How flies and people age
The aging of the brain of humans and flies was united by the work of 50 genes
Anna Muravyeva, N+1
Comparative analysis of the RNA of human prefrontal cortex cells and drosophila head tissues allowed us to identify 50 conservative genes associated with aging.
Article by Webb et al. Identification of preserved transcriptome features between humans and Drosophila in the aging brain utilizing machine learning on combined data from the NIH Sequence Read Archive is published in the journal PLoS One.
Data on the expression of these genes made it possible to reliably predict the age of the owner using machine learning algorithms. Based on the expression, it was also possible to classify samples by age groups with an accuracy of 93 percent for drosophila and 88 percent for humans. These results show conservative aging genes, with the help of which this process can be studied on model organisms.
Biologists are trying to invent a "cure for old age," but it is becoming increasingly obvious that it is more effective to fight against individual manifestations of aging. For example, old age is the biggest risk factor in the development of neurodegenerative diseases. In turn, it is affected by the expression of genes: for example, proteins of interneuronal transmission or signals of cell death. In order to investigate conservative (preserved in evolution) aging processes in the brain, a comparative analysis can be carried out between evolutionarily distant species – their common features can be considered conservative and used for further research.
That 's exactly what the researchers from University of Iowa under the leadership of Joe L. Webb and Simon Moe (Simon M. Moe). They analyzed data on the work of genes in human prefrontal cortex cells and drosophila head tissue cells (brain and visual system). Biologists used data from open source articles that contained RNA sequencing (transcriptome) of the necessary tissues in healthy subjects for whom the age was known. The transcriptomes for analysis consisted of a large number of short RNA reads that passed the filter on the reliability of reading each nucleotide.
The resulting sequences were aligned to the genomes of the corresponding organisms – that is, the sequencing sequences of the entire DNA set of cells were compared and the gene to which the fragment belongs was determined. Then all the samples were randomly divided into training and test data for several types of machine learning models. The algorithms solved a regression problem – that is, according to sequencing data, they predicted the age of a person or a fly. The XGBoost algorithm coped best of all: for a human, its accuracy was 80 percent, and for a fly – 93. It was chosen for subsequent analysis.
All genes were ranked according to the degree of correlation of their expression with age from the strongest correlation to the weakest. Both humans and flies were selected for 1000 genes, the expression of which was most correlated with age. Further, matches were searched in these lists to identify 50 matching (evolutionarily conservative) ones.
The resulting genes were analyzed using databases to better understand their functions. Biologists have built a map of the interaction of genes using the confidence index STRING, and also identified the signaling pathways to which the genes belong. The most frequent of them were PI3K-akt, a universal signaling pathway responsible for cell development and metabolism, as well as MAPK, another universal pathway that also affects expression. Among the genes also found those associated with Alzheimer's and Parkinson's diseases.
The aging processes of the brain are also studied in species closer to humans. For example, biologists have recently discovered that fecal microbiota transplantation promotes brain rejuvenation in old mice. The researchers transplanted fecal samples from young mice to old ones, causing them to improve their immunity, hippocampal metabolites and cognitive abilities.
Portal "Eternal youth" http://vechnayamolodost.ru