eXplainable Artificial Intelligence (XAI) in aging clock models.

XAI is a rapidly progressing field of machine learning, aiming to unravel the predictions of complex models. XAI is especially required in sensitive applications, e.g. in health care, when diagnosis, recommendations and treatment choices might rely on the decisions made by artificial intelligence systems. AI approaches have become widely used in aging research as well, in particular, in developing biological clock models and identifying biomarkers of aging and age-related diseases. However, the potential of XAI here awaits to be fully appreciated. We discuss the application of XAI for developing the "aging clocks" and present a comprehensive analysis of the literature categorized by the focus on particular physiological systems.

Epigenetics of the far northern Yakutian population.

BACKGROUND: Yakuts are one of the indigenous populations of the subarctic and arctic territories of Siberia characterized by a continental subarctic climate with severe winters, with the regular January average temperature in the regional capital city of Yakutsk dipping below - 40 °C. The epigenetic mechanisms of adaptation to such ecologies and environments and, in particular, epigenetic age acceleration in the local population have not been studied before. RESULTS: This work reports the first epigenetic study of the Yakutian population using whole-blood DNA methylation data, supplemented with the comparison to the residents of Central Russia. Gene set enrichment analysis revealed, among others, geographic region-specific differentially methylated regions associated with adaptation to climatic conditions (water consumption, digestive system regulation), aging processes (actin filament activity, cell fate), and both of them (channel activity, regulation of steroid and corticosteroid hormone secretion). Further, it is demonstrated that the epigenetic age acceleration of the Yakutian representatives is significantly higher than that of Central Russia counterparts. For both geographic regions, we showed that epigenetically males age faster than females, whereas no significant sex differences were found between the regions. CONCLUSIONS: We performed the first study of the epigenetic data of the Yakutia cohort, paying special attention to region-specific features, aging processes, age acceleration, and sex specificity.

Small immunological clocks identified by deep learning and gradient boosting.

Background: The aging process affects all systems of the human body, and the observed increase in inflammatory components affecting the immune system in old age can lead to the development of age-associated diseases and systemic inflammation. Results: We propose a small clock model SImAge based on a limited number of immunological biomarkers. To regress the chronological age from cytokine data, we first use a baseline Elastic Net model, gradient-boosted decision trees models, and several deep neural network architectures. For the full dataset of 46 immunological parameters, DANet, SAINT, FT-Transformer and TabNet models showed the best results for the test dataset. Dimensionality reduction of these models with SHAP values revealed the 10 most age-associated immunological parameters, taken to construct the SImAge small immunological clock. The best result of the SImAge model shown by the FT-Transformer deep neural network model has mean absolute error of 6.94 years and Pearson ρ = 0.939 on the independent test dataset. Explainable artificial intelligence methods allow for explaining the model solution for each individual participant. Conclusions: We developed an approach to construct a model of immunological age based on just 10 immunological parameters, coined SImAge, for which the FT-Transformer deep neural network model had proved to be the best choice. The model shows competitive results compared to the published studies on immunological profiles, and takes a smaller number of features as an input. Neural network architectures outperformed gradient-boosted decision trees, and can be recommended in the further analysis of immunological profiles.

Disease classification for whole-blood DNA methylation: Meta-analysis, missing values imputation, and XAI.

BACKGROUND: DNA methylation has a significant effect on gene expression and can be associated with various diseases. Meta-analysis of available DNA methylation datasets requires development of a specific workflow for joint data processing. RESULTS: We propose a comprehensive approach of combined DNA methylation datasets to classify controls and patients. The solution includes data harmonization, construction of machine learning classification models, dimensionality reduction of models, imputation of missing values, and explanation of model predictions by explainable artificial intelligence (XAI) algorithms. We show that harmonization can improve classification accuracy by up to 20% when preprocessing methods of the training and test datasets are different. The best accuracy results were obtained with tree ensembles, reaching above 95% for Parkinson's disease. Dimensionality reduction can substantially decrease the number of features, without detriment to the classification accuracy. The best imputation methods achieve almost the same classification accuracy for data with missing values as for the original data. XAI approaches have allowed us to explain model predictions from both populational and individual perspectives. CONCLUSIONS: We propose a methodologically valid and comprehensive approach to the classification of healthy individuals and patients with various diseases based on whole-blood DNA methylation data using Parkinson's disease and schizophrenia as examples. The proposed algorithm works better for the former pathology, characterized by a complex set of symptoms. It allows to solve data harmonization problems for meta-analysis of many different datasets, impute missing values, and build classification models of small dimensionality.

Accelerated epigenetic aging and inflammatory/immunological profile (ipAGE) in patients with chronic kidney disease.

Chronic kidney disease (CKD) is defined by a reduced estimated glomerular filtration rate (eGFR). This failure can be related to a phenotype of accelerated aging. In this work, we considered 76 patients with end-stage renal disease (ESRD) and 83 healthy controls. We concomitantly evaluated for the first time two measures that can be informative of the rate of aging, i.e., whole blood DNA methylation using the Illumina Infinium EPIC array and plasma levels of a selection of inflammatory/immunological proteins using multiplex immunoassays. First of all, we demonstrated accelerated aging in terms of the most common epigenetic age estimators in CKD patients. Moreover, we developed a new clock/predictor of age based on the inflammatory/immunological profile (ipAGE) and identified the inflammatory/immunological biomarkers differentially expressed between cases and controls. IpAGE appeared to be more sensitive than epigenetic clocks in quantifying the accelerated aging phenotype of ESRD patients. Interestingly, we did not find any correlation between the age acceleration evaluated according to the epigenetic clocks and ipAGE in either the ESRD group or the control group. On the whole, our data show a consistent accelerated aging phenotype in ESRD patients, which is better appreciated by quantifying the underlying inflammatory processes (inflammaging) by ipAGE than by using epigenetic clocks.

A Meta-Analysis of Brain DNA Methylation Across Sex, Age, and Alzheimer's Disease Points for Accelerated Epigenetic Aging in Neurodegeneration.

Alzheimer's disease (AD) is characterized by specific alterations of brain DNA methylation (DNAm) patterns. Age and sex, two major risk factors for AD, are also known to largely affect the epigenetic profiles in brain, but their contribution to AD-associated DNAm changes has been poorly investigated. In this study we considered publicly available DNAm datasets of four brain regions (temporal, frontal, entorhinal cortex, and cerebellum) from healthy adult subjects and AD patients, and performed a meta-analysis to identify sex-, age-, and AD-associated epigenetic profiles. In one of these datasets it was also possible to distinguish 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC) profiles. We showed that DNAm differences between males and females tend to be shared between the four brain regions, while aging differently affects cortical regions compared to cerebellum. We found that the proportion of sex-dependent probes whose methylation is modified also during aging is higher than expected, but that differences between males and females tend to be maintained, with only a few probes showing age-by-sex interaction. We did not find significant overlaps between AD- and sex-associated probes, nor disease-by-sex interaction effects. On the contrary, we found that AD-related epigenetic modifications are significantly enriched in probes whose DNAm varies with age and that there is a high concordance between the direction of changes (hyper or hypo-methylation) in aging and AD, supporting accelerated epigenetic aging in the disease. In summary, our results suggest that age-associated DNAm patterns concur to the epigenetic deregulation observed in AD, providing new insights on how advanced age enables neurodegeneration.

Transforming Lindblad Equations into Systems of Real-Valued Linear Equations: Performance Optimization and Parallelization of an Algorithm.

With their constantly increasing peak performance and memory capacity, modern supercomputers offer new perspectives on numerical studies of open many-body quantum systems. These systems are often modeled by using Markovian quantum master equations describing the evolution of the system density operators. In this paper, we address master equations of the Lindblad form, which are a popular theoretical tools in quantum optics, cavity quantum electrodynamics, and optomechanics. By using the generalized Gell-Mann matrices as a basis, any Lindblad equation can be transformed into a system of ordinary differential equations with real coefficients. Recently, we presented an implementation of the transformation with the computational complexity, scaling as O(N5logN) for dense Lindbaldians and O(N3logN) for sparse ones. However, infeasible memory costs remains a serious obstacle on the way to large models. Here, we present a parallel cluster-based implementation of the algorithm and demonstrate that it allows us to integrate a sparse Lindbladian model of the dimension N=2000 and a dense random Lindbladian model of the dimension N=200 by using 25 nodes with 64 GB RAM per node.

Age-related DNA methylation changes are sex-specific: a comprehensive assessment.

The existence of a sex gap in human health and longevity has been widely documented. Autosomal DNA methylation differences between males and females have been reported, but so far few studies have investigated if DNA methylation is differently affected by aging in males and females. We performed a meta-analysis of 4 large whole blood datasets, comparing 4 aspects of epigenetic age-dependent remodeling between the two sexes: differential methylation, variability, epimutations and entropy. We reported that a large fraction (43%) of sex-associated probes undergoes age-associated DNA methylation changes, and that a limited number of probes show age-by-sex interaction. We experimentally validated 2 regions mapping in FIGN and PRR4 genes and showed sex-specific deviations of their methylation patterns in models of decelerated (centenarians) and accelerated (Down syndrome) aging. While we did not find sex differences in the age-associated increase in epimutations and entropy, we showed that the number of probes having an age-related increase in methylation variability is 15 times higher in males compared to females. Our results can offer new epigenetic tools to study the interaction between aging and sex and can pave the way to the identification of molecular triggers of sex differences in longevity and age-related diseases prevalence.