Poster Presentation 49th Lorne Conference on Protein Structure and Function 2024

We do know what we don’t know: unlocking archaeal genomes using artificial intelligence. (#222)

Rebecca S Bamert 1 , Giovanni Leandri 1 , Marie C Schoelmerich 2 , Rohan Sachdeva 2 , Gavin J Knott 1 , Jillian F Banfield 1 2
  1. Biomedicine Discovery Institute, Monash University, Clayton, VIC, Australia
  2. Innovative Genomics Institute, University of California, Berkeley, California, USA

The comprehensive sequencing of both Borgs1 (huge archaeal extrachromosomal elements) and archaea themselves from diverse environments has revealed a limitation of annotation. Using sequence-based analysis for many of these novel ORFs finds little that reveals much of the protein's function and role within the cell, and larger environment. Intriguingly Borgs contain many multicopy novel proteins that at first pass can only be ascribed limited function, as a generic-ase, or defence element, or the dreaded domain of unknown function (DUF). Using AlphaFold22 to generate high confidence models, and then searching for structural homologues is unlocking function and sketching an increasingly interesting picture of the interactions between Borgs and their archaeal hosts. However this structure-driven approach alone is not a panacea, and as these examples show, the synthesis of a combination of structural, sequence and synteny analysis  is required to begin to understand the biology of this understudied system.

 

 

  1. 1.Al-Shayeb B, Schoelmerich MC, West-Roberts J, Valentin-Alvarado LE, Sachdeva R, Mullen S, Crits-Christoph A, Wilkins MJ, Williams KH, Doudna JA, Banfield JF. Borgs are giant genetic elements with potential to expand metabolic capacity. Nature. 2022 Oct;610(7933):731-736. doi: 10.1038/s41586-022-05256-1.
  2. 2.Jumper, J., Evans, R., Pritzel, A. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021). https://doi.org/10.1038/s41586-021-03819-2