Paper ID: 269
University of Aveiro (Portugal)
The concept of statistical proteins was proposed in 1965 by Carl Woese1. He defined statistical proteins as mixtures of polypeptides whose primary structures are related to some theoretical average primary structure. This concept was overlooked for many years due to its association in the original publication with primordial life forms with poorly defined genetic codes. However, recent studies show that it also applies to existing organisms. We have discovered a statistical proteome in the main human fungal pathogen Candida albicans2. Its genome encodes 6198 protein coding genes (haploid genome) - similar to other fungi -, but ambiguous gene translation by the ribosome diversifies stochastically its proteome producing millions of different proteins that are not degraded by the protein quality control machinery. In other words, in Candida albicans there is no correlation between gene and protein numbers despite the lack of alternative splicing. I will illustrate in my talk that global re-programming of the genetic code produces statistical proteins that have specific cellular functions. Such proteins increase dramatically phenotypic and genetic diversity, expand adaptation capacity in changing ecological landscapes and influence virulence, biofilm formation and drug resistance.