise to dissect this by identifying complex intermediate phenotypes – herein called dimensional neuroimaging endophenotypes (DNEs) which subtype various neurologic neuropsychiatric diseases. We investigate the presence of nine such DNEs derived from independent yet harmonized studies on Alzheimer’s disease (AD1-2) 1 , autism spectrum disorder (ASD1-3) 2 late-life depression (LLD1-2) 3 schizophrenia (SCZ1-2) 4 general population 39,178 participants UK Biobank study. Phenome-wide associations revealed prominent between related brain other human organ systems. This phenotypic landscape aligns with SNP-phenotype genome-wide associations, revealing 31 genomic loci associated (Bonferroni corrected P-value < 5×10 -8 /9). The exhibited genetic correlations, colocalization, causal relationships multiple systems chronic A effect (odds ratio=1.25 [1.11, 1.40], P-value=8.72×10 -4 ) was established AD2, characterized focal medial temporal lobe atrophy, AD. their polygenic risk scores significantly improved prediction accuracy 14 systemic categories mortality. These findings underscore potential identify individuals at high developing four diseases during preclinical stages diagnostics. All results are publicly available at: http://labs.loni.usc.edu/medicine/ .

Language: Английский

Citations

6
Accelerated brain age in young to early middle-aged adults after mild to moderate COVID-19 infection DOI Creative Commons
Shelli R. Kesler, Oscar Y. Franco‐Rocha, Alexa De La Torre Schutz

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: March 7, 2024

Abstract Cognitive decline is a common adverse effect of the Coronavirus Disease 2019 (COVID-19), particularly in post-acute disease phase. The mechanisms cognitive impairment after COVID-19 (COGVID) remain unclear, but neuroimaging studies provide evidence brain changes, many that are associated with aging. Therefore, we calculated Brain Age Gap (BAG), which difference between age and chronological age, cohort 25 mild to moderate survivors (did not experience breathlessness, pneumonia, or respiratory/organ failure) 24 non-infected controls (mean = 30 +/− 8) using magnetic resonance imaging (MRI). BAG was significantly higher group (F 4.22, p 0.046) by 2.65 years. Additionally, 80% demonstrated an accelerated compared 13% control (X 2 20.0, < 0.001). Accelerated correlated lower function (p 0.041). Females 99% decreased risk males (OR 0.015, 95% CI: 0.001 0.300). There also small (1.4%) significant decrease for longer time since diagnosis 0.986, 0.977 0.995). Our findings novel biomarker COGVID point aging as potential mechanism this effect. results offer further insight regarding gender-related disparities morbidity COVID-19.

Language: Английский

Citations

1
Five dominant dimensions of brain aging are identified via deep learning: associations with clinical, lifestyle, and genetic measures DOI Creative Commons
Zhijian Yang, Junhao Wen, Güray Erus

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: Dec. 30, 2023

Abstract Brain aging is a complex process influenced by various lifestyle, environmental, and genetic factors, as well age-related often co-existing pathologies. MRI and, more recently, AI methods have been instrumental in understanding the neuroanatomical changes that occur during large diverse populations. However, multiplicity mutual overlap of both pathologic processes affected brain regions make it difficult to precisely characterize underlying neurodegenerative profile an individual from scan. Herein, we leverage state-of-the art deep representation learning method, Surreal-GAN, present methodological advances extensive experimental results allow us elucidate heterogeneity cohort 49,482 individuals 11 studies. Five dominant patterns neurodegeneration were identified quantified for each their respective (herein referred as) R-indices. Significant associations between R-indices distinct biomedical, factors provide insights into etiology observed variances. Furthermore, baseline showed predictive value disease progression mortality. These five contribute MRI-based precision diagnostics, prognostication, may inform stratification clinical trials.

Language: Английский

Citations

2
AgeML: Age modelling with Machine Learning DOI Creative Commons
Jorge Garcia Condado, Iñigo Tellaetxe Elorriaga, Jesús M. Cortés

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: May 5, 2024

An approach to age modeling involves the supervised prediction of using machine learning from subject features. The derived metrics are used study relationship between healthy and pathological aging in multiple body systems, as well interactions them. We lack a standard for this type modeling. In work we developed AgeML, an OpenSource software age-prediction any tabular clinical data following well-established tested methodologies. objective is set standards reproducibility standardization reporting tasks. AgeML does modeling, calculates deltas, difference predicted chronological age, measures correlations deltas factors, visualizes differences different populations classifies based on deltas. With able reproduce published unveil novel relationships organs polygenetic risk scores. made easy reproducibility.

Language: Английский

Citations

0