Carnes · PloS one 2015 · experimental evolution with pooled DNA and RNA sequencing · n=10 fly lines (5 selected lines, 5 control lines)

The Genomic Basis of Postponed Senescence in Drosophila melanogaster.

Cited 55 times in the scientific literature.

Level 5 - mechanism / opinion, no new human data

Animal model genomic study (Level 5 by design analogy, no clinical data)

PubMed 26378456 · doi:10.1371/journal.pone.0138569 · record verified 2026-08-30

What was done

Researchers evaluated genetic and transcriptomic divergence between five Drosophila melanogaster lines selected for postponed senescence over more than 170 generations (O lines) and five control lines maintained at a two-week generation interval for over 850 generations (B lines). They performed pooled population DNA sequencing across all lines to identify divergent variants and conducted sex-specific RNA sequencing at one and five weeks of age to assess expression differences related to age, population, and their interaction, integrating both datasets to identify top candidate genes.

What was found

O lines lived an average of 70% longer than B lines, maintained higher productivity at all ages, and showed delayed senescence across non-reproductive traits. Genomic sequencing revealed 6,394 divergent variants in or near 1,928 genes at a false discovery rate (FDR) of 0.068, including a 2.6 Mb region on the X chromosome showing fixed alternative alleles. Transcriptomic profiling (FDR < 0.05) combined with genomic divergence identified 98 top candidate genes in females and 175 top candidate genes in males associated with postponed senescence.

Why it matters

This work provides a high-resolution map of the polygenic architecture underlying evolved longevity in a model organism, highlighting extensive sex-specific transcriptional responses and generating a catalog of candidate genes for future aging research.

Limits

The findings are derived entirely from laboratory-evolved Drosophila melanogaster and cannot be directly extrapolated to human aging. The abstract reports computational candidate identification without functional validation (such as targeted knockdowns or overexpression), and the exact number of individual flies within the sequenced pools is not specified.

Cited by