Why an integrated view of gene expression studies on hematopoiesis in mouse aging is better than the sum of their parts

Leonid V Bystrykh

DOI: https://doi.org/10.1002/1873-3468.14869

Abstract:Globally, the human population is aging, with an increased proportion of people in "old age" (over 60 years). This trend leads to a growing demand in aging research, stimulating studies in animal models such as mice, fish, and invertebrates. Recently, we published a research summary on the aging of hematopoietic stem cells (HSCs) in C57BL/6 mice based on 12 gene expression datasets. Here, I discuss in greater detail the added value of taking an integrated view, rather than considering each publication separately, to determine genes involved in aging. Considerable variation exists between lists of differentially expressed (DE) genes in HSCs, comparing young and old mice. This variation can result from factors such as inconsistent definitions of "young" and "old", technical variations and variations between laboratory mouse strains. We previously demonstrated that the variation between gene lists could be circumvented by forming a unified list of DE genes-the "aging list"-with citation indexes attached. The most frequently detected DE genes [approximately 200 most cited, which we named the "aging signature" (AS)] were highly consistent across publications. Gene Ontology classification of the AS list identified additional sources of variation between studies: one comes from the specifics of how the data are collected and analyzed; another comes from inconsistencies between how we define the gene categories. As discussed, overcoming these variations is the next challenge toward an integral approach to our systematic knowledge of the aging process.