Author(s): Surkont, Jarosław
Date: 2016
Persistent ID: http://hdl.handle.net/10362/45364
Origin: Repositório Institucional da UNL
Project/scholarship: info:eu-repo/grantAgreement/FCT/SFRH/SFRH%2FBD%2F51880%2F2012/PT;
Subject(s): Biology; Bioinformatics; Biology/Bioinformatics
Evolutionary processes can be considered at multiple levels of biological organization. The work developed in this thesis focuses on protein molecular evolution. Although proteins are linear polymers composed from a basic set of 20 amino acids, they generate an enormous variety of form and function. Proteins that have arisen by a common descent are classified into families; they often share common properties including similarities in sequence, structure, and function. Multiple methods have been developed to infer evolutionary relationships between proteins and classify them into families. Yet, those generic methods are often inaccurate, especially when specific protein properties limit their applications. In this thesis, we analyse two protein classes that are often difficult for the evolutionary analysis: the coiled-coils – repetitive protein domains defined by a simple widespread peptide motif (chapters 2 and 3) and Rab small GTPases – a large family of closely related proteins (chapters 4 and 5). In both cases, we analyse the specific properties that determine protein structure and function and use them to improve their evolutionary inference.
