Dohmen E, Klasberg S, Bornberg-Bauer E, Perrey S and Kemena C
The modular nature of protein evolution: Domain rearrangement rates across eukaryotic life
PlosCompBio, 2018

[Login to Download]


The recombination and reuse of existing units is an important principle in evolution.
Genes can fuse or split and thereby rearrange functional units in proteins they are
coding for. These units of proteins are called domains and their rearrangements can
alter a protein’s function. For this reason the underlying genetic rearrangements can be
used to study protein innovation and adaptation during the evolution of phylogenetic
Previous studies, performed with a variety of methods, gave first insights into these
rearrangement events in small data sets. In this study we use an improved methodology
to analyse five major eukaryotic clades (vertebrates, insects, fungi, monocots and
eudicots) in unprecedented depth and breadth, making it possible for the first time to
directly compare findings between different clades and draw general conclusions about
conserved evolutionary mechanisms for eukaryotic protein innovation.
We analyse domain arrangement changes along phylogenetic trees by reconstructing
ancestral domain content in combination with possible single step events (e.g. fusion or
fission). Using this method we are able to explain up to 70% of all gained arrangements,
while the results in all clades are highly consistent, suggesting the findings can be
considered to be universally valid across eukaryotes.
In agreement with previous studies, fusions seem to be in general the most frequent
event leading to new arrangements. Interestingly, in fungi the loss of single domain
arrangements is the most frequent event, representing a distinct signal. Furthermore,
our methodology allows us to link emerging domains at specific nodes in the
phylogenetic tree to important functional developments, such as the origin of hair in
The results of this study further our understanding of the fundamental mechanisms
of protein innovation. In addition, our methodology should aid phylogenetic
reconstruction based on protein domain changes.