On Two Problems in Comparative Genomics of Eukaryotes
Semeiks, Jeremy Raymond
MetadataShow full item record
The recent advent of whole-genome sequencing allows us to use novel comparative methods to explore the genetic bases for traits of interest. Here, I present two case studies of such methods applied to eukaryote genomes. The first study regards the evolution of longevity in the mammalian proteome. Evolutionary theory suggests that the force of natural selection decreases with age. To explore the extent to which this prediction directly affects protein structure and function, I used computational methods to identify positions of proteins conserved in long-lived but not in short-lived mammal species. I analyzed 7,590 orthologous protein families in 33 mammalian species, accounting for body mass, phylogeny, and species-specific mutation rate. Overall, I found that the number of longevity-selected positions in the mammalian proteome is much greater than would be expected by chance. Further, these positions are enriched in domains of several proteins that interact with one another in inflammation and other aging-related processes, as well as in organismal development. I present as an example the kinase domain of anti-Müllerian hormone type-2 receptor (AMHR2). AMHR2 inhibits ovarian follicle recruitment and growth, and my results show that its longevity-selected positions cluster near a SNP associated with delayed human menopause. Distinct from its canonical role in development, this region of AMHR2 may function to regulate the protein's activity in a lifespan-specific manner. The second study concerns the genetic basis for toxin production in the black mold genus Stachybotrys, which produces several diverse toxins that can damage human health. Its strains comprise two mutually-exclusive toxin chemotypes, one producing satratoxins (a subclass of trichothecenes) and the other producing the less-toxic atranones. To determine the genetic bases for chemotype-specific differences in toxin production, I sequenced and assembled de novo four Stachybotrys genomes, including two from atranone strains and two from satratoxin strains. Comparative analysis of these four 35-Mbp genomes revealed several chemotype-specific gene clusters that are predicted to make atranones and satratoxins, based on several lines of evidence. I show that chemotype-specific gene clusters are likely the genetic basis for the mutually-exclusive toxin chemotypes of Stachybotrys. I then present a unified biochemical model for Stachybotrys toxin production.