Information Technology Reference
In-Depth Information
A spectrin molecule is a heterodimer composed of domains: D and E connected by
noncovalent bonds with each other. Chains of those domains are winded round the same
axis, anti-parallel. The tetramer is 200 nm long with the diameter 3 nm and it is the basic
functional unit of spectrin molecule.
E-Spectrins concern 19 segments; the first of them is a high conservative N-terminal
domain composed of two neighbouring domains that are calponin homologues. The centre
of the molecule is occupied by 17 triple helical repeats and the terminal part - by a domain
concerning PH - motif - a region homologic to pleckstrin. The actin binding sites are
localized in the middle part of the tetramer (segment 15).
In this paper the E-spectrin family of proteins was subjected to a theoretical analysis
of the variability mechanisms in their primary structure, through examinations of
similarities and differences within the family. The correlation of the variability of several
positions and functions of the proteins was also analysed. On the basis of the global
multiple alignment and consensus sequence that was constructed, a level of semihomology
and identity was estimated.
1. Materials and methods
The initial material used it this research were E-spectrin amino acid sequences obtained
from Swiss Prot
1
database and through BLAST
2
programme. The first step was to find a
model sequence of human erythroid E-spectrin (access number: P11277) and then using
BLAST programme - to find its homologues. 67 sequences were colleted; 12 of them were
rejected - they were recurrent sequences, incomplete or totally distinct in their amino acid
composition than other sequences. The algorithm of genetic semihomology [6] was used
for the analysis of the correlation between amino acids at semihomologous and non-
homologous positions, mechanism of variability, location of gaps, the multiple alignment
and the consensus sequence construction. The multiple alignment was constructed
tentatively by the Test
3
programme aligning two sequences, based on the algorithm of
genetic semihomology. The graphic figure of the multiple alignment was created using
Protein Calculator
4
(v0.901 Beta version) programme, colouring the conservative positions
and constructing tentatively the consensus sequence.
2. Results and discussion
2.1 General characteristics of algorithm of genetic semihomology
The algorithm of genetic semihomology assumes, that the basic (but not the only)
mechanism of evolutionary diversity of proteins is single point mutation that may lead to
the replacement one amino acid residue by another in one or more positions of homologous
sequences. It is based on simple, clearly defined rules; connections between codons of
several amino acid residues make it's sense. The main part of the algorithm is a three-
dimensional diagram showing a network of genetic relations among amino acids (Fig. 1.).
1
www.expasy.ch/sprot/
2
www.ncbi.nlm.nih.gov/blast/Blast.cgi
3
Programmes are properties of Interdysciplinary Centre for Mathematical and Computational Modelling,
Warsaw University.
