The Structural History of Eukarya
|The Structural History of Eukarya
|NAISS Medium Compute
|Patrick Bryant <firstname.lastname@example.org>
|2023-12-21 – 2024-07-01
Comparative genomics suffers from varying evolutionary rates across genes and genomes. Therefore, ribosomal RNA genes have been the main focus to construct phylogenetic relationships since these are more stable. This approach ignores most of the evolutionary changes in an organism arising across its genome. In addition, sequence studies only capture genetic changes that may take place but that do not necessarily impact the resulting protein structures.
To capture a more complete picture of evolution, we aim to construct evolutionary relationships based on predictions of all protein structures in an organism. We define a similarity metric between proteomes based on the structural bit score of aligned protein regions and compute the evolutionary history for all Eukarya with available proteomes in UniProt. The result is both a complete picture of single protein- and proteome structural similarity throughout eukaryotic evolution.
We study what proteomes change the fastest and what drives this. We also study instances of highly similar proteins arising in different evolutionary branches, suggesting convergent evolution. Our resource, the Structural History of Eukarya, will show the complete evolutionary relationships between eukaryotic organisms’ proteomes.
We believe that our approach to organism comparison and the resource we provide will help accelerate evolutionary comparisons and elucidate hidden aspects of the history of Eukarya. The vast amount of comparisons necessary to construct this resource can only be made by specialised research groups, which is why we will provide a web interface that is accessible free of charge to query this resource. Complete evolutionary histories of selected proteins can be generated and entire proteomes compared on the structural level directly in the web interface.
This will be a valuable resource for evolutionary comparisons and we plan to expand the resource further in the future to incorporate all complete proteomes in UniProt. The method to be created here serves as a stepping stone in whole proteome comparison on the structural level. We believe that this is a starting point for proteome-wide structural comparisons that have been made available through recent algorithmic developments in structure prediction and search.