Kemena, C, Bornberg-Bauer, E
A roadmap to domain based proteomics
, 2018

[Login to Download]


Abstract

Protein domains are reusable segments of proteins and play an impor-
tant role in protein evolution. By combining the elements from a relatively
small set of domains into unique arrangements, a large number of proteins
can be generated. Since domains often have specific functions, changes
in their arrangement usually affect the overall protein function. Further-
more, domains are well amenable to computational representations, e.g.,
by HMMs (Hidden Markov Models) and these HMMs are widely repre-
sented in various data bases. Therefore, domains can be efficiently used for
proteomic analyses. Here, we describe how domains are annotated using
different domain databases and then how to assess the annotation quality of
proteomes. We next show how functional annotations of domains in large
scale data such as whole genomes or transcriptomes can be used to analyze
molecular differences between species. Furthermore, we describe methods
to analyze the changes in domain content of proteins which significantly
helps to characterize and reconstruct the modular evolution of proteins.
Altogether, domain based methods offer a computationally highly effective
approach to analyze large amounts of proteomic data in an evolutionary
setting.