An Aging Cell study compared gene activity across 25 thousand samples: age-related changes differed among organs, while most tumor types showed a more even distribution of activity across genes
An Aging Cell study compared gene activity across 25 thousand samples: age-related changes differed among organs, while most tumor types showed a more even distribution of activity across genes
In an Aging Cell article published on 4 September, the authors calculated Shannon entropy from RNA data for more than 25 thousand human and mouse tissue samples. This measure indicates whether most gene activity is concentrated in a small number of genes or distributed across many genes.
The authors first tested whether this measure changes with age in the same way throughout the body. After adjustment for the available sample characteristics, entropy decreased in the brain, stomach, and blood. It increased in the salivary gland, heart, skin, adipose tissue, and skeletal muscle, and remained stable in the other tissues. The age-related pattern therefore depends on the organ.
Conventional RNA analysis measures the aggregate signal from a tissue, combining changes within cells with shifts in the proportions of different cell types. The authors estimated this cellular composition mathematically. The age-related increase in entropy in skin and skeletal muscle remained after adjustment. In single-cell data from liver cancer and melanoma, tumor samples often had higher entropy even when cells of the same type were compared.
In paired samples of tumors and adjacent normal tissue, primary tumors of most types had higher entropy. In a separate comparison of melanoma samples, entropy increased from skin that had not been exposed to the sun, through primary tumors, to late metastases. The skin and tumor data came from different datasets. Entropy increased after the tumors acquired resistance to therapy in five of six paired melanoma samples. Among the cancer types in which the association with overall survival was significant, high entropy was associated with worse survival in approximately 70%.
Entropy also increased in cellular reprogramming experiments. The less successful chemical protocol produced a higher value than the more successful protocol. The same calculation reflects both the state of the cells and their relative proportions within a tissue. Its interpretation therefore depends on the specific organ and the cellular composition of the sample.