Browse
Submit Data
Databases
API
Help

Dataset Information

0 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

Accurate age prediction from blood using a small set of DNA methylation sites and a cohort-based machine learning algorithm

ABSTRACT: Chronological age prediction from DNA methylation sheds light on human aging, indicates poor health and predicts lifespan. Previous studies developed methylation clocks based on linear regression models on methylation array data. While accurate, these models are limited to fixed-rate changes in methylation levels across age. Moreover, the high cost of methylation arrays, compared to targeted-PCR sequencing, hinders widespread utility of such predictors. We present an AI-based alternative termed GP-age, which uses a non-parametric approach based on Gaussian Process Regression of a large cohort of ~12K blood methylomes. Given a new blood sample, methylation levels are compared to the cohort samples, which are then weighted to predict the query age. Using only 30 CpG sites, our approach outperforms state-of-the-art methylation clocks that use hundreds of sites, with a median error of 2.1 years (on held-out data). Our model was also applied to sequencing-based data yielding highly accurate predictions. Overall, we provide an accessible alternative to current array-based methylation clocks, with future applications in aging research, forensic profiling, and monitoring epigenetic processes in transplantation medicine and cancer.

ORGANISM(S): Homo sapiens

PROVIDER: GSE207605 | GEO | 2023/01/20

REPOSITORIES: GEO

ACCESS DATA

Json Xml

Dataset's files

Source:

			Action	DRS
		Other

Items per page:

1 - 1 of 1

Similar Datasets

A multi-tissue full lifespan epigenetic clock for mice [II]

Project description:Human DNA-methylation data have been used to develop highly accurate biomarkers of aging ("epigenetic clocks"). Recent studies demonstrate that similar epigenetic clocks for mice (Mus Musculus) can be slowed by gold standard anti-aging interventions such as calorie restriction and growth hormone receptor knock-outs. Using DNA methylation data from previous publications with data collected in house for a total 1189 samples spanning 193,651 CpG sites, we developed 4 novel epigenetic clocks by choosing different regression models (elastic net- versus ridge regression) and by considering different sets of CpGs (all CpGs vs highly conserved CpGs). We demonstrate that accurate age estimators can be built on the basis of highly conserved CpGs. However, the most accurate clock results from applying elastic net regression to all CpGs. While the anti-aging effect of calorie restriction could be detected with all types of epigenetic clocks, only ridge regression based clocks replicated the finding of slow epigenetic aging effects in dwarf mice. Overall, this study demonstrates that there are trade-offs when it comes to epigenetic clocks in mice. Highly accurate clocks might not be optimal for detecting the beneficial effects of anti-aging interventions.

2018-11-29 | GSE120136 | GEO

A multi-tissue full lifespan epigenetic clock for mice [I]

2018-11-29 | GSE120132 | GEO

Development of a novel aging clock based on chromatin accessibility [RNA-seq]

Project description:The establishment of highly accurate aging clocks based on DNA methylation highlighted the strong link between epigenetic alterations and aging. However, connecting methylation clocks to physiological changes is not straightforward. Transcriptomics and proteomics clocks on the other hand are directly connected to cellular function, yet they do not allow us to understand underlying epigenetic mechanisms. We hypothesize that chromatin accessibility, a readout that integrates many epigenetic mechanisms, may bridge epigenetic changes with their downstream effects in aging. To demonstrate that chromatin accessibility can predict age, we generated chromatin accessibility profiles from peripheral blood mononuclear cells (PBMCs) of 157 human donors, aged 20 to 74 and used them to construct a novel aging clock. Our clock predicted age with a root mean squared error of 7.69 years, a median absolute error of 5.69 years and a correlation of 0.83. Moreover, by comparing our chromatin accessibility data to matched transcriptomic profiles, we show that the genomic sites selected by our clock drive changes in transcription during aging. This chromatin accessibility clock could therefore be used to better understand the mechanisms driving the process of aging and evaluate anti-aging interventions.

2022-12-01 | GSE193141 | GEO

Development of a novel aging clock based on chromatin accessibility [ATAC-seq]

2022-12-01 | GSE193140 | GEO

BayesAge: A Maximum Likelihood Algorithm To Predict Epigenetic Age

Project description:DNA methylation is a reaction that results in the formation of 5-methylcytosine when a methyl group is added to the cytosine’s C5 position. As organisms age, DNA methylation patterns change in a reproducible fashion. This phenomenon has established DNA methylation as a valuable biomarker in aging studies. Epigenetic clocks based on weighted combinations of methylation sites have been developed to accurately predict the age of an individual from their methylome. However, many epigenetic clocks, particularly those that utilize penalized regression, model the changes in methylation linearly with age. Moreover, these models, which use methylation levels as features, are not robust to missing data and do not account for the count-based nature of bisulfite sequence data. Additionally, the models are generally not interpretable. To overcome these challenges, we present BayesAge, an extension of the previously developed scAge approach that was developed for the analysis of single cell DNA methylation datasets. BayesAge utilizes maximum likelihood estimation (MLE) to infer ages, models count data using binomial distributions, and uses LOWESS smoothing to capture the non-linear dynamics between methylation and age. Our approach is designed for use with bulk bisulfite sequencing datasets. BayesAge outperforms scAge in several respects. Specifically, BayesAge’s age residuals are not age associated, thus providing a less biased representation of epigenetic age variation across populations. Moreover, BayesAge enables the estimation of error bounds on age inference and, when run on downsampled data, its coefficient of determination between predicted and actual ages surpasses both scAge and penalized regression.

2024-03-20 | GSE261769 | GEO

Age prediction from human blood plasma using proteomic and small RNA data: a comparative analysis

Project description:Aging clocks, built from comprehensive molecular data, have emerged as promising tools in medicine, forensics, and ecological research. However, few studies have compared the suitability of different molecular data types to predict age in the same cohort and whether combining them would improve predictions. Here, we explored this at the level of proteins and small RNAs in 103 human blood plasma samples. First, we used a two-step mass spectrometry approach measuring 612 proteins to select and quantify 21 proteins that changed in abundance with age. Notably, proteins increasing with age were enriched for components of the complement system. Next, we used small RNA sequencing to select and quantify a set of 315 small RNAs that changed in abundance with age. Most of these were microRNAs (miRNAs), downregulated with age, and predicted to target genes related to growth, cancer, and senescence. Finally, we used the collected data to build age-predictive models. Among the different types of molecules, proteins yielded the most accurate model (R² = 0.59 ± 0.02), followed by miRNAs as the best-performing class of small RNAs (R² = 0.54 ± 0.02). Interestingly, the use of protein and miRNA data together improved predictions (R2 = 0.70 ± 0.01). Future work using larger sample sizes and a validation dataset will be necessary to confirm these results. Nevertheless, our study suggests that combining proteomic and miRNA data yields superior age predictions, possibly by capturing a broader range of age-related physiological changes. It will be interesting to determine if combining different molecular data types works as a general strategy to improve future aging clocks.

2023-07-03 | PXD028295 | Pride

Age prediction from human blood plasma using proteomic and small RNA data: a comparative analysis

2023-07-03 | PXD028281 | Pride

A whole lifespan mouse multi-tissue DNA methylation clock

Project description:Age predictors based on DNA methylation levels at a small set of CpG sites, DNAm clocks, have been developed for humans and extended to several other species. Three currently available versions of mouse DNAm clocks were either created for individual tissues or tuned towards young ages. Here, we constructed a robust multi-tissue age predictor based on 435 CpG sites, which covers the entire mouse lifespan and remains unbiased with respect to any particular age group. It can successfully detect the effects of several lifespan-modulating interventions on biological age as well as the rejuvenation effect related to the transition from fibroblasts to iPSCs. We have carried out comparative analyses of available mouse DNAm clocks, which revealed their broad applicability, but also certain limitations to the use of tissue-specific and multi-tissue age predictors. Together, these tools should help address diverse questions in aging research.

2018-11-20 | GSE121141 | GEO

Methylation arrays (MethylationEPIC) of peripheral whole blood samples from adults aged between 19-96 years

Project description:Epigenetic remodeling is central to understanding the molecular mechanisms that occur and drive human aging. In fact, DNA methylation clocks are capable of estimating biological age, which determines human lifespan and healthspan. Here, using Illumina's MethylationEPIC array technology, we have profiled the DNA methylation landscape of peripheral whole blood samples from adult individuals ranging in age from 19 to 96 years.

2024-09-29 | E-MTAB-13431 | biostudies-arrayexpress

Development of an Epigenetic Clock Resistant to Changes in Immune Composition

Project description:Epigenetic clocks are age predictors that use machine-learning models trained on DNA CpG methylation values to predict chronological or biological age. Increases in predicted epigenetic age relative to chronological age (epigenetic age acceleration) are connected to aging-associated pathologies, and changes in epigenetic age are linked to canonical aging hallmarks. However, epigenetic clocks rely on training data from bulk tissues whose cellular composition changes with age. We found that human naive CD8+ T cells, which decrease during aging, exhibit an epigenetic age 15–20 years younger than effector memory CD8+ T cells from the same individual. Importantly, homogenous naive T cells isolated from individuals of different ages show a progressive increase in epigenetic age, indicating that current epigenetic clocks measure two independent variables, aging and immune cell composition. To isolate the age-associated cell intrinsic changes, we created a new clock, the IntrinClock, that did not change among 10 immune cell types tested. IntrinClock showed a robust predicted epigenetic age increase in a model of replicative senescence in vitro and age reversal during OSKM-mediated reprogramming

2023-12-27 | GSE252045 | GEO

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data