Article

ISSN 0006-2979, Biochemistry (Moscow), 2024, Vol. 89, No. 2, pp. 313-321 © The Author(s) 2024. This article is an open access publication.

313

AgeMeta: Quantitative Gene Expression Database

of Mammalian Aging

Stanislav Tikhonov

1,2

, Mikhail Batin

, Vadim N. Gladyshev

Sergey E. Dmitriev

1,2

, and Alexander Tyshkovskiy

1,4,a

Belozersky Institute of Physico-Chemical Biology, Lomonosov Moscow State University, 119234 Moscow, Russia

Faculty of Bioengineering and Bioinformatics, Lomonosov Moscow State University, 119234 Moscow, Russia

Open Longevity, 15260 Ventura Blvd, STE 2230, Sherman Oaks, CA 91403, USA

Division of Genetics, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School,

Boston, MA 02115, USA

e-mail: atyshkovskii@bwh.harvard.edu

Received October 12, 2023

Revised December 12, 2023

Accepted February 13, 2024

Abstract— AgeMeta is a database that provides systemic and quantitative description of mammalian aging at

the level of gene expression. It encompasses transcriptomic changes with age across various tissues of humans,

mice, and rats, based on a comprehensive meta-analysis of 122 publicly available gene expression datasets from

26 studies. AgeMeta provides an intuitive visual interface for quantification of aging-associated transcriptomics

at the level of individual genes and functional groups of genes, allowing easy comparison among various species

and tissues. Additionally, all the data in the database can be downloaded and analyzed independently. Overall, this

work contributes to the understanding of the complex network of biological processes underlying mammalian

aging and supports future advancements in this field. AgeMeta is freely available at: https://age-meta.com/.

DOI: 10.1134/S000629792402010X

Keywords: aging, ageing, gene expression, differential expression, transcriptomics, mammalian, meta-analysis,

AgeMeta, age-meta, database, GSEA

* To whom correspondence should be addressed.

DEDICATION

We dedicate this work to the cherished memory of

Vladimir P. Skulachev – our esteemed teacher, mentor,

colleague, and extraordinary scientist. His profound

influence extended beyond the confines of academia,

as he played a pivotal role in orchestrating scientific

endeavors. Academician Skulachev is renowned for

his groundbreaking contributions to the fields of cell

bioenergetics and mitochondrial function. His unwav-

ering passion has, in recent years, also encompassed

the biology of aging, yielding many remarkable dis-

coveries. Academician Skulachev’s legacy is not only

defined by his scientific achievements but also by

the enduring impression he left through his bril-

liance, wisdom, kindness, and remarkable character.

In our hearts, he will forever occupy a special place,

his memory serving as an inspiration for current and

future generations.

INTRODUCTION

Aging is the major risk factor for multiple chronic

diseases and, therefore, poses a global medical prob-

lem [1]. Understanding the molecular basis and in-

ter-species differences of the aging process is thus a

crucial step towards developing effective geroprotec-

tors and overcoming the difficulties aging brings to

our society. The systems approach to characterize ag-

ing has gained popularity in the recent years with the

advent of high-throughput sequencing and due to its

comprehensiveness, resulting in a plethora of omics

data available and a great demand for its aggregation

into databases and meta-analysis.

TIKHONOV et al.314

BIOCHEMISTRY (Moscow) Vol. 89 No. 2 2024

To date, a number of databases containing ag-

ing genomics and transcriptomics data exist, includ-

ing Aging Atlas [2], HAGR [3], Digital Ageing Atlas [4],

AGEMAP [5], and Open Genes [6]. Most of these data-

bases accommodate plenty of information in addition

to aging transcriptomics, e.g., lifespan-extending inter-

ventions, aging phenotype, and epigenetics, but when

it comes to gene expression, the information is most-

ly a collection of age-related differentially expressed

genes identified from different studies. To the best of

our knowledge, there are currently no databases that

would present meta-analysis-derived quantitative

general trends of gene expression changes with age,

including species-specific, tissue-specific, and global

trends, allowing easy comparison and visualization

for each gene at the level of individual datasets and

individual samples within each dataset. To fill this gap,

we developed the AgeMeta database that provides an

interactive visual interface to investigate these bio-

markers in particular tissues and mammalian species

as well as cellular functions enriched in the transcrip-

tomic aging signatures, both at the level of general

trends and individual datasets. To facilitate the use of

our resource, we developed a comprehensive manual

for each section of the database and made the data-

base contents downloadable.

MATERIALS AND METHODS

The full list of datasets used to construct the tran-

scriptomic signatures of mammalian aging integrated

in AgeMeta included the following study IDs from GEO

[7], ArrayExpress [8], and SRA [9] databases: GSE9103,

GSE123981, GSE3150, GSE6591, GSE74463, GSE53960,

GSE66715, GSE11291, GSE34378, GSE27625, GSE12480,

GSE36192, GSE1572, GSE28422, GSE25941, GSE53890,

GSE38718, GSE674, GSE17612, GSE21935, GSE362,

GSE132040, E-MTAB-3374, PRJNA281127, PRJNA516151.

In addition to these studies, we also used data from

GTEx [10]. The datasets contained RNA-Seq and mi-

croarray samples obtained from different tissues of

Mus musculus, Rattus norvegicus, and Homo sapiens.

The tissue content varied across species; 23 tissues

used across all three species included adipose tissue

[in certain cases labeled specifically as brown adipose

tissue (BAT), mesenteric adipose tissue (MAT), gonadal

adipose tissue (GAT) and subcutaneous adipose tissue

(SCAT)], adrenal gland, blood vessel, bone and bone

marrow, brain (in certain cases olfactory bulbs were

labeled separately as OB and cerebellum and fron-

tal cortex were labeled separately), esophagus, heart,

kidney, liver, lung, muscle, nerve, pancreas, pituitary,

prostate, salivary gland, skin, small intestine, spleen,

testis, thyroid and whole blood (in some cases white

blood cells (WBC) were labeled separately). An anno-

tation of every dataset used in this study, that includes

information about the number of samples, age range,

strain, tissue, and sex, can be found in the supplemen-

tary materials (TableS1 in the Online Resource1).

The data preprocessing pipelines for RNA-Seq and

microarray datasets were described in detail in [11].

For both types of data, all platform IDs of genes were

standardized to Entrez IDs and mapped to mouse one-

to-one orthologs (in the case of rat and human samples).

These two preprocessing steps resulted in some genes

being discarded from the data. Genes that were not de-

tected in more than 40% of the datasets used to build a

particular signature were also filtered out prior to the

corresponding meta-analysis. Consequently, if a given

gene has no entry in AgeMeta, this could have been

caused by either of the above-stated data processing

steps.

The gene expression log fold changes (logFC) uti-

lized to construct the signatures of aging were calcu-

lated as the slope coefficient of a linear model when

the sample ages included more than two unique values

and as the difference in means otherwise, using the

limmaR package [12]. Mixed-effect model was utilized

to aggregate the expression changes from all datasets

into aging signatures with the metafor R package [13].

The logFCs and their respective standard errors were

used as an input, and study ID, tissue, and species were

introduced into the model as random terms. Gene set

enrichment analysis was performed using the fgseaR

package [14] with 10,000 permutations for the individ-

ual datasets and the signatures of aging. The AgeMeta

website was created using the shiny R package [15].

Overview of the 7 transcriptomic signatures of aging

Criterion Human Mouse Rat Brain Muscle Liver Global

Number of datasets 51 52 19 25 17 11 122

Total number of genes 15,959 13,459 10,382 16,125 15,339 12,805 15,170

Number of statistically significant

age-associated genes (p adjusted<0.05)

649 184 501 1526 379 1219 7

AgeMeta: GENE EXPRESSION IN AGING 315

BIOCHEMISTRY (Moscow) Vol. 89 No. 2 2024

DATABASE CONTENT

The contents of the database are based on the

7 transcriptomic signatures of aging identified in our

recent study [11]. The transcriptomic signatures rep-

resent general trends of gene expression changes with

age and were constructed by the meta-analysis of 122

datasets from 26 studies (table). Tissue-specific signa-

tures of brain, skeletal muscle, and liver were derived

from the corresponding human, mouse, and rat data-

sets, while species-specific signatures summarize the

data across multiple different tissues of a certain spe-

cies. The global signature encompasses all of the data

from the three species and 23 tissues (see “Materials

and Methods” section).

The database consists of 5 major modules (Fig. 1).

The first one is the comparative signature interface,

which provides aggregated quantitative estimates of

the age-related expression changes (normalized logFC)

in a particular tissue and species along with the cor-

responding metrics of statistical significance (p-value

before and after the Benjamini–Hochberg adjustment

[16]) for every gene. The user can choose the main sig-

nature to work with, and sort and filter genes accord-

ing to their significance in the chosen signature. One

can easily compare the differential expression values

of chosen genes and their significance between all

7signatures of aging using the right part of the inter-

active table, where the colored arrows next to logFC val-

ues illustrate the direction of gene expression changes

and the cell hues show statistical significance (Fig. 2).

The side menu allows the user to filter genes based on

their presence and significance in all 7 signatures and

on whether they are up- or downregulated with age

according to each of the signatures.

Should one wish to inspect the origin of each

logFC value in the interactive table, one can click the

table cell with the value, which would direct them to

the mixed-effects model plot corresponding to the gene

and signature represented by the clicked cell (Fig. 3a).

The mixed-effects model was used to aggregate dif-

ferential expression data from individual datasets

into signatures (see “Materials and Methods” section).

In the plot, each point stands for a dataset, namely a

data point with a unique combination of species, tis-

sue, sex, strain, and study. The dataset specifications

are displayed for each point on mouse hover. The red

line shows the aggregated average logFC value for a

given gene presented in the comparative signature ta-

ble (this value is the output of the mixed-effects mod-

el), while the y-coordinate of each point shows the

logFC from the corresponding dataset. The color and

shape of the points can be set to discriminate between

species, tissues, studies, sexes, platforms, and types of

models used for obtaining the differential expression

values, which allows the user to tailor the plot to their

specific needs. A click of a point in this plot directs the

user to the dataset expression plot, which shows how

the expression of the selected gene changes with age

in the clicked dataset (Fig.3b). The goal of the dataset

expression plot is to display the raw data, utilized to

calculate the logFC value and its standard error for a

particular dataset.

The last two modules of the database essentially

mirror the first two, but the genes are replaced with

the functional groups of genes. Functional enrichment

for the signatures and datasets was performed using

Gene Set Enrichment Analysis [17, 18], where quanti-

tative normalized enrichment scores (NES) and Ben-

jamini–Hochberg adjusted p-values were calculated.

In the functional signature interface module, one can

compare enrichment scores and corresponding statis-

tical significances across all 7 signatures. The function-

al terms can be filtered based on their presence, statis-

tical significance, up- or downregulation according to

the signatures, as well as on the ontology these terms

came from. The utilized ontologies included GO Biolog-

ical Process [19, 20], KEGG [21-23], Reactome [24], and

Biocarta [25]. Since many functional groups of genes

are nested within each other, we introduced a feature

that allows one to simultaneously examine all parent

and child gene sets for a selected functional term, i.e.,

groups that contain the selected gene set or are sub-

sets of this gene set, respectively. Clicking a function-

al term name will reduce the functional signatures

interface table, keeping only the selected term and its

parent and child terms. The “Back to full table” button

under the table will reset the table to the default state

with all functional groups from the database. Clicking

a table cell navigates the user to the functional dataset

plot, which displays enrichment scores and statistical

significances for all individual datasets corresponding

to the clicked functional term and signature (Fig. 4).

Although aggregated functional enrichment scores of

signatures were not obtained from the scores of the

individual datasets, this plot provides a general idea

about the consistency of the age-related up- or down-

regulation of genes associated with the selected path-

way across datasets and helps to justify the values in

the interactive table. Each bar in the plot represents

one dataset, with the dataset specifications shown on

mouse hover. Green and red bars represent statistical-

ly significant up- or downregulation with age, respec-

tively, while gray bars stand for enrichment scores

that failed to cross the statistical significance threshold

set in the side menu (by default, the adjusted p-value

threshold is equal to 0.1).

To make the use of the database simple and con-

venient, we developed the “Manual” section where

each database module is explained in detail with ex-

ample tables and plots. In addition, the whole database

is available in two languages: English and Russian.

TIKHONOV et al.316

BIOCHEMISTRY (Moscow) Vol. 89 No. 2 2024

Fig. 1. The structure of the AgeMeta database. AgeMeta provides a visual interface for the 7 transcriptomic signatures of aging,

consisting of two interactive tables which display quantitative age-related transcriptomic changes for individual genes (the com-

parative signature interface) and functional groups of genes (the functional signature interface) and allow comparison between

the signatures; as well as three types of plots depicting the differential expression of a gene in a signature at the dataset level

(the mixed-effects model plot), the change of expression with age for a gene within an individual dataset (the dataset expres-

sion plot), and enrichment scores of a functional group of genes for individual datasets associated with a certain signature (the

functional dataset plot). Clicking the table cells navigates the user to the corresponding plots, while clicking the points in the

mixed-effects model plot directs the user to the dataset expression plot related to the clicked dataset. Both tables can be down-

loaded in a separate section of the database.

Fig. 2. The comparative signature interface table. The arrow colors represent the direction of gene expression changes in a given

signature (the signatures are labeled in the column names), the cell hues represent the statistical significance of those chang-

es. Adjusted p-value less than the soft threshold results in a dull hue, adjusted p-value less than the strict threshold results in

a brighter hue. Both thresholds can be set in the menu on the left-hand side of the page. All cells in the right part of the table

with the signature logFCs are clickable. Clicking them transfers the user to the corresponding mixed-effects model plot.

AgeMeta: GENE EXPRESSION IN AGING 317

BIOCHEMISTRY (Moscow) Vol. 89 No. 2 2024

Fig. 3. Examples of the mixed-effects model plot and the differential gene expression plot. a)An example of the mixed-effects

model plot. Each point represents a single dataset. Shape and color represent species and tissue, respectively. The data are mean

normalized slope/logFC ±SE. The details of each dataset are specified in the text box that appears on mouse hover. b)An exam-

ple of the dataset gene expression plot corresponding to a point clicked on the mixed-effects model plot. The slope of the regres-

sion line corresponds to the logFC or slope value for the gene in the clicked dataset (the y-axis coordinate of the clicked point),

while the width of the shaded area around the line reflects the standard error for the slope.

To allow independent analysis of our data, we made

the differential gene expression data and the function-

al enrichment data fully downloadable. They can be

accessed through the “Downloads” section.

DISCUSSION

With the use of AgeMeta, the aging-associated dif-

ferential expression patterns and their inter-species or

inter-tissue differences can be easily visualized for any

of the thousands of genes or functional groups present

in the database. The interactive tables and plots allow

users to study genes of interest at the transcriptom-

ic signature level as well as at the level of individual

datasets. Additionally, the application of these and oth-

er related transcriptomic signatures for prediction and

characterization of novel lifespan-extending and re-

juvenative interventions [11, 26] suggests that the sig-

natures of aging integrated in AgeMeta might facilitate

TIKHONOV et al.318

BIOCHEMISTRY (Moscow) Vol. 89 No. 2 2024

Fig. 4. An example of the functional enrichment dataset plot. The plot corresponds to “GOBP: Antigen processing and presen-

tation” function in the mouse signature. The bars are colored based on statistical significance: up- and downregulated func-

tions are shown in green and red, respectively, while gray bars correspond to non-significant enrichments. Dataset specification

isdisplayed in a text box on mouse hover.

the discovery and development of new potential gero-

protectors.

Several databases of aging genomics and tran-

scriptomics exist to date. Therefore, it is important to

highlight some of the differences between the most

relevant existing databases and AgeMeta. GenAge,

a database that is a part of a collection of databases

known as Human Ageing Genomic Resources, or

HAGR [3], presents effects of genomic interventions

(e.g., gene knockouts) on lifespan for model organ-

isms such as Mus musculus. The association of these

effects with age-related transcriptomic changes is

not trivial, i.e., genomic interventions that result in a

longer lifespan do not always counteract the gene ex-

pression changes that happen with age. This could be

explained by some of the age-related gene expression

changes being detrimental, while some being adap-

tive and having a positive effect on the aging pheno-

type [11]. Anexample of this phenomenon is Pparg, a

gene which is significantly upregulated with age in the

mouse signature but whose knockout significantly de-

creases the mean lifespan of mice [27].

There are other databases with data on aging

transcriptomics, e.g., Aging Atlas [2] and AGEMAP [5].

Both of them contain differential expression informa-

tion from individual studies (there is only one study

in AGEMAP), but none of them offer meta-analysis-

derived differential expression values, aggregated

across multiple studies with multiple tissues and spe-

cies, which is the key distinguishing feature of AgeMeta.

Aging Atlas also contains multi-omics data, including

protein–protein interactions, ChIP-Seq data, and single

cell transcriptomics, which were outside of the scope

of our study. Another database, Open Genes [6], con-

tains descriptions of genes with many different types

of associations with aging, including knockouts affect-

ing lifespan and age-related differential expression,

but this database also does not contain results of a

quantitative meta-analysis.

A couple of limitations of our work should be not-

ed. Firstly, the signatures of aging used in AgeMeta

were based on linear differential expression models,

which could imply that some non-linear age-related

gene expression changes were not identified as statis-

tically significant in this study. Secondly, differential

co-expression was shown to occur with age in vari-

ous mammalian species [28-30], whereas in our work

only expression changes of individual genes were ex-

amined. Expansion of AgeMeta with a meta-analysis

of non-linear differential gene expression and gene

co-expression patterns may further improve the com-

prehensiveness and the application scope of our da-

tabase as well as help enhance the understanding of

molecular mechanisms of mammalian aging by the

scientific community.

Supplementary information. The online version

contains supplementary material available at https://

doi.org/10.1134/S000629792402010X.

Contributions. A.T., S.E.D., and M.B. conceived

the study; S.T. and A.T. designed the database; S.T. per-

formed data analysis and developed the database; S.T.,

M.B., V.G., S.E.D., and A.T. interpreted the data; A.T. su-

pervised the study; S.T. and A.T. wrote the manuscript

with contributions from all other authors. All authors

read and approved the final version.

Funding. The work was supported by the Russian

Science Foundation (grant no. 23-14-00218 to S.E.D.;

data processing and creating transcriptomic signa-

tures) and Open Longevity (developing the database).

AgeMeta: GENE EXPRESSION IN AGING 319

BIOCHEMISTRY (Moscow) Vol. 89 No. 2 2024

Ethics declarations. This work does not contain

any studies involving human and animal subjects.

The authors of this work declare that they have no

conflicts of interest.

Open access. This article is licensed under a Cre-

ative Commons Attribution 4.0 International License,

which permits use, sharing, adaptation, distribution,

and reproduction in any medium or format, as long

as you give appropriate credit to the original au-

thor(s) and the source, provide a link to the Creative

Commons license, and indicate if changes were made.

Theimages or other third-party material in this article

are included in the article’s Creative Commons license,

unless indicated otherwise in a credit line to the mate-

rial. If material is not included in the article’s Creative

Commons license and your intended use is not permit-

ted by statutory regulation or exceeds the permitted

use, you will need to obtain permission directly from

the copyright holder. To view a copy of this license,

visit http://creativecommons.org/licenses/by/4.0/.

REFERENCES

1. Harman,D. (1991) The aging process: Major risk factor

for disease and death, Proc. Natl. Acad. Sci. USA, 88,

5360-5363, doi:10.1073/pnas.88.12.5360.

2. Aging Atlas Consortium (2021) Aging Atlas: a multi-om-

ics database for aging biology, Nucleic Acids Res., 49,

825-830, doi:10.1093/nar/gkaa894.

3. Tacutu,R., Thornton,D., Johnson,E., Budovsky,A., Ba-

rardo,D., Craig,T., Diana, E., Lehmann, G., Toren,D.,

Wang,J., De Magalh,P., and Fraifeld,E. (2018) Human

Ageing Genomic Resources: new and updated data-

bases, Nucleic Acids Res., 46, 1083-1090, doi:10.1093/

nar/gkx1042.

4. Craig, T., Smelick, C., Tacutu, R., Wuttke, D., Wood,

S. H., Stanley, H., Janssens, G., Savitskaya, E., Moska-

lev,A., De Magalh,P., and Arking,R. (2015) The Digital

Ageing Atlas: integrating the diversity of age-related

changes into a unified resource, Nucleic Acids Res.,

43, 873-878, doi:10.1093/nar/gku843.

5. Zahn, J. M., Poosala, S., Owen, A. B., Ingram, D. K.,

Lustig, A., Carter, A., Weeraratna, A. T., Taub, D. D.,

Gorospe, M., Mazan-mamczarz, K., Lakatta, E. G.,

Boheler, K. R., Xu, X., Mattson, M. P., Falco, G., Ko,

M.S.H., Schlessinger,D., Firman,J., Kummerfeld, S.K.,

Iii, W. H. W., Zonderman, A.B., Kim, S.K., and Beck-

er, K.G. (2007) AGEMAP: a gene expression database

for aging in mice, PLoS Genet., 3, e201, doi: 10.1371/

journal.pgen.0030201.

6. Rafikova, E., Nemirovich-Danchenko, N., Ogmen, A.,

Parfenenkova, A., Velikanova, A., Tikhonov, S., Pe-

shkin, L., Rafikov, K., Spiridonova, O., Belova,Y., Gli-

nin,T., Egorova,A., and Batin,M. (2023) Open Genes–

a new comprehensive database of human genes as-

sociated with aging and longevity, Nucleic Acids Res.,

52, D950-D962, doi:10.1093/nar/gkad712.

7. Edgar,R., Domrachev,M., and Lash, A.E. (2002) Gene

expression omnibus: NCBI gene expression and hy-

bridization array data repository, Nucleic Acids Res.,

30, 207-210, doi:10.1093/nar/30.1.207.

8. Parkinson, H., Kapushesky, M., Shojatalab, M., Abey-

gunawardena,N., Coulson,R., Farne,A., Holloway,E.,

Kolesnykov,N., Lilja,P., Lukk,M., Mani,R., Rayner,T.,

Sharma, A., William, E., Sarkans, U., and Brazma, A.

(2007) ArrayExpress – a public database of microar-

ray experiments and gene expression profiles, Nucleic

Acids Res., 35, 747-750, doi:10.1093/nar/gkl995.

9. Leinonen, R., Sugawara,H., and Shumway, M. (2011)

The sequence read archive, Nucleic Acids Res., 39,

2010-2012, doi:10.1093/nar/gkq1019.

10. Lonsdale, J., Thomas, J., Salvatore, M., Phillips, R.,

Lo, E., Shad, S., Hasz, R., Walters, G., Garcia, F.,

Young,N., Foster,B., Moser,M., Karasik,E., Gillard,B.,

Ramsey, K., Sullivan, S., Bridge, J., Magazine, H., Sy-

ron,J., Fleming,J., Siminoff,L., Traino,H., Mosavel,M.,

Barker,L., Jewell,S., Rohrer,D., Maxim,D., Filkins,D.,

Harbach, P., Cortadillo, E., Berghuis, B., Turner, L.,

Hudson,E., Feenstra,K., Sobin,L., Robb,J., Branton,P.,

Korzeniewski,G., Shive,C., Tabor,D., Qi,L., Groch,K.,

Nampally,S., Buia,S., Zimmerman,A., Smith,A., Burg-

es,R., Robinson,K., Valentino,K., Bradbury,D., Cosen-

tino,M., Diaz-Mayoral,N., Kennedy,M., Engel,T., Wil-

liams,P., Erickson,K., Ardlie,K., Winckler,W., Getz,G.,

DeLuca, D., MacArthur, D., Kellis, M., Thomson, A.,

Young,T., Gelfand,E., Donovan,M., Meng,Y., Grant,G.,

Mash,D., Marcus,Y., Basile,M., Liu, J., Zhu,J., Tu,Z.,

Cox, N. J., Nicolae, D. L., Gamazon, E. R., Im, H. K.,

Konkashbaev, A., Pritchard, J., Stevens, M., Flutre, T.,

Wen,X., Dermitzakis, E.T., Lappalainen,T., Guigo,R.,

Monlong,J., Sammeth,M., Koller,D., Battle,A., Mosta-

favi, S., McCarthy, M., Rivas, M., Maller, J., Rusyn, I.,

Nobel, A., Wright, F., Shabalin, A., Feolo, M., Sharop-

ova,N., Sturcke,A., Paschal,J., Anderson, J.M., Wild-

er, E.L., Derr, L.K., Green, E.D., Struewing, J.P., Tem-

ple,G., Volpi,S., Boyer, J.T., Thomson, E.J., Guyer, M.S.,

Ng,C., Abdallah,A., Colantuoni,D., Insel, T.R., Koester,

S.E., Roger Little,A., Bender, P.K., Lehner,T., Yao,Y.,

Compton, C.C., Vaught, J.B., Sawyer,S., Lockhart, N.C.,

Demchok, J., and Moore, H. F. (2013) The Genotype-

Tissue Expression (GTEx) project, Nat. Genet., 45, 580-

585, doi:10.1038/ng.2653.

11. Tyshkovskiy, A., Ma, S., Shindyapina, A. V., Tikhon-

ov, S., Lee, S. G., Bozaykut, P., Castro, J. P., Seluan-

ov, A., Schork, N. J., Gorbunova, V., Dmitriev, S. E.,

Miller, R.A., and Gladyshev, V.N. (2023) Distinct lon-

gevity mechanisms across and within species and

their association with aging, Cell, 186, 2929-2949.e20,

doi:10.1016/j.cell.2023.05.002.

12. Ritchie, M.E., Phipson,B., Wu,D., Hu,Y., Law, C. W.,

Shi, W., and Smyth, G. K. (2015) Limma powers

TIKHONOV et al.320

BIOCHEMISTRY (Moscow) Vol. 89 No. 2 2024

differential expression analyses for RNA-sequencing

and microarray studies, Nucleic Acids Res., 43, e47,

doi:10.1093/nar/gkv007.

13. Viechtbauer,W. (2010) Conducting meta-analyses in R

with the metafor, J.Stat. Softw., 36, 1-48, doi:10.18637/

jss.v036.i03.

14. Korotkevich, G., Sukhov, V., and Sergushichev, A.

(2021) Fast gene set enrichment analysis, bioRxiv,

doi:10.1101/060012.

15. Chang, W., Cheng, J., Allaire, J., Sievert, C., Schloerke,

B., Xie, Y., Allen, J., McPherson, J., Dipert, A., and Borg-

es, B. (2022) Shiny: Web Application Framework for R..

16. Benjamini, Y., and Hochberg, Y. (1995) Controlling

the false discovery rate: a practical and powerful ap-

proach to multiple testing, J. R. Stat. Soc. Ser. B, 57,

289-300, doi:10.1111/j.2517-6161.1995.tb02031.x.

17. Subramanian,A., Tamayo, P., Mootha, V.K., Mukher-

jee,S., Ebert, B.L., Gillette, M.A., Paulovich,A., Pome-

roy, S. L., Golub, T. R., Lander, E. S., and Mesirov,

J. P. (2005) Gene set enrichment analysis: A knowl-

edge-based approach for interpreting genome-wide

expression profiles, Proc. Natl. Acad. Sci. USA, 102,

15545-15550, doi:10.1073/pnas.0506580102.

18. Mootha, V.K., Lindgren, C.M., Eriksson, K.F., Subrama-

nian,A., Sihag,S., Lehar,J., Puigserver,P., Carlsson,E.,

Ridderstråle, M., Laurila, E., Houstis, N., Daly, M. J.,

Patterson, N., Mesirov, J. P., Golub, T. R., Tamayo, P.,

Spiegelman,B., Lander, E.S., Hirschhorn, J.N., Altshul-

er,D., and Groop, L.C. (2003) PGC-1α-responsive genes

involved in oxidative phosphorylation are coordi-

nately downregulated in human diabetes, Nat. Genet.,

34, 267-273, doi:10.1038/ng1180.

19. Ashburner,M., Ball, C.A., Blake, J.A., Botstein,D., But-

ler,H., Cherry, J.M., Davis, A.P., Dolinski,K., Dwight,

S.S., Eppig, J.T., Harris, M.A., Hill, D.P., Issel-Tarver,L.,

Kasarskis,A., Lewis,S., Matese, J.C., Richardson, J.E.,

Ringwald, M., Rubin, G. M., and Sherlock, G. (2000)

Gene Ontology: tool for the unification of biology,

Nat. Genet., 25, 25-29, doi:10.1038/75556.

20. Carbon,S., Douglass,E., Good, B.M., Unni, D.R., Har-

ris, N.L., Mungall, C.J., Basu,S., Chisholm, R.L., Dod-

son, R. J., Hartline, E., Fey, P., Thomas, P. D., Albou,

L.P., Ebert,D., Kesling, M.J., Mi,H., Muruganujan,A.,

Huang, X., Mushayahama, T., LaBonte, S. A., Siegele,

D. A., Antonazzo, G., Attrill, H., Brown, N. H., Gara-

pati, P., Marygold, S. J., Trovisco, V., dos Santos, G.,

Falls,K., Tabone,C., Zhou,P., Goodman, J.L., Strelets,

V. B., Thurmond, J., Garmiri, P., Ishtiaq, R., Rodrí-

guez-López,M., Acencio, M.L., Kuiper,M., Lægreid,A.,

Logie, C., Lovering, R. C., Kramarz, B., Saverimuttu,

S.C.C., Pinheiro, S.M., Gunn,H., Su,R., Thurlow, K.E.,

Chibucos,M., Giglio,M., Nadendla,S., Munro,J., Jack-

son,R., Duesbury, M.J., Del-Toro,N., Meldal, B.H.M.,

Paneerselvam, K., Perfetto, L., Porras, P., Orchard, S.,

Shrivastava,A., Chang, H.Y., Finn, R.D., Mitchell, A.L.,

Rawlings, N. D., Richardson, L., Sangrador-Vegas, A.,

Blake, J.A., Christie, K.R., Dolan, M.E., Drabkin, H.J.,

Hill, D.P., Ni,L., Sitnikov, D.M., Harris, M.A., Oliver,

S. G., Rutherford, K., Wood, V., Hayles, J., Bähler, J.,

Bolton, E. R., de Pons, J. L., Dwinell, M. R., Hayman,

G. T., Kaldunski, M. L., Kwitek, A. E., Laulederkind,

S.J.F., Plasterer,C., Tutaj, M.A., Vedi,M., Wang, S.J.,

D’Eustachio, P., Matthews, L., Balhoff, J. P., Aleksand-

er, S. A., Alexander, M. J., Cherry, J. M., Engel, S. R.,

Gondwe,F., Karra,K., Miyasato, S.R., Nash, R.S., Simi-

son,M., Skrzypek, M.S., Weng,S., Wong, E.D., Feuer-

mann, M., Gaudet, P., Morgat,A., Bakker, E., Berardi-

ni, T.Z., Reiser,L., Subramaniam,S., Huala,E., Arighi,

C. N., Auchincloss, A., Axelsen, K., Argoud-Puy, G.,

Bateman,A., Blatter, M.C., Boutet,E., Bowler,E., Breu-

za,L., Bridge, A., Britto,R., Bye-A-Jee,H., Casas, C.C.,

Coudert, E., Denny, P., Es-Treicher, A., Famiglietti,

M.L., Georghiou, G., Gos, A. N., Gruaz-Gumowski,N.,

Hatton-Ellis, E., Hulo, C., Ignatchenko, A., Jungo, F.,

Laiho,K., Le Mercier,P., Lieberherr,D., Lock,A., Lus-

si,Y., MacDougall,A., Ma-Grane,M., Martin, M.J., Mas-

son,P., Natale, D.A., Hyka-Nouspikel,N., Orchard,S.,

Pedruzzi,I., Pourcel,L., Poux,S., Pundir,S., Rivoire,C.,

Speretta, E., Sundaram, S., Tyagi, N., Warner, K.,

Zaru,R., Wu, C.H., Diehl, A.D., Chan, J.N., Grove,C.,

Lee, R. Y. N., Muller, H. M., Raciti, D., van Auken, K.,

Sternberg, P.W., Berriman, M., Paulini,M., Howe,K.,

Gao,S., Wright,A., Stein,L., Howe, D.G., Toro,S., West-

erfield, M., Jaiswal, P., Cooper, L., and Elser, J. (2021)

The gene ontology resource: enriching a GOld mine,

Nucleic Acids Res., 49, D325-D334, doi: 10.1093/nar/

gkaa1113.

21. Kanehisa, M. (2000) KEGG: kyoto encyclopedia of

genes and genomes, Nucleic Acids Res., 28, 27-30,

doi:10.1093/nar/28.1.27.

22. Kanehisa,M. (2019) Toward understanding the origin

and evolution of cellular organisms, Protein Sci., 28,

1947-1951, doi:10.1002/pro.3715.

23. Kanehisa, M., Furumichi, M., Sato, Y., Ishiguro-Wata-

nabe, M., and Tanabe, M. (2021) KEGG: Integrating

viruses and cellular organisms, Nucleic Acids Res., 49,

D545-D551, doi:10.1093/nar/gkaa970.

24. Jassal,B., Matthews,L., Viteri,G., Gong,C., Lorente,P.,

Fabregat, A., Sidiropoulos, K., Cook, J., Gillespie, M.,

Haw,R., Loney,F., May,B., Milacic,M., Rothfels,K., Se-

villa, C., Shamovsky, V., Shorser, S., Varusai, T., Weis-

er, J., Wu, G., Stein, L., Hermjakob, H., and D’Eusta-

chio,P. (2020) The reactome pathway knowledgebase,

Nucleic Acids Res., 48, D498-D503, doi: 10.1093/nar/

gkz1031.

25. Rouillard, A. D., Gundersen, G. W., Fernandez, N. F.,

Wang, Z., Monteiro, C. D., McDermott, M. G., and

Ma’ayan, A. (2016) The harmonizome: a collec-

tion of processed datasets gathered to serve and

mine knowledge about genes and proteins, Data-

base (Oxford)

, 2016, baw100, doi: 10.1093/database/

baw100.

AgeMeta: GENE EXPRESSION IN AGING 321

BIOCHEMISTRY (Moscow) Vol. 89 No. 2 2024

26. Zhang, B., Lee, D. E., Trapp, A., Tyshkovskiy, A., Lu,

A.T., Bareja,A., Kerepesi,C., McKay, L.K., Shindyapi-

na, A.V., Dmitriev, S.E., Baht, G.S., Horvath,S., Glady-

shev, V. N., and White, J.P. (2023) Multi-omic rejuve-

nation and lifespan extension on exposure to youthful

circulation, Nat. Aging, 3, 948-964, doi:10.1038/s43587-

023-00451-9.

27. Argmann, C., Dobrin, R., Heikkinen, S., Auburtin, A.,

Pouilly,L., Cock, T.A., Koutnikova,H., Zhu,J., Schadt,

E.E., and Auwerx,J. (2009) Pparγ2 is a key driver of

longevity in the mouse, PLoS Genet., 5, e1000752,

doi:10.1371/journal.pgen.1000752.

28. Bernard,G., Teulière,J., Lopez,P., Corel,E., Lapointe,

F. J., and Bapteste, E. (2022) Aging at evolutionary

crossroads: longitudinal gene co-expression network

analyses of proximal and ultimate causes of aging

in bats, Mol. Biol. Evol., 39, msab302, doi: 10.1093/

molbev/msab302.

29. Southworth, L. K., Owen, A. B., and Kim, S.K. (2009)

Aging mice show a decreasing correlation of gene

expression within genetic modules, PLoS Genet., 5,

e1000776, doi:10.1371/journal.pgen.1000776.

30. Hu,Y., Xu,Y., Mao,L., Lei,W., Xiang,J., Gao,L., Jiang,J.,

Huang, L., Luo, O. J., Duan, J., and Chen, G. (2021)

Gene expression analysis reveals age and ethnicity

signatures between young and old adults in human

PBMC, Front. Aging, 2, 797040, doi: 10.3389/fragi.

2021.797040.

Publisher’s Note. Pleiades Publishing remains

neutral with regard to jurisdictional claims in pub-

lished maps and institutional affiliations.