132

10.5

Protein Evolution: Recognizing Domains

As we have just learned, protein families are subject to evolutionary processes such as natu­

ral selection. However, most mutations have no direct influence on fitness, i.e. they do not

lead to any advantage or disadvantage (neutral evolution). In addition, random changes in

the sequence (gene drift) or larger sequence regions or even entire genes (gene shift) can

also occur, which can influence the function, e.g. catalytic domain or functional side. Thus,

it is possible that a domain occurs in several proteins, but the remaining functional domains

in the protein differ, which may contribute to new functions, for example. Domains thus

provide an important clue to the origin and function of a protein. Proteins are thus grouped

into protein families on the basis of their domains and similar functions and stored in data­

bases or can be used bioinformatically to predict domains and functions of unknown

sequences (see tutorials). Known functional domains of proteins can be found, for example,

on UniProt (https://www.uniprot.org/), bioinformatically predicted with InterPro (https://

www.ebi.ac.uk/interpro/), Pfam (https://pfam.xfam.org/), SMART (https://smart.embl-­

heidelberg.de/) or with Eukaryotic Linear Motif (ELM; https://elm.eu.org/).

­

­

10.2

2003

Fig. 10.2

Fig. 10.2  Metabolic pathway

evolution. (Figure from

Schmidt et al. 2003)

10  Understand Evolution Better Applying the Computer