Python Programming for Biology: Bioinformatics and Beyond

Figure 15.1.  The hierarchical levels at which protein structures are commonly

described. Proteins may be described in various ways, from a simple sequence of amino

acids to a full folded, three-dimensional structure. The primary structure of a protein is the

covalently bound chain of amino acid residues. The secondary structure represents the

formation of characteristic hydrogen-bonding patterns (and thus also angles of twist)

between its backbone peptide groups. The two most common secondary-structure types

are the α-helix and the β-sheet (formed of β-strands, represented here as arrows). The

tertiary structure represents the three-dimensional shape of one protein chain and the

tertiary structure is how multiple three-dimensional protein chains combine to form larger

assemblies. The quaternary structure represents the association of multiple molecules,

which may or may not be of different types, into a larger complex.

---

The  membrane-spanning  part  of  a  membrane  protein  is  commonly  referred  to  as  a

transmembrane  domain.  The  transmembrane  part  may  only  be  a  small  part  of  a  larger

protein, where the other parts are in water, but in some cases it may comprise almost all of

the  protein.  The  folding  of  a  transmembrane  protein  domain  is  naturally  somewhat

different to that of aqueous domains. Indeed, most membrane proteins are inserted through

a hole in a membrane as they are made. However, it is not the insertion into a membrane

that makes a protein a membrane protein, but rather its amino acid composition. Firstly, a

special signal (a special kind of amino acid sequence at the start of the protein chain) is

needed to get them to the membrane when they are made. Secondly, in order to persist in a

membrane  the  protein  needs  to  have  a  large  number  of  hydrophobic  amino  acids  (those

that cannot form especially favourable interactions with water) to form the transmembrane

domain. Such protein domains are much more hydrophobic than those that sit in water; if

they  were  not  so  they  would  simply  dissolve  in  the  water,  leaving  the  membrane.

Accordingly,  you  can  often  spot  a  transmembrane  span  by  looking  at  the  amino  acid

sequence, because it has so many hydrophobic residues in a concentrated region.