Structure-based functional predictions for un-annotated proteins of uropathogenic E. coli
One of the key challenges in modern biology is that of utilizing our now-massive knowledge of protein and nucleic acid sequences to obtain functional insight. For many organisms of medical or biotechnological importance, large fractions of the proteome remain un-annotated, and existing methods can offer little in the way of reliable predictions for the functions of the un-annotated proteins.
We have recently developed a structure-based functional annotation pipeline that combines protein structure prediction with more established methods to provide highly reliable consensus functional predictions. Here we will apply our pipeline to annotate the complete proteome of E. coli UTI89, a model for uropathogenic E. coli (UPEC). Urinary tract infections (primarily caused by UPEC) affect approximately 10% of women and 2% of men in the United States every year, with total annual costs estimated to be $2.5 billion. By understanding the complete set of functionalities encoded in the genome of a typical UPEC strain, and comparing with the non-pathogenic E. coli K12 strain that we have already studied, we will identify new potential therapeutic targets for prevention or treatment of urinary tract infections.