7 Combinatories and Topology of theβ-Sandwich andβ-Barrel Proteins 133Fig. 7.2.Schematic representation of the barrel structure (1bia). The mainβ-sheet
strands are numbered as they presented in the sequence. The strand 2 is divided
into 2′and 2′′parts. The long strand 2 is bent into two parts: 2′and 2′′.The2′
part forms H-bonds with strand 1 in one part of theβ-sheet, while the 2′′part has
hydrogen contacts with strand 3 in the second part of theβ-sheet. The strands 1,
2 ′, and 5 form one subsheet and the strands 2′′, 3, and 4 form the second subsheet
k+ 1 is also fixed: residues at position 6 ini+ 1 and position 8 ink+ 1 strands
are located opposite to each other in the sheet, but do not form H-bonds. Side
chains of both residues look inside the hydrophobic interior of SP.
7.3.6 Method of the Sequence Determinants for Identification
of Proteins
Knowledge of sequence determinants of protein groups allows us to develop
a computer algorithm for classification of proteins. A set of sequence deter-
minants is characterized by (1) a number of the sequence determinants for
a given protein group (usually there are 8–12 sequence determinants); (2) a
set of residues that characterize each sequence determinant, and (3) intervals
(a number of residues) between the sequence determinants in sequences. This
data will be used to distinguish all proteins of a particular protein group and
to predict the secondary and tertiary structure.
For the search procedure, we implemented an algorithm based on appro-
priate modification of the dynamic programming [33]. This algorithm matches
one-by-one sequence determinants of a given protein group with residues of
the query sequence. Once a match is found for the sequence determinant clos-
est to the beginning of the sequence, the algorithm starts to look for a match
for the second determinant in the query sequence, and so on. If all sequence
determinants are matched, then the protein is assigned to the group.